• Research article
  • Open access
  • Published: 29 March 2021

Prevalence and factors related to urinary incontinence in older adults women worldwide: a comprehensive systematic review and meta-analysis of observational studies

  • Sedighe Batmani 1 ,
  • Rostam Jalali 1 ,
  • Masoud Mohammadi   ORCID: orcid.org/0000-0002-5722-8300 1 &
  • Shadi Bokaee 2  

BMC Geriatrics volume  21 , Article number:  212 ( 2021 ) Cite this article

22k Accesses

45 Citations

1 Altmetric

Metrics details

A Correction to this article was published on 25 May 2022

This article has been updated

Urinary incontinence is a common condition in the general population and, in particular, the older adults population, which reduces the quality of life of these people, so this study aims to systematically examine and meta-analyse the overall prevalence of urinary incontinence in older women around the world and the related and influential factors.

This report is a comprehensive systematic review and meta-analysis of the findings of research on urinary incontinence in older adults people across the world through looking for MEDLINE, Cochrane Library Sciencedirect, Embase, Scopus, ProQuest and Persian databases, namely iranmedex, magiran, and SID from January 2000 to April 2020, the heterogeneity of the experiments was measured using the I 2 index and the data processing was done in the Systematic Meta-Analysis programme.

In 29 studies and the sample size of 518,465 people in the age range of 55–106 years, urinary incontinence in older adults’ women in the world based on a meta-analysis of 37.1% (95% CI: 29.6–45.4%) was obtained. The highest prevalence of urinary incontinence was reported in older adults’ women in Asia with 45.1% (95% CI: 36.9–53.5%). Meta-regression also showed that with increasing the sample size and year of the study, the overall prevalence of urinary incontinence in the older adults women of the world decreased and increased, respectively, which were statistically significant differences ( P  <  0.05). According to studies, the most important factors influencing the incidence of urinary incontinence in older women are women’s age ( p  <  0.001), obesity (p <  0.001), diabetes (p <  0.001), women’s education (p <  0.001), delivery rank (p <  0.001), hypertension (p <  0.001), smoking (p <  0.001). They also have urinary tract infections (p <  0.001).

Given the high prevalence of urinary incontinence in older women around the world, health policy makers must consider control and diagnostic measures in older women and prioritize treatment and rehabilitation activities.

Peer Review reports

The World Health Organization (WHO) finds citizens 65 years of age to be older adults and the United Nations deems people with 60 years or above to be older adults [ 1 , 2 ]. The world’s population is aging rapidly, with 703 million people now over the age of 65, and this number is projected to reach 1.5 billion by 2050 [ 3 ]. Urinary incontinence is a common condition in the general population, especially the older adults, which reduces the quality of life so that ten to 20 % of all women and 77% of women living in nursing homes have urinary incontinence [ 4 ]. According to the International Association of Urinary Incontinence (ICS), any involuntary leakage of urine is called urinary incontinence (UI) [ 5 ].

Urinary incontinence is divided into three categories: stress, urgency and combination. Stress urinary incontinence (SUI) refers to the leakage of urine due to increased intra-abdominal pressure such as exercise and cough, which is due to the poor functional urethra. In connection with the reduction of anatomical support due to trauma, vaginal delivery, obesity and increased intra-abdominal pressure due to chronic constipation, lifting heavy objects and exercise is called urinary excretion with or above the distance after the sensation of excretion, urgent urinary incontinence (UUI) Called; If both urgency and stress are present together, it is called a hybrid type (MUI) [ 6 , 7 ].

Urinary incontinence has been identified as a World Health Organization health priority [ 8 ]. Urinary incontinence has many physical, mental and social effects on women’s lives [ 9 , 10 ], common mental problems in these people include anxiety and depression [ 11 , 12 ]. Physical consequences include pressure sores [ 12 ], sleep disturbances and decreased sleep quality [ 13 ], urinary tract infections [ 14 ], falls and fractures, which are the leading causes of death in people over 65 [ 15 ].

Urinary incontinence has a great impact on daily and social activities such as work, travel, physical exercise and sexual function [ 16 , 17 ] and thus reduces the quality of life [ 18 ]. Urgent incontinence is more common in nervous system disorders such as Parkinson’s, multiple sclerosis, and spinal and pelvic nerve damage [ 19 , 20 ]. Age-related changes in the lower urinary tract include decreased bladder capacity and a feeling of fullness, decreased detrusor muscle contraction rate, decreased pelvic floor muscle strength, and increased residual urine volume [ 21 ].

The prevalence of urinary incontinence among older women has been reported in different studies, with an overall prevalence of 14% in US studies [ 22 , 23 ]. In studies conducted in European countries, the prevalence of urinary incontinence has been estimated at 37% [ 24 , 25 ]. In studies conducted in different regions of Asia, the prevalence of urinary incontinence in older adults was estimated at 13% [ 26 , 27 ] and in Africa 45.3% [ 28 ]. In the study conducted in Middle Eastern countries, the prevalence of urinary incontinence was reported to be 52% [ 29 , 30 , 31 ].

In a study conducted in Iran, in a study in northern Iran (2016), one-third of older adults’ women in the city of Babol had urinary incontinence [ 32 ], in a study conducted in Yazd (2015) among women over 60 years, the prevalence of urinary incontinence was 62.2% [ 31 ]. Given the different prevalence reported and the need for consistent doses for intervention measures, and given that women cannot avoid aging and childbirth, awareness of the risk factors for urinary incontinence should be promoted.

On the other hand, studies in this field provide opaque and different information and the effective factors affecting urinary incontinence in older adults women in different studies report different reporting amounts and heterogeneity. Therefore, this study aims to answer the questions of the prevalence of urinary incontinence in older women in the world and what are the factors affecting this incontinence?

Registration number

This study has been registered with the code (IR.KUMS.REC.1399.455), in the deputy of research and technology of Kermanshah University of Medical Sciences.

Search method and time domain

This study is a systematic review and meta-analysis and is the result of extracting the findings of studies conducted in this field. First, articles published in domestic and foreign journals were retrieved by searching in databases, MEDLINE, Cochrane Library, Sciencedirect, Embase, Scopus, ProQuest, and Persian databases including iranmedex, magiran and SID in the period January 2000 to April 2020.

The researcher uses the keywords urinary incontinence, women, the older adults, urinary disorders, or similar words in Persian sources and examines English-language databases using the words: Incontinence, women, older adults, urinary disorders, Prevalence, risk factor Urinary.

Also in the google scholar search engine, both words will be done in Persian and English, and the AND, OR and NOT operators will be used in combination for more comprehensive access to all articles, so the OR pragmatist will be used to check common letters about a disorder such as (Urinary incontinence OR Urinary disorders OR Urinary Reflex Incontinence OR Urinary Urge Incontinence), (Older adults OR Aging).

As well as the word AND among the keywords: (Urinary incontinence AND older adults AND Women) will be used through word matching in the MeSH Browser.

Each article was read by two browsers independently and if the article was rejected, the reason for its rejection was mentioned and in case of disagreement between the two browsers, the article was judged by the third browser and the third referee was considered. Prevalence of study disorder based on PRISMA diagram for entering meta-analysis and to manage articles and remove duplicate articles the EndNote software has been used (version X7, for Windows, Thomson Reuters).

Selection criteria and entry and exit criteria

Articles in Persian and English are taken from cross-sectional studies as well as case-control articles, all in the group to select the factors affecting urinary incontinence in older adults’ women had the selection criteria to enter the study. And review articles, articles that do not have access to full text despite the relationship with the author of the article and lack of proper response, as well as articles that are of low quality in the evaluation of quality evaluation were removed from the review list.

Quality assessment and evaluation of the risk of bias

The Newcastle-Ottawa Scale (NOS) is a quality assessment tool for observational studies that are recommended by the Cochrane Collaboration [ 21 ]. The NOS assigns up to a maximum of nine points for the least risk of bias in three domains: 1) selection of study groups (four points); 2) comparability of groups (two points); and 3) ascertainment of exposure and outcomes (three points) for case-control and cohort studies, respectively [ 21 ], and 11 scores possible. Eventually, articles were classified as high quality (scoring ≥5 points) or low quality (scoring< 5 points). In this meta-analysis, all the articles that obtained five or more points were included.

Statistical analysis

Data were analysed using Comprehensive Meta-analysis software (Biostat, Englewood, NJ, USA version 3). To evaluate the heterogeneity of selected studies, the I 2 index test was used. If high heterogeneity is obtained in studies (75% < I 2 ), random effects model will be used for meta-analysis of studies, and if low heterogeneity is obtained (I 2  < 25%), the fixed effects model will be used for the analysis of studies [ 21 ]. also, to investigate the publication bias and regarding the high volume of samples included in the study, The Begg and Mazumdar test and its corresponding Funnel plot were used at a significance level of 0.1. the meta-regression test was used to investigate the effects of potential factors influencing the heterogeneity of the studies.

Search output

Based on studies on the prevalence and factors related to urinary incontinence in older women and including articles published in domestic and foreign journals and search in Cochrane Library Sciencedirect, Embase, Scopus, ProQuest and Persian databases including iranmedex, magiran and SID and in total searches: 2791 items were found. Then, the articles that had the initial conditions for inclusion in the study, based on the initial reviews by deleting 2522 duplicate articles and deleting 235 articles unrelated to the subject of study and deleting 5 articles during the secondary reviews due to lack of access to abstracts and main articles and low quality of articles (This number of deleted items from articles due to lack of access to the full text of articles and their abstracts due to being old or removed from the site of some journals and also their low quality in quality evaluation, of course, the deleted items due to low quality in the study is very limited.). The article entered the meta-analysis process (Fig.  1 ) (Table  1 ).

figure 1

The flowchart on the stages of including the studies in the systematic review and meta-analysis (PRISMA 2009)

Review of publication bias and meta-analysis

The heterogeneity of the studies was investigated using the I 2 test and based on this test, the amount of heterogeneity (I 2  = 99.9%) was obtained and shows high heterogeneity in the included studies, so the random-effects model was used to combine the results of the studies. Also, the results of the study of publication bias in the studies were evaluated due to the high sample size entered in the studies with Begg and Manzumdar test and with a significance level of 0.1, which indicates that the diffusion bias was not significant in the present study ( P  = 0.252) (Fig.  2 ).

figure 2

Funnel Plot Results of urinary incontinence in older adults’ women worldwide

A review of 29 studies and the sample size of 518,465 people in the age range of 55–106 years, urinary incontinence in the older adults’ women of the world based on a meta-analysis of 37.1% (95% CI: 29.6–45.4%) was obtained. The highest prevalence of urinary incontinence in older adults’ women in Egypt with 80% (95% CI:72.2–86%) in 2020 [ 50 ] and the lowest prevalence of urinary incontinence in older adults’ women in Mexico with 9.5% (95% CI:8–11.2%) was achieved in 2017 [ 33 ] (Fig.  3 ).

figure 3

Overall prevalence of urinary incontinence in older adults’ women worldwide based on a random effects model

In this figure, the prevalence of urinary incontinence is shown based on the random-effects model, in which the black square, the colour of the prevalence, and the length of the line segment on which the square is placed are 95% confidence intervals in each study.

Sensitivity analysis

A sensitivity analysis was perfumed to ensure the stability results, after removing each study results did not change (Fig.  4 ).

figure 4

Results of sensitivity analysis

Meta-regression test

To investigate the effects of potential factors influencing the heterogeneity of the overall prevalence of urinary incontinence in older women around the world, meta-regression was used for two factors: sample size and year of study (Figs.  5 and  6 ). According to Fig.  5 , with increasing sample size, the overall prevalence of urinary incontinence in the older adults omen of the world decreases ( P  <  0.05). It was also reported in Fig.  6 that with increasing the year of the study, the overall prevalence of urinary incontinence in the older adults women of the world increases ( P  <  0.05).

figure 5

Meta-regression diagram of the overall prevalence of urinary incontinence in older adults’ women worldwide by sample size

figure 6

Meta-regression diagram of the overall prevalence of urinary incontinence in older adults’ women worldwide by year of release

Subgroup analysis by continent

Based on the results of Table  2 , the highest prevalence of urinary incontinence in older adults women was reported in Asia with 45.1% (95% CI: 36.9–53.5%). The results of this table also report that no diffuse bias was observed in the study by continent, and the study of metallic mercury was also reported in each continent.

Effective and related factors in urinary incontinence in older adults’ women

According to a systematic review of studies, various factors affect the incidence of urinary incontinence in older women, the most important of which are the age of women [ 25 , 26 , 38 , 49 , 50 , 58 , 59 , 60 , 61 , 62 , 63 , 64 ], obesity based on BMI index [ 25 , 37 , 48 , 49 , 52 , 58 , 59 , 62 , 63 , 65 , 66 , 67 ], diabetes [ 25 , 26 , 37 , 49 , 52 , 58 , 62 , 66 , 67 , 68 ], women’s education [ 26 , 30 , 36 , 48 , 52 , 58 , 61 ], delivery rate [ 23 , 37 , 59 , 60 , 62 , 67 ], hypertension [ 26 , 66 , 67 ], smoking [ 30 , 36 , 37 , 52 , 60 , 62 ] as well as urinary tract infections [ 23 , 49 , 52 ]. Based on the results reported in Table  3 , all these factors have a significant difference in the incidence of urinary incontinence in older adults’ women ( p  <  0.05).

Urinary incontinence is a very common condition that usually increases with age in women. Having general information about the prevalence of this disorder and identifying risk factors is useful and even necessary that can play an effective role in improving the quality of life and general health of society [ 4 , 57 ]. This meta-analysis study was performed on 518,465 older adults women and the prevalence of urinary incontinence in older adults women was 37.1%. However, in the study of the prevalence of incontinence in older adults women by continents, the highest prevalence of urinary incontinence was reported in older adults women in Asia with 45.1%.

In a study conducted in Egypt (2020), the prevalence of incontinence among older women was 80% [ 50 ]. In the study of Summer Khan et al. in India (2018) the overall prevalence of urinary incontinence was 46.3% [ 30 ], in a study in Russia (2018) the prevalence of incontinence in older adults women was 40.2 [ 57 ].

In a study conducted in Iran (2017), it was reported that one-third of older women (33%) have urinary incontinence [ 34 ]. In another study conducted in Iran as a systematic review and meta-analysis (2018), the overall prevalence of urinary incontinence in women was estimated at 46% [ 64 ].

Based on the results, the highest prevalence of urinary incontinence in older adults’ women was reported in Asia with 45.1% and the lowest prevalence of urinary incontinence in older adults’ women was reported in America with 25.8%, By observing the prevalence in different regions, it can be concluded that the prevalence of urinary incontinence in different populations is completely different, which can be due to differences in culture or tools and methods of study.

It can also show the effect of ethnoreligious factor on the insignificance of urinary incontinence in older adults’ women in Asian countries, this issue has been stated and reported in the study of Touhidi Nezhad and et al. this study is about rectovaginal fistula and explains the importance and says that The rectovaginal fistula is a complex and multifaceted problem with social, individual, familial, religious, and ethnic-environmental dimensions [ 69 ], this can embarrass Asian women and hide and increase the prevalence of urinary incontinence in older women.

The high prevalence obtained in this study shows the need to investigate and follow up this condition, due to the significant impact of this disorder on depression and quality of life of older adults’ women, requires special attention and screening for urinary incontinence in treatment and care programs in the country. Various studies have mentioned various factors in the incidence of urinary incontinence in women, such as age, menopause, delivery and number of deliveries, obesity, and diabetes are among the most important of these factors [ 25 , 70 ].

Age is one of the important factors in the prevalence of urinary incontinence. Changes related to aging in the lower urinary system include: decreased bladder capacity and feeling of fullness, decreased rate of detrusor muscle contraction, decreased pelvic floor muscle resistance and increased residual urine volume [ 21 ].

In a study by Marland Lind et al. in the Netherlands and a study by Nazli et al. in Turkey, aging was one of the most influential factors in urinary incontinence [ 48 , 49 ], while in a study in Brazil [ 46 ] In Iranian older adults women, no relationship was observed between urinary incontinence and aging [ 34 ]. Menopause, with a decrease in estrogen and a decrease in collagen, reduces the elasticity of the detrusor muscle of the ductus arteriosus and atrophic changes in the pelvic floor muscles and increases urinary incontinence in women [ 71 ].

In the study conducted in Turkey, menopause is one of the most important factors influencing female incontinence [ 48 ], while in the study of Aquarius et al. in Brazil, no significant relationship was reported between menopause and the increased prevalence of urinary incontinence [ 72 ]. Urinary incontinence is higher in women with more deliveries and vaginal deliveries. These two factors seem to be one of the most important risk factors for urinary incontinence in women [ 73 ]. In the study conducted among Chinese women, there is a type of delivery and the possibility of urinary incontinence [ 52 ], also in the study of Marland Lind et al. there was a significant relationship between delivery history, number and type of delivery with increased urinary incontinence [ 49 ]. However, in a study in India, no association was found between childbirth and urinary incontinence [ 30 ].

Obesity is an exacerbating condition of urinary incontinence, which can be caused by the accumulation of excess weight on the urinary tract during life [ 22 ]. Many studies have shown an association between obesity and increased urinary incontinence. In a study by Ninomia et al. in Japan [ 59 ] and a study by Hong et al. in the United States [ 74 ], a significant relationship was found between weight gain and increased incidence of urinary incontinence.

Also, the level of education is considered as one of the components of individual and social development and its role in personal health and also a factor in increasing the quality of life [ 9 ]. In his study by Espanyo et al. in Mexico and the United States [ 36 ] and in the study by Marcos et al. in Brazil [ 23 ], increasing the level of education was reported to be an important factor in reducing the incidence of urinary incontinence. No urinary incontinence was reported between education levels [ 66 ].

Diabetes can cause UI by several mechanisms, hyperglycaemia causes increased urine volume and increased activity of the bladder muscle, and ultimately causes dysfunction of this muscle. Diabetic cytopathic and bladder nerve damage are other effective complications [ 75 ]. In a study by Absen et al. in Norway, it was reported that there was a significant association between diabetes and urinary incontinence [ 60 ], while a German study found no association between diabetes and urinary incontinence [ 25 ].

Chronic respiratory diseases are associated with symptoms such as a cough that can cause urinary incontinence [ 76 ]. In a study based on the population of Jinge Ge et al. in China, a significant relationship was reported between lung disease and incidence [ 39 ]. However, in the study of Ralph Souher et al. in Germany [ 4 ] and the study of Sohan et al. in Korea [ 26 ], no significant relationship was observed between urinary incontinence and respiratory disease.

Nervous system disorders are seen as an important factor in the prevalence of urgent incontinence [ 19 , 20 ]. There were mental illnesses, cancer and conditions such as living alone [ 25 ]. A study by Kasik et al. in Turkey also reported obesity, smoking, a history of constipation, UTI, family history, chronic illness, chronic cough, a history of hormone therapy, genital prolapse, a history of urology, and a history of communication impairment. Have significance with incontinence [ 46 ].

In a promising study by colleagues in Iran, it was reported that urinary incontinence is directly and significantly related to factors such as marital status, constipation, and corticosteroid medications, while urinary incontinence is associated with factors such as age, obesity, education, number of children, diabetes, hypertension, and Respiratory disorders were not associated [ 34 ].

In a 2016 study by Aquarius et al. in Brazil, the factors that increased urinary incontinence in women included: number of pregnancies, deliveries, genital prolapse, anxiety, depression, and obesity [ 72 ]. In a study by Marcos et al. in Brazil, there was a significant relationship between age, education, physical activity, dependence, cognitive problems, symptoms of anomia, bronchitis, asthma, cardiovascular disease, diabetes, hypertension, stroke and ischemia, nutritional status, polypharmacy, self-Urinary incontinence was reported [ 23 ].

Given the above, it is necessary for physicians and specialists to consider adults’ women in the age group of 55 to 106 years according to the criteria recommended by the International Continence Society (ICS) and to standardize the criteria so that diagnostic and treatment strategies are more effective.


The most important limitations of the present study are the high heterogeneity of studies, which can be due to sampling size, age groups, geographical areas, races, and other different factors in the studies, which can be controversial in the study.

Given the high prevalence of urinary incontinence in older women around the world, health policy makers must considerand diagnostic measures in older women and prioritize treatment and rehabilitation activities.

Availability of data and materials

Datasets are available through the corresponding author upon reasonable request.

Change history

25 may 2022.

A Correction to this paper has been published: https://doi.org/10.1186/s12877-022-03111-6


World Health Organization

International Association of Urinary Incontinence

urinary incontinence

Stress urinary incontinence

Urgent urinary incontinence

The Newcastle-Ottawa Scale

Scientific information database

Preferred reporting items for systematic reviews and meta-analysis

Strengthening the reporting of observational studies in epidemiology for cross- sectional study

Vaisi-Raygani AA, Mohammadi M, Jalali R, Ghobadi A, Salari N. The prevalence of obesity in older adults in Iran: a systematic review and meta-analysis. BMC Geriatr. 2019;19(1):371.

Lukacz ES, Santiago-Lastra Y, Albo ME, Brubaker L. Urinary incontinence in women: a review. JAMA. 2017;318(16):1592–604. https://doi.org/10.1001/jama.2017.12137 .

Article   PubMed   Google Scholar  

Hu JS, Pierre EF. Urinary Incontinence in Women: Evaluation and Management. Am Fam Physician. 2019;100(6):339–48.

PubMed   Google Scholar  

Forde JC, Chughtai B, Cea M, Stone BV, Te A, Bishop TF. Trends in ambulatory management of urinary incontinence in women in the United States. Female Pelvic Med Reconstr Surg. 2017;23(4):250–5. https://doi.org/10.1097/SPV.0000000000000365 .

Peate I. Urinary incontinence in women: treatment recommendations. Br J Nurs. 2019;28(22):1486–8. https://doi.org/10.12968/bjon.2019.28.22.1486 .

Barat S, Javadian Y, Talebi G, Aghajani M. Frequency of urinary incontinence in female patients attending pelvic floor disorders clinic. J Mazandaran Univ Med Sci. 2015;24(121):189–96.

Google Scholar  

Nygaard IE, Shaw JM. Physical activity and the pelvic floor. Am J Obstet Gynecol. 2016;214(2):164–71. https://doi.org/10.1016/j.ajog.2015.08.067 .

Agarwal BK, Agarwal N. Urinary incontinence: prevalence, risk factors, impact on quality of life and treatment seeking behaviour among middle aged women. Int Surg J. 2017;4(6):953–8.

Article   Google Scholar  

Javadifar N, Komeilifar R, Afshary P, Haghighy M. Urinary incontinence and its predisposing factors in reproductive age women. Sci J Ilam Univ Med Sci. 2016;25(6):45–53.

Goforth J, Langaker M. Urinary incontinence in women. N C Med J. 2016;77(6):423–5. https://doi.org/10.18043/ncm.77.6.423 .

Lim Y-M, Lee SR, Choi EJ, Jeong K, Chung HW. Urinary incontinence is strongly associated with depression in middle-aged and older Korean women: data from the Korean longitudinal study of ageing. Eur J Obstetr Gynecol Reprod Biol. 2018;220:69–73. https://doi.org/10.1016/j.ejogrb.2017.11.017 .

Melotti IGR, Juliato CRT, Tanaka M, Riccetto CLZ. Severe depression and anxiety in women with overactive bladder. Neurourol Urodyn. 2018;37(1):223–8. https://doi.org/10.1002/nau.23277 .

Winkelman WD, Warsi A, Huang AJ, Schembri M, Rogers RG, Richter HE, Myers DL, Kraus SR, Johnson KC, Hess R, Gregory T, Bradley CS, Arya LA, Brown JS, Stone KL, Subak LL. Sleep quality and daytime sleepiness among women with urgency predominant urinary incontinence. Female Pelvic Med Reconstr Surg. 2018;24(2):76–81. https://doi.org/10.1097/SPV.0000000000000547 .

Gibson W, Hunter KF, Camicioli R, Booth J, Skelton DA, Dumoulin C, Paul L, Wagg A. The association between lower urinary tract symptoms and falls: forming a theoretical model for a research agenda. Neurourol Urodyn. 2018;37(1):501–9. https://doi.org/10.1002/nau.23295 .

Soliman Y, Meyer R, Baum N. Falls in the elderly secondary to urinary symptoms. Rev Urol. 2016;18(1):28–32.

PubMed   PubMed Central   Google Scholar  

Lim R, Liong ML, Leong WS, Lau YK, Khan NAK, Yuen KH. The impact of stress urinary incontinence on individual components of quality of life in Malaysian women. Urology. 2018;112:38–45. https://doi.org/10.1016/j.urology.2017.10.019 .

Tzur T, Yohai D, Weintraub AY. The role of local estrogen therapy in the management of pelvic floor disorders. Climacteric. 2016;19(2):162–71. https://doi.org/10.3109/13697137.2015.1132199 .

Article   CAS   PubMed   Google Scholar  

Nasaj G, Omidpour P, Shah Ali S, ZS. Relationship between pelvic floor muscle strength and severity of urinary incontinence on quality of life in women with urinary incontinence; 2017.

Schneeweiss J, Koch M, Umek W. The human urinary microbiome and how it relates to urogynecology. Int Urogynecol J. 2016;27(9):1307–12. https://doi.org/10.1007/s00192-016-2944-5 .

Pearce MM, Zilliox MJ, Rosenfeld AB, Thomas-White KJ, Richter HE, Nager CW, et al. The female urinary microbiome in urgency urinary incontinence. Am J Obstetr Gynecol. 2015;213(3):347. e1-. e1.

Alpoim PN, de Barros PM, Junqueira DR, Freitas LG, das Gracas Carvalho M, Fernández AP, et al. Preeclampsia and ABO blood groups: a systematic review and metaanalysis. Mol Biol Rep. 2012;40:2253–61.

Suskind AM, Cawthon PM, Nakagawa S, Subak LL, Reinders I, Satterfield S, Cummings S, Cauley JA, Harris T, Huang AJ, Health ABC Study. Urinary incontinence in older women: the role of body composition and muscle strength: from the health, aging, and body composition study. J Am Geriatr Soc. 2017;65(1):42–50. https://doi.org/10.1111/jgs.14545 .

Marques LP, Schneider IJC, Giehl MWC, Antes DL, d'Orsi E. Demographic, health conditions, and lifestyle factors associated with urinary incontinence in elderly from Florianópolis, Santa Catarina, Brazil. Rev Bras Epidemiol. 2015;18(3):595–606. https://doi.org/10.1590/1980-5497201500030006 .

Stickley A, Santini ZI, Koyanagi A. Urinary incontinence, mental health and loneliness among community-dwelling older adults in Ireland. BMC Urol. 2017;17(1):29. https://doi.org/10.1186/s12894-017-0214-6 .

Article   PubMed   PubMed Central   Google Scholar  

Suhr R, Lahmann NA. Urinary incontinence in home care: a representative multicenter study on prevalence, severity, impact on quality of life, and risk factors. Aging Clin Exp Res. 2018;30(6):589–94. https://doi.org/10.1007/s40520-017-0816-6 .

Sohn K, Lee CK, Shin J, Lee J. Association between female urinary incontinence and geriatric health problems: results from Korean longitudinal study of ageing (2006). Korean J Fam Med. 2018;39(1):10–4. https://doi.org/10.4082/kjfm.2018.39.1.10 .

Eshkoor S, Hamid TA, Shahar S, Mun C. Factors related to urinary incontinence among the Malaysian elderly. J Nutr Health Aging. 2017;21(2):220–6. https://doi.org/10.1007/s12603-016-0779-x .

Gallas S, Frioui S, Rabeh H, Rejeb MB. Prevalence and risk factors for urinary and anal incontinence in Tunisian middle aged women. Afr J Urol. 2018;24(4):368–73. https://doi.org/10.1016/j.afju.2018.08.002 .

Kaşıkçı M, Kılıç D, Avşar G, Şirin M. Prevalence of urinary incontinence in older Turkish women, risk factors, and effect on activities of daily living. Arch Gerontol Geriatr. 2015;61(2):217–23. https://doi.org/10.1016/j.archger.2015.06.008 .

Khan S, Ansari MA, Vasenwala SM, Mohsin Z. The influence of menopause on urinary incontinence in the women of the community: a cross-sectional study from North India. Int J Reprod Contracept Obstetr Gynecol. 2017;6(3):911–8. https://doi.org/10.18203/2320-1770.ijrcog20170555 .

Morowatisharifabad MA, Rezaeipandari H, Mazyaki A, Bandak Z. Prevalence of urinary incontinence among elderly women in Yazd, Iran: a population-based study. Elderly Health J. 2015;1(1):27–31.

Bijani A, Ghadimi R, Mikaniki E. Cohort Profile Update: The Amikola Health and Ageing Project. Caspian J Intern Med. 2017;8(3):205–12 The original cohort The Amirkola Health and Ageing Project (AHAP) is the first comprehensive cohort project of the elderly in Iran (1) that composed.1616:2011–2.

Townsend MK, Lajous M, Medina-Campos RH, Catzin-Kuhlmann A, López-Ridaura R, Rice MS. Risk factors for urinary incontinence among postmenopausal Mexican women. Int Urogynecol J. 2017;28(5):769–76. https://doi.org/10.1007/s00192-016-3196-0 .

Moudi E, Samadi F, Hosseini SR, Bijani A, Ghadimi R. Urinary incontinency in elderly women and the potential risk factors: a cohort study among the elderly women of Amirkola. J Babol Univ Med Sci. 2017;19(2):14–9.

Khanighaleejogh R, Akbari Kaji M, Shamsi A, Norrighoshki H. Prevalence of Urinary Tract Disorders in Residents of Kahrizak Elderly House. J Fac Nurs Midwifery. 2011;17:No 2.

Espino DV, Palmer RF, Miles TP, Mouton CP, Lichtenstein MJ, Markides KP. Prevalence and severity of urinary incontinence in elderly Mexican-American women. J Am Geriatr Soc. 2003;51(11):1580–6. https://doi.org/10.1046/j.1532-5415.2003.51503.x .

Maggi S, Minicuci N, Langlois J, Pavan M, Enzi G, Crepaldi G. Prevalence rate of urinary incontinence in community-dwelling elderly individuals: the Veneto study. J Gerontol Ser A Biol Med Sci. 2001;56(1):M14–M8. https://doi.org/10.1093/gerona/56.1.M14 .

Article   CAS   Google Scholar  

Ko Y, Lin S-J, Salmon JW, Bron MS. The impact of urinary incontinence on quality of life of the elderly. Am J Manag Care. 2005;11(4 Suppl):S103–11.

Ge J, Yang P, Zhang Y, Li X, Wang Q, Lu Y. Prevalence and risk factors of urinary incontinence in Chinese women: a population-based study. Asia Pac J Public Health. 2015;27(2):NP1118–NP31. https://doi.org/10.1177/1010539511429370 .

Burti JS, Santos AMB, Pereira RMR, Zambon JP, Marques AP. Prevalence and clinical characteristics of urinary incontinence in elderly individuals of a low income. Arch Gerontol Geriatr. 2012;54(2):e42–e6. https://doi.org/10.1016/j.archger.2011.04.004 .

Sumardi R, Mochtar CA, Junizaf J, Santoso BI, Setiati S, Nuhonni SA, et al. Prevalence of urinary incontinence, risk factors and its impact: multivariate analysis from Indonesian nationwide survey. Acta medica Indonesiana. 2016;46(3):175–82.

Masenga GG, Shayo BC, Msuya S, Rasch V. Urinary incontinence and its relation to delivery circumstances: a population-based study from rural Kilimanjaro, Tanzania. Plos one. 2019;14(1):e0208733. https://doi.org/10.1371/journal.pone.0208733 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Wu JM, Matthews CA, Vaughan CP, Markland AD. Urinary, fecal, and dual incontinence in older US adults. J Am Geriatr Soc. 2015;63(5):947–53. https://doi.org/10.1111/jgs.13385 .

Zhang L, Zhu L, Xu T, Lang J, Li Z, Gong J, Liu Q, Liu X. A population-based survey of the prevalence, potential risk factors, and symptom-specific bother of lower urinary tract symptoms in adult Chinese women. Eur Urol. 2015;68(1):97–112. https://doi.org/10.1016/j.eururo.2014.12.012 .

Pinkas J, Gujski M, Humeniuk E, Raczkiewicz D, Bejga P, Owoc A, Bojar I. State of health and quality of life of women at advanced age. Med Sci Monit. 2016;22:3095–105. https://doi.org/10.12659/MSM.900572 .

Jerez-Roig J, Santos MM, Souza DL, Amaral FLJ, Lima KC. Prevalence of urinary incontinence and associated factors in nursing home residents. Neurourol Urodyn. 2016;35(1):102–7. https://doi.org/10.1002/nau.22675 .

Reigota RB, Pedro AO, de Souza Santos Machado V, Costa-Paiva L, Pinto-Neto AM. Prevalence of urinary incontinence and its association with multimorbidity in women aged 50 years or older: a population-based study. Neurourol Urodyn. 2016;35(1):62–8. https://doi.org/10.1002/nau.22679 .

Sensoy N, Dogan N, Ozek B, Karaaslan L. Urinary incontinence in women: prevalence rates, risk factors and impact on quality of life. Pak J Med Sci. 2013;29(3):818–22. https://doi.org/10.12669/pjms.293.3404 .

Linde JM, Nijman RJ, Trzpis M, Broens PM. Urinary incontinence in the Netherlands: prevalence and associated risk factors in adults. Neurourol Urodyn. 2017;36(6):1519–28. https://doi.org/10.1002/nau.23121 .

Aly WW, Sweed HS, Mossad NA, Tolba MF. Prevalence and risk factors of urinary incontinence in frail elderly females. J Aging Res. 2020;2020:1–8. https://doi.org/10.1155/2020/2425945 .

Prabhu SA, Shanbhag S. Prevalence and risk factors of urinary incontinence in women residing in a tribal area in Maharashtra, India; 2013.

Liu B, Wang L, Huang S-S, Wu Q, Wu D-L. Prevalence and risk factors of urinary incontinence among Chinese women in Shanghai. Int J Clin Exp Med. 2014;7(3):686–96.

Ramage-Morin PL, Gilmour H. Urinary incontinence and loneliness in Canadian seniors: Statistics Canada. Health Rep. 2013;24(10):3–10.

Luo R, Dai W, Tay LH, Ng FC, Koh L-T. Urinary incontinence in female outpatients in Singapore. Int Urogynecol J. 2018;29(4):579–84. https://doi.org/10.1007/s00192-017-3488-z .

Matthews CA, Whitehead WE, Townsend MK, Grodstein F. Risk factors for urinary, fecal or dual incontinence in the nurses’ health study. Obstet Gynecol. 2013;122(3):539–45. https://doi.org/10.1097/AOG.0b013e31829efbff .

Schumpf LF, Theill N, Scheiner DA, Fink D, Riese F, Betschart C. Urinary incontinence and its association with functional physical and cognitive health among female nursing home residents in Switzerland. BMC Geriatr. 2017;17(1):17. https://doi.org/10.1186/s12877-017-0414-7 .

Tkacheva ON, Runikhina NK, Ostapenko VS, Sharashkina NV, Mkhitaryan EA, Onuchina JS, Lysenkov SN, Yakhno NN, Press Y. Prevalence of geriatric syndromes among people aged 65 years and older at four community clinics in Moscow. Clin Interv Aging. 2018;13:251–9. https://doi.org/10.2147/CIA.S153389 .

Schreiber Pedersen L, Lose G, Høybye MT, Elsner S, Waldmann A, Rudnicki M. Prevalence of urinary incontinence among women and analysis of potential risk factors in Germany and Denmark. Acta Obstet Gynecol Scand. 2017;96(8):939–48. https://doi.org/10.1111/aogs.13149 .

Ninomiya S, Naito K, Nakanishi K, Okayama H. Prevalence and risk factors of urinary incontinence and overactive bladder in J apanese women. LUTS. 2018;10(3):308–14. https://doi.org/10.1111/luts.12185 .

Ebbesen MH, Hunskaar S, Rortveit G, Hannestad YS. Prevalence, incidence and remission of urinary incontinence in women: longitudinal data from the Norwegian HUNT study (EPINCONT). BMC Urol. 2013;13(1):27. https://doi.org/10.1186/1471-2490-13-27 .

Cohn JA, Shah AS, Goggins KM, Simmons SF, Kripalani S, Dmochowski RR, Schnelle JF, Reynolds WS. Health literacy, cognition, and urinary incontinence among geriatric inpatients discharged to skilled nursing facilities. Neurourol Urodyn. 2018;37(2):854–60. https://doi.org/10.1002/nau.23368 .

Minassian VA, Hagan KA, Erekson E, Austin AM, Carmichael D, Bynum JP, et al. The natural history of urinary incontinence subtypes in the Nurses’ Health Studies. Am J Obstetr Gynecol. 2020;222(2):163. e1–8.

Dellú MC, Schmitt ACB, Cardoso MRA, Pereira WMP, Pereira ECA, Vasconcelos ÉSF, et al. Prevalence and factors associated with urinary incontinence in climacteric. Rev Assoc Méd Bras. 2016;62(5):441–6. https://doi.org/10.1590/1806-9282.62.05.441 .

Rashidi F, Hajian S, Darvish S, Alavi MH. Prevalence of urinary incontinence in Iranian women: systematic review and meta-analysis. Iran J Obstetr Gynecol Infertility. 2019;21(12):94–102.

Lu S, Zhang H-L, Zhang Y-J, Shao Q-C. Prevalence and risk factors of urinary incontinence among perimenopausal women in Wuhan. J Huazhong Univ Sci Technol. 2016;36(5):723–6.

Park J, Hong G-RS, Yang W. Factors associated with self-reported and medically diagnosed urinary incontinence among community-dwelling older women in Korea. Int Neurourol J. 2015;19(2):99–106. https://doi.org/10.5213/inj.2015.19.2.99 .

Altintas R, Beytur A, Oguz F, Tasdemir C, Kati B, Cimen S, Colak C, Gunes A. Assessment of urinary incontinence in the women in eastern Turkey. Int Urogynecol J. 2013;24(11):1977–82. https://doi.org/10.1007/s00192-013-2137-4 .

Wehrberger C, Madersbacher S, Jungwirth S, Fischer P, Tragl KH. Lower urinary tract symptoms and urinary incontinence in a geriatric cohort–a population-based analysis. BJU Int. 2012;110(10):1516–21. https://doi.org/10.1111/j.1464-410X.2012.11022.x .

Touhidi Nezhad F, Jalali R, Karimi F. Women's experiences of rectovaginal fistula: an ethno- religious experience. BMC Womens Health. 2020;20(1):130. https://doi.org/10.1186/s12905-020-00992-w .

Ramezanzadeh K, Sharifzadeh G, Saljughi M, Moodi M, Ibrahimzadeh A. An epidemiological study of the infectious diseases of older adults hospitalized in hospitals affiliated to Birjand University of Medical Sciences, in 2016. Iran J Ageing. 2019;14(3):298–309.

Traish AM, Vignozzi L, Simon JA, Goldstein I, Kim NN. Role of androgens in female genitourinary tissue structure and function: implications in the genitourinary syndrome of menopause. Sex Med Rev. 2018;6(4):558–71. https://doi.org/10.1016/j.sxmr.2018.03.005 .

Vakili M, Amoqadiri M, Mohammadi M, Modaressi M. Prevalence and factors associated with stress urinary incontinence after childbirth in Yazd in 2015. SSU_J. 2016;24(9):716–23.

Mansourghanaie M, Asgari A. Related factors associated with urinary incontinence in women who had cystocel and rectocel surgery. J Holistic Nurs Midwifery. 2015;25(1):73–7.

Hagan KA, Erekson E, Austin A, Minassian VA, Townsend MK, Bynum JP, et al. A prospective study of the natural history of urinary incontinence in women. Am J Obstetr Gynecol. 2018;218(5):502. e1–8.

Feeney C, Gorman G, Stefanetti R, McFarland R, Turnbull D, Harding C, Sachdeva A. Lower urinary tract dysfunction in adult patients with mitochondrial disease. Neurourol Urodyn. 2020;39(8):2253–63.

Burge AT, Lee AL, Kein C, Button B, Sherburn M, Miller B, et al. Prevalence and impact of urinary incontinence in men with chronic obstructive pulmonary disease: a questionnaire survey. Physiotherapy. 2017;103(1):53–8. https://doi.org/10.1016/j.physio.2015.11.004 .

Download references


We hereby express our gratitude and appreciation to the school of nursing and midwifery of Kermanshah university of medical sciences.

Funding for this research was provided by the deputy of research and technology –Kermanshah University of Medical Sciences, (990423), the deputy of research and technology –Kermanshah University of Medical Sciences had no role in the design of the study and collection, analysis, and interpretation of data and in writing of the manuscript.

Author information

Authors and affiliations.

Department of Nursing, School of Nursing and Midwifery, Kermanshah University of Medical Sciences, Kermanshah, Iran

Sedighe Batmani, Rostam Jalali & Masoud Mohammadi

Faculty of Health and Life Sciences, School of Life Sciences, Coventry University, Coventry, UK

Shadi Bokaee

You can also search for this author in PubMed   Google Scholar


RJ and SB1 and MM contributed to the design, MM statistical analysis, participated in most of the study steps. SB1 and MM prepared the manuscript. SB1 and RJ and SB2 assisted in designing the study, and helped in the, interpretation of the study. All authors have read and approved the content of the manuscript.

Corresponding author

Correspondence to Masoud Mohammadi .

Ethics declarations

Ethics approval and consent to participate.

This study was recorded in the ethics committee of Kermanshah University of medical sciences with the ethics code of (IR.KUMS.REC.1399.455) .

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Cite this article.

Batmani, S., Jalali, R., Mohammadi, M. et al. Prevalence and factors related to urinary incontinence in older adults women worldwide: a comprehensive systematic review and meta-analysis of observational studies. BMC Geriatr 21 , 212 (2021). https://doi.org/10.1186/s12877-021-02135-8

Download citation

Received : 11 August 2020

Accepted : 07 March 2021

Published : 29 March 2021

DOI : https://doi.org/10.1186/s12877-021-02135-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

  • Urinary incontinence
  • Older adults
  • Meta-analysis

BMC Geriatrics

ISSN: 1471-2318

nursing research articles on incontinence

  • Research article
  • Open access
  • Published: 20 April 2020

An evidence-based self-management package for urinary incontinence in older women: a mixed methods feasibility study

  • E. Andrea Nelson 2 &
  • Linda McGowan 3  

BMC Urology volume  20 , Article number:  43 ( 2020 ) Cite this article

2332 Accesses

6 Citations

4 Altmetric

Metrics details

Urinary incontinence (UI) is a distressing condition that limits women’s quality of life and places a heavy burden on health care services. Behavioural treatments are recommended as a first-line treatment. An evidence-based self-management package was developed following the Medical Research Council (MRC) framework for complex interventions. This study aimed to evaluate the feasibility and acceptability of the intervention.

A mixed-methods approach was undertaken, namely a randomised controlled feasibility study with nested qualitative study. Fifty women aged 55 or over living with UI, recruited from community centres were randomly assigned to either a 3-month course with the package with a support session or a control group to receive the same package only 3 months later. Principal outcome measures were: self-reported quality of life, UI severity, self-efficacy and psychological status. Analysis of covariance was undertaken to estimate within- and between- group changes for all outcomes. Acceptability was explored using individual interviews at follow-up.

Fifty women were randomised (24 to intervention, 26 to control); mean age of 69.7 (±9.1) years and mean UI frequency 2.2 (±2.2) episodes/day at baseline. Overall, 49 women (98%) completed 3-month follow-up (24 in the intervention, 25 in the control). A positive trend was detected in the impact of UI on their personal relationships (− 3.89, p  = 0.088), symptom severity (− 1.77, p  = 0.025), UI symptoms scale (− 1.87, p  = 0.031) and anxiety status (− 2.31, p  = 0.001), respectively. Changes in quality of life and self-efficacy did not differ significantly between groups. Majority of women (71%) in the intervention group reported subjective improvement after 3 months. Spearman correlation coefficient was 0.43 ( p  < 0.05) between their subjective perception of change and self-efficacy. Women perceived the package being acceptable and described that the package had the potential to increase their knowledge and confidence to manage symptoms and improve quality of life.


The study demonstrated that the self-management package is feasible and acceptable for older women with UI. Further studies are needed with a large sample size in clinical settings to evaluate the effectiveness of this package.

Trial registration

ISRCTN17194896 . Registered on 11th September 2019 (retrospectively registered).

Peer Review reports

Urinary incontinence (UI) is a distressing condition that limits people’s quality of life and places a burden on those affected and health and social care services [ 1 , 2 ]. Prevalence in women is estimated to be around about 40%, with an increase with age [ 3 ]. Women living with UI often experience functional limitations and social embarrassment, negatively impacting their mental health [ 4 ]. Although several options are available for treating and managing UI, behavioural treatments are recommended as a first-line treatment therapies for stress, urgency and mixed UI by international UI guidelines developed worldwide [ 5 ], prior to considering more intensive treatments.

Behavioural treatment programmes focusing on single elements of behavioural strategies have been extensively researched, e.g. pelvic floor muscle exercises (PFME) [ 6 ]. These programmes are sometimes challenging to deliver as they need multiple appointments, the involvement of specialised practitioners, and may lack flexibility to respond to individual needs associated with comorbidities in older women. Many women therefore choose to disengage with the service. Only a third of women with UI consult a doctor in European countries such as France, Germany, Spain and the UK, and 20 to 25% of those experiencing significant clinical symptoms seek care and less than half of them receive treatment [ 7 ]. A multifaceted intervention involving behavioural strategies may be more effective than a single component for the management of UI in older women [ 8 , 9 ].

Self-management for chronic conditions is multidimensional and defined as an intervention designed to develop individuals’ knowledge, skills or psychological and social resources and their ability to manage their health condition and consequences, through education, training and support [ 10 ]. Many self-management interventions have been developed to support people to cope with their health conditions and improve quality of life. Positive outcomes reported include a higher degree of self-efficacy, reduced physical interference, and improved mental health status at follow-up [ 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 ]. Although evidence for self-management programmes incorporating multifaceted behavioural treatments for use in UI among older women is currently limited, our systematic review concluded that multifaceted self-management interventions including PFME, bladder retraining, or combination behavioural techniques are beneficial.

Following the Medical Research Council (MRC) framework for developing and evaluating complex interventions [ 22 ], a self-management package was co-developed with older women with UI and health professionals providing treatment and care. The initial draft of the intervention was informed by synthesis of data from a systematic review and stakeholder interviews with women and health professionals. Both groups preferred an evidence-based self-management package that had the capacity to meet women’s individual needs with the flexibility to modify behaviours in coping with UI without significant service provision and/or intensive interaction with health professionals. The initial draft was reviewed and discussed in detail with an expert group which consisted of four women with UI, five health professionals and two lay members. Consensus was reached on the content and format of this self-management package for older women with UI using a normal group technique [ 23 ]. However, the feasibility and acceptability of the intervention remained unknown.

The aim of this study was to evaluate the feasibility, acceptability, and preliminary outcomes of an evidence-based self-management package for UI in older women. Specifically, it was to 1) assess the feasibility of the intervention to evaluate the effectiveness of a self-management package; 2) assess the variation of the main outcome measures to inform sample size considerations for a full randomised controlled trial (RCT); 3) provide preliminary data on the effect of the intervention; 4) understand the benefits/limitations and acceptability of the self-management package compared to no active treatment.

A mixed-methods approach comprising a two-arm RCT feasibility study with a nested qualitative study was undertaken. The qualitative study was conducted to understand how the intervention might work and explore facilitators and barriers to acceptance of the intervention [ 24 ]. This study presented adheres to CONSORT guidelines.

Patient and public engagement

A project advisory group comprising three older women (aged 55 plus) living with UI and one nurse working in a community continence clinic was set up prior to the commencement of this study, to ensure this study addressed issues that were important and relevant to women. The meeting was led by YF and facilitated by LMc/EAN every 6 months. Participants were provided with background information and clinical guidelines for UI management and consulted for their current experiences and expectations of managing UI. They also reviewed the findings of the systematic review and interview transcripts. All highlighted the need for evidence-based practice that has the ability to engage the wider population and indicated a willingness to use it for UI self-management. The group also supported the application of ethics approval and development of interview schedules.

This is a single-centre, randomised controlled feasibility study of self-management package versus wait-list control with 12- week post-intervention follow up. Participants were recruited using flyers posted in community centres of a local Forum for older people in West Yorkshire, UK. A short presentation was also given by the project researcher in most centres.


Women were eligible if they were aged 55 or over, had self-reported symptom of involuntary leakage of urine and were able to read and speak English. Individuals who self-reported their UI were caused by neurological diseases affecting the brain and spinal cord, or were cognitively impaired were excluded.

A consecutive sampling strategy was applied to recruit participants in this study. Assuming an attrition rate of 20%, a target of 25 per arm was sufficient to have 20 participants in each arm by the end of the study. This has been recommended as acceptable for a feasibility trial assuming at worst a medium effect size (Cohen’s d = 0.2) for a continuous outcome and 80% power [ 25 ].


All participants were recruited prior to randomisation to the intervention. Eligible women were randomised using a 1:1 ratio to either the intervention or control group. The randomisation procedure was performed by a web-based randomisation service ( https://sealedenvelope.com/ ). Due to the nature of the intervention, this study is considered open-label and is not allocation concealed because the researcher delivering the interventions and the participants were aware of group allocation at the time of implementation.


The experimental intervention was the self-management package, co-developed with older women living with UI, health professionals and lay members. The aim of the package was to provide information and practical skills for women to self-manage their UI and other symptoms. Following elements were included: recognition and awareness, getting the support you need, understanding the cause, learning to manage your UI, developing a self-management plan and how can you find out more. Descriptions of self-management techniques such as PFME, bladder training and lifestyle interventions were also provided. The researcher acted as a facilitator and delivered the intervention in person immediately after the completion of baseline data collection. A self-management brochure was also distributed. The intervention group could request a single one-hour support session with the researcher if they had any difficulties in using the package.

The control group did not receive the self-management package or the support session. However they had been given a copy of the same package upon the completion of their follow-up data collection (at 3 months).

Outcome measures

As this was a feasibility study it was appropriate to explore a range of outcome measures. Standardised self-reported measures were used to assess participants’ generic and disease-specific quality of life, UI severity, self-efficacy and psychological health respectively: the EuroQol (EQ-5D-5 L) [ 26 ], King’s Health Questionnaire (KHQ) [ 27 ], International Consultation on Incontinence Questionnaire – urinary incontinence short form (ICIQ-UI SF) [ 28 , 29 ], Geriatric Self-Efficacy index for urinary incontinence (GSE-UI) [ 30 ], and Hospital Anxiety and Depression Scale (HADS) [ 31 ]. Data were collected at baseline and three-month follow-up. Patient Global Impression: Improvement (PGI-I) was obtained from the intervention group only at three-month follow up [ 32 ]. These measures have been commonly used in research and practice for women with UI.

Data analysis

Descriptive analyses were undertaken to establish recruitment, drop-out rates and the distribution of baseline and follow-up characteristics and outcomes. For each of the above measures, analysis of covariance (ANCOVA) was conducted adjusting for their baseline values to take account of outcome imbalance at baseline and estimate the impact of intervention compared with control [ 33 , 34 ]. T-tests and Chi-square tests were applied to compare two groups at baseline and follow-up for continuous and categorical variables, respectively. Intention-to-treat analysis was performed for this study. Data from all subjects were included in the analysis as randomised. To test the appropriateness of the analysis, complete case analysis was undertaken for each of the above outcome measures to identify the relative treatment effects using a linear (for continuous variables) and logistic (for binary outcomes) mixed model. For all analyses, a two-tailed p -value less than 0.05 was considered statistically significant and a p-value between 0.05 and 0.1 was interpreted as indicating a trend [ 35 , 36 ].

Nested qualitative study

Women in the intervention group were eligible to participate in this subsequent qualitative study aiming to understand the acceptability of the package. They were purposively recruited considering different types of UI, number of years living with UI, to enable wider discussions on the self-management package and their experience. The concept of data saturation guided sample size for this qualitative study [ 37 ].

Semi-structured individual interviews using open-ended questions were undertaken at follow-up. Interviews were conducted either face-to-face in the participants’ homes, in a meeting room at the University, or by telephone, based on participants’ preferences. Interviews were digitally recorded with permission and transcribed verbatim for analysis. Interviews focused on exploring participants’ experience of managing symptoms guided by the intervention, facilitators and barriers to the use of the package, comments on the content and format, feedback on outcome measures used and suggestions on dissemination strategy.

Data analysis was commenced during the interview phase and the transcription, and continued during the analysis of the transcriptions, hence early commencement of analysis facilitated the development of subsequent interviews [ 38 ]. Data were analysed using thematic analysis [ 39 ], involving a six-step procedure: familiarising the data, generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report. Although they were presented as a step-by-step procedure, the analysis was an iterative and reflexive process to finalise the themes. To ensure trustworthiness and rigour of the analysis, we double coded the data as a validity check and explored alternative interpretations of the data and through discussion with members of the research team. Interview transcripts were managed and analysed using Nvivo, a qualitative software programme for organising and coding the data [ 40 ].

Participants characteristics

Overall 50 participants ( M age =69.7, SD  = 9.1), were randomly allocated to either intervention ( n  = 24, M age =69.5, SD  = 8.9) or control group ( n  = 26, M age =69.8, SD  = 9.5). During the 3 months follow-up, no one in the intervention group requested the support session with the researcher. One woman in the control group dropped out due to admission to the hospital for other health conditions. A total of 49 participants with a mean age 69.6 (SD = 9.1) were included in the complete case analysis. No significant differences were found for all demographics and UI characteristics collected at baseline, indicating that randomisation was achieved between the intervention and control groups. The majority were white British (77.6%), completed secondary and higher education (91.8%), were retired (71.4%), and had delivered 2 or more children (71.5%). The number of participants was similar across different UI types. Please see participant’s characteristics in Table  1 .

Medians, interquartile range (IQR), mean, standard deviation (SD) and p -values of all baseline and follow-up outcomes are shown in Tables  2 and 3 , respectively. There were no differences shown in general quality of life, UI symptoms scale and emotional health status at baseline, however groups differed with regard to KHQ severity measures (34.03 vs 48.67, p  = 0.028) and self-efficacy (84.96 vs 57.76, p  < 0.001). At follow-up, differences were detected in KHQ symptom severity (6.38 vs 9.04, p  = 0.006), UI symptoms scale (5.38 vs 7.76, p  = 0.022), self-efficacy (89.13 vs 69.08, p  = 0.007), and anxiety (5.13 vs 7.80, p  = 0.049) at 3-month follow-up.

The effect of the self-management intervention on outcome measures between intervention and control groups were demonstrated in Table  4 . The difference captured on each outcome measure was further shown in Figs.  1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 and 9 , respectively. Participants’ general quality of life were improved within both groups after the follow-up, however no significant difference was identified (see Figs. 1 and 2 ). Similarly, no difference was shown in UI specific quality of life (see Figs.  3 and 4 ), self-efficacy (see Fig.  7 ), or depression status (see Fig.  9 ) between groups. However, significant differences were detected in participants’ KHQ symptom severity (− 1.77, p  = 0.025; also see Fig.  5 ), ICIQ-UI SF symptoms scale (− 1.87, p  = 0.031; also see Fig.  6 ), and their anxiety status (− 2.31, p  = 0.001; also see Fig.  8 ), respectively. A positive trend was also observed in the participants’ KHQ personal relationships (− 3.89, p  = 0.088; also see Fig.  4 ).

figure 1

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for EQ-5D

figure 2

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for EQ-5D utility and VAS scores

figure 3

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for KHQ, Part I

figure 4

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for KHQ, Part II

figure 5

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for KHQ, Part III

figure 6

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for ICIQ-UI SF

figure 7

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for GSE-UI

figure 8

Mean (95 and 90% confidence intervals) of pre-post intervention differences and treatment effect for HADS, Anxiety

figure 9

The relationship between the participants’ perceived improvement and the difference in outcomes within the intervention group was assessed using Spearman correlations (Table  5 ). There was a significant correlation shown between PGI-I and GSE-UI ( r s = 0.43, p  < 0.05). Mean change in GSE-UI for participants who responded “better” on the PGI-I were significantly higher than those who responded “no change” or “worse”.

Qualitative findings

Individual interviews were undertaken with 15 participants from the intervention group at their homes. Participants ranged in age from 55 to 86 years. The majority were retired (11; 73%) and had suffered from UI for more than 5 years (11; 73%), but had received no support (13; 87%) from health services.

Each interview lasted between 15 and 48 min. A total of four themes were generated. Facilitators and barriers to the use of the self-management package followed by suggestions were also explored under each theme.

Raising awareness and gaining/refreshing knowledge

All participants found that this package had raised their attention to UI management as well as the resources and support available. Some participants learned new knowledge and management skills, while others found it useful to help refresh their memory. Positive comments were received on the fact that this package gathered both information and practical skills in a systematic and structured manner, accompanied by tips shared by women and health professionals.

“Yeah, I found the information on how to do the pelvic floor muscle exercises really good, it describes exactly what you should do and how much you should do..” (W5)

A few participants also expressed the need for more detailed information and tips to meet their needs. For instance, information and explanation were desired for biofeedback, vaginal cones, pads and how certain lifestyles impacted on UI symptoms and severity. In addition, some participants shared strategies and experiences of managing UI, including yoga, weight management programmes, and using reminders to practice PFME.

Half of the participants also identified the bladder diary as a potential barrier to the use of this book. For people who were in employment or regularly involved in outdoor activities, they felt that it was difficult and infeasible to measure intake of fluids and volume voided.

Feeling more confident and motivated to management UI

Participants generally felt more equipped and confident in managing their UI with resources and support supplied in the package.

“…we can do this, at least we know where to contact, we can tell the doctors also but if it happens to somebody you can tell them and they can do this.”(W24)

Participants also expressed their desire to motivate for self-management, however many of them did not practice as much as they would like to due to other commitments. Hence they suggested the inclusion of information and strategies on how to stay motivated and develop adherence.

“I suspect a lot of people would read the book, and think, oh yeah, that’s a really good idea, I ought to be doing that, and then never, ever actually do it.”(W46)

Being a useful and user-friendly package

All participants preferred the fact that this package was colour coded, making it easy to read and follow. They found the texts being easy to read and understand and the illustrations being appropriate, indicating that there was no need to book the support session. Some participants described benefit gained as a result of using this package.

“The book was very well put together, and very clear, very easy to read, and it was very informative.” (W6)
“I’m still having the occasional accident, but now I’m sort of going oh well, I will get there. It’s not a major setback. But it’s not affecting my life like it used to.” (W5)

Some of them preferred a smaller size of the package so they could carry in their handbags, whereas some believed that people tended to put it away if it was a small book. A few concerned about the phrase “older women” used in the title of the package and suggested to remove it or simply indicate “women over 55”.

Exploring dissemination strategies

All participants found the self-reported questionnaires in the feasibility study acceptable and considered them being necessary. Some appreciated being involved in this study. All participants recommended places where appropriate to distribute the package, including doctor’s surgeries, hospitals, pharmacies, workplaces, public libraries, local leisure centres, and other meeting places for women.

Some participants also expressed the need to design the package in multiple languages groups who may have limited access to support and resources. In addition to delivering the package to individuals, a couple of participants expressed their preference for a group to enable peer support.

This results of this study provide supporting evidence for the feasibility and acceptability of the self-management package as a complementary strategy for the management of UI in older women. This results of the feasibility RCT have confirmed that recruitment of women with UI from local community centres who were not actively engaged with health services is feasible and the drop-out rate is low. Completion and follow up of the questionnaires were successful. The findings of the acceptability study have suggested that the package may be a useful resource in improving women’s knowledge and practical skills for the management of UI. Women also expressed the need for such intervention for this condition and commented that questionnaires used to capture information were acceptable.

The feasibility RCT showed that women’s UI severity, symptoms and anxiety status were improved after the intervention for 3 months. This provides initial evidence that this self-management package has the capacity to reduce both physical and emotional impact of UI on women. There is an increasing interest in the self-management of UI consisting in both research and practice, however many interventions only focused on one particular skill or technique [ 6 ]. The UK Continence Society has recommended a support package for patients with UI as a minimum standard for continence care [ 41 ]. With no evidence-based package being available, this study developed the intervention consisting of multiple components and also demonstrated the acceptability of the intervention and potential benefit on women’s physical and psychological health. Therefore, future studies are needed to evaluate the intervention in other settings in a larger multicentre trial in order to gain robust evidence.

Although the difference in women’s quality of life or self-efficacy was not detected, women described that they felt more confident and motivated to self-manage their condition and some of them gained improvement after using the package. Qualitative interviews further supported these findings and participants reported that the package was informative and useful in raising their awareness. A significant correlation was observed between women’s self-efficacy and their subjective perception of improvement. This suggests that women who gained a higher level of self-efficacy were likely to experience an improved subjective perception of their UI condition. Therefore, individual’s self-efficacy level needs to be assessed as a way to stratify participants before randomisation process. For those with extremely low self-efficacy, interventions with a purpose to build their confidence may be needed before providing self-management package.

Existing literature has shown modest benefit of self-management interventions in older women with UI. Compared to pharmacological therapies alone, nonpharmacological interventions, such as behavioural therapy either alone or combined with other intervention, are more effective in achieving cure or improvement both stress and urge UI [ 42 ]. Clinical effectiveness of self-management of UI has also been with respect to symptoms, UI severity, quality of life and perceptions of improvement when being delivered in group format with effects maintained (12 months [ 43 ]. Similarly, this study has observed improvement in symptoms and severity with the self-management package, and effects on psychological health. This package appears feasible to modify women’s quality of life living with UI. Further studies may be needed to investigate the effectiveness of self-management interventions on both physical and psychosocial outcomes in women with UI, at both short and long term.

Strengths and limitations

The intervention was co-developed with key stakeholders including women with UI and health professionals. The feasibility and acceptability of the intervention were evaluated with women who were not actively engaged with clinical services for UI. Potential positive effects were observed in women’s UI symptoms and their psychological status, which were further supported by the women’s experiences. However, careful interpretation is needed. The study was not allocation concealed and the sample size was not sufficiently powered to make robust inferences about the effectiveness of the intervention. Further studies are needed with a large sample size in multiple clinical settings to evaluate the effectiveness of this package. There is a possibility of bias associated with the self-reported measures. Objective information such as pad test or measure of PFM strength may be needed. It is worth noting that pad tests are not recommended in the routine assessment of women with UI by the NICE [ 44 ]. Also, women may be less in favour of the invasiveness of the measures such as vaginal dynamometer.

The self-management package consisting of multifaceted behavioural techniques appears feasible to improve UI symptoms, severity and anxiety status in older women living with UI. Women also found it useful for increasing their awareness, motivation and knowledge of UI management. This study suggested that this package has the potential to be implemented in routine practice with further evaluation of effectiveness in clinical settings.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.


Medical Research Council

Pelvic floor muscle exercises

  • Randomised controlled trial
  • Urinary incontinence

Abrams P, Andersson KE, Apostolidis A, Birder L, Bliss D, Brubaker L, Cardozo L, Castro-Diaz D, O'Connell PR, Cottenden A, et al. 6th international consultation on incontinence. Recommendations of the international scientific committee: evaluation and treatment of urinary incontinence, pelvic organ prolapse and faecal incontinence. Neurourol Urodyn. 2018;37(7):2271–2.

Article   PubMed   Google Scholar  

Rubilotta E, Balzarro M, D’Amico A, Cerruto MA, Bassi S, Bovo C, Iacovelli V, Bianchi D, Artibani W, Finazzi Agrò E. Pure stress urinary incontinence: analysis of prevalence, estimation of costs, and financial impact. BMC Urol. 2019;19(1):44.

Article   PubMed   PubMed Central   Google Scholar  

Cooper J, Annappa M, Quigley A, Dracocardos D, Bondili A, Mallen C. Prevalence of female urinary incontinence and its impact on quality of life in a cluster population in the United Kingdom (UK): a community survey. Primary Health Care Res Dev. 2015;16(4):377–82.

Article   Google Scholar  

Abrams P, Smith AP, Cotterill N. The impact of urinary incontinence on health-related quality of life (HRQoL) in a real-world population of women aged 45–60 years: results from a survey in F rance, G ermany, the UK and the USA. BJU Int. 2015;115(1):143–52.

Syan R, Brucker BM. Guideline of guidelines: urinary incontinence. BJU Int. 2016;117(1):20–33.

Dumoulin C, Hay-Smith J, Habée-Séguin GM, Mercier J. Pelvic floor muscle training versus no treatment, or inactive control treatments, for urinary incontinence in women: a short version Cochrane systematic review with meta-analysis. Neurourol Urodyn. 2015;34(4):300–8.

O’Donnell M, Lose G, Sykes D, Voss S, Hunskaar S. Help-seeking behaviour and associated factors among women with urinary incontinence in France, Germany, Spain and the United Kingdom. Eur Urol. 2005;47(3):385–92.

Tannenbaum C, Agnew R, Benedetti A, Thomas D, van den Heuvel E. Effectiveness of continence promotion for older women via community organisations: a cluster randomised trial. BMJ Open. 2013;3(12):e004135.

Tannenbaum C, Fritel X, Halme A, van den Heuvel E, Jutai J, Wagg A. Long-term effect of community-based continence promotion on urinary symptoms, falls and healthy active life expectancy among older women: cluster randomised trial. J Age Ageing. 2019;48(4):526–32. https://doi.org/10.1093/ageing/afz038 .

Taylor SJ, Pinnock H, Epiphaniou E, Pearce G, Parke HL, Schwappach A, Purushotham N, Jacob S, Griffiths CJ, Greenhalgh T. A rapid synthesis of the evidence on interventions supporting self-management for people with long-term conditions: PRISMS–practical systematic review of self-management support for long-term conditions; 2014.

Google Scholar  

Gurden M, Morelli M, Sharp G, Baker K, Betts N, Bolton J. Evaluation of a general practitioner referral service for manual treatment of back and neck pain. Primary Health Care Res Dev. 2012;13(3):204.

Lorig K. Self-management of chronic illness: a model for the future. Generations. 1993;17(3):11–4.

Lorig K. Self-management education: more than a nice extra. Med Care. 2003;41(6):699–701.

PubMed   Google Scholar  

Lorig K, González VM, Laurent DD, Morgan L, Laris B. Arthritis self-management program variations: three studies. Arthritis Rheumatism. 1998;11(6):448–54.

Article   CAS   Google Scholar  

Lorig K, Holman H. Arthritis self-management studies: a twelve-year review. Health Educ Q. 1993;20(1):17–28.

Barlow J, Wright C, Sheasby J, Turner A, Hainsworth J. Self-management approaches for people with chronic conditions: a review. Patient Educ Couns. 2002;48(2):177–87.

Barlow JH, Turner AP, Wright CC. A randomized controlled study of the arthritis self-management Programme in the UK. Health Educ Res. 2000;15(6):665–80.

Article   CAS   PubMed   Google Scholar  

Bourbeau J, Van Der Palen J. Promoting effective self-management programmes to improve COPD. Eur Respir J. 2009;33(3):461–3.

Effing TW, Bourbeau J, Vercoulen J, Apter AJ, Coultas D, Meek P, van der Valk P, Partridge MR, van der Palen J. Self-management programmes for COPD moving forward. Chronic Respiratory Dis. 2012;9(1):27–35.

Lennon S, McKenna S, Jones F. Self-management programmes for people post stroke: a systematic review. Clin Rehabil. 2013;0269215513481045.

Smith-Turchyn J, Morgan A, Richardson J. The effectiveness of group-based self-management Programmes to improve physical and psychological outcomes in patients with Cancer: a systematic review and meta-analysis of randomised controlled trials. Clin Oncol. 2015;28(5):292–305.

Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. Int J Nurs Stud. 2013;50(5):587–92.

Lennon R, Glasper A, Carpenter D: Nominal Group Technique: its utilisation to explore the rewards and challenges of becoming a mental health nurse, prior to the introduction of the all graduate nursing curriculum in England. In . : Working Papers in Health Sciences 1: 2 ISSN 2051–6266/20120000; 2012.

Campbell M, Fitzpatrick R, Haines A, Kinmonth AL, Sandercock P, Spiegelhalter D, Tyrer P. Framework for design and evaluation of complex interventions to improve health. Bmj. 2000;321(7262):694–6.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Whitehead AL, Julious SA, Cooper CL, Campbell MJ. Estimating the sample size for a pilot randomised trial to minimise the overall trial sample size for the external pilot and main trial for a continuous outcome variable. Stat Methods Med Res. 2016;25(3):1057–73.

Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, Bonsel G, Badia X. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res. 2011;20(10):1727–36.

Kelleher C, Cardozo L, Khullar V, Salvatore S. A new questionnaire to assess the quality of life of urinary incontinent women. BJOG Int J Obstet Gynaecol. 1997;104(12):1374–9.

Avery K, Donovan J, Peters TJ, Shaw C, Gotoh M, Abrams P. ICIQ: a brief and robust measure for evaluating the symptoms and impact of urinary incontinence. Neurourol Urodynamics. 2004;23(4):322–30.

Klovning A, Avery K, Sandvik H, Hunskaar S. Comparison of two questionnaires for assessing the severity of urinary incontinence: the ICIQ-UI SF versus the incontinence severity index. Neurourol Urodynamics. 2009;28(5):411–5.

Tannenbaum C, Brouillette J, Korner-Bitensky N, Dumoulin C, Corcos J, Tu LM, Lemieux MC, Ouellet S, Valiquette L. Creation and testing of the geriatric self-efficacy index for urinary incontinence. J Am Geriatr Soc. 2008;56(3):542–7.

Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361–70.

Yalcin I, Bump RC. Validation of two global impression questionnaires for incontinence. Am J Obstet Gynecol. 2003;189(1):98–101.

Vickers AJ, Altman DG. Analysing controlled trials with baseline and follow up measurements. Bmj. 2001;323(7321):1123–4.

Senn SS: Statistical issues in drug development, vol. 69. United States: Wiley; 2008.

Drummond GB, Tom BD. Statistics, probability, significance, likelihood: words mean what we define them to mean. Adv Physiol Educ. 2011;35(4):361–4.

Halsey LG, Curran-Everett D, Vowler SL, Drummond GB. The fickle P value generates irreproducible results. Nat Methods. 2015;12(3):179.

Guest G, Bunce A, Johnson L. How many interviews are enough? An experiment with data saturation and variability. Field Methods. 2006;18(1):59–82.

Morse JM, Field PA. Principles of data analysis. In: Nursing Research edn. United States: Springer; 1996. p. 103–23.

Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3(2):77–101.

QSR International: NVivo qualitative data analysis software: QSR international Pty ltd; 2018.

Minimum Standards for Continence Care in the United Kingdom. http://www.ukcs.uk.net/wp-content/uploads/2016/04/15091716_Revised_Min_Standards_for_CC_in_UK.pdf . Accessed 16 Nov 2019.

Balk EM, Rofeberg VN, Adam GP, et al. Pharmacologic and nonpharmacologic treatments for urinary incontinence in women: a systematic review and network meta-analysis of clinical outcomes. Ann Intern Med. 2019;170:465–79. [Epub ahead of print 19 March 2019]. https://doi.org/10.7326/M18-3227 .

Diokno AC, Newman DK, Low LK, Griebling TL, Maddens ME, Goode PS, Raghunathan TE, Subak LL, Sampselle CM, Boura JA, et al. Effect of group-administered behavioral treatment on urinary incontinence in older women: a randomized clinical TrialEffect of group-administered behavioral treatment on urinary incontinence in older WomenEffect of group-administered behavioral treatment on urinary incontinence in older women. JAMA Intern Med. 2018;178(10):1333–41.

NICE. Urinary incontinence in women: management. In . Edited by National Institute for health and care excellence. London: NICE; 2013.

Download references


We thank women and health professionals’ involvement and their contribution to this study.

This work was supported by Leeds Benevolent Society for Single Ladies (LBSSL) at UK, a Charitable Incorporated Organisation (registered charity number 1155794). The funder had no role in the design of the study, collection, analysis and interpretation of data, the views expressed are those of the authors and not necessarily those of the LBSSL.

Author information

Authors and affiliations.

Leeds Institute of Health Sciences, University of Leeds, Leeds, LS2 9NL, UK

School of Health and Life Sciences, Glasgow Caledonian University, Glasgow, UK

E. Andrea Nelson

School of Healthcare, University of Leeds, Leeds, UK

Linda McGowan

You can also search for this author in PubMed   Google Scholar


All authors have read the manuscript, discussed the results and commented on the manuscript. All authors have met the criteria for authorship and approved the final. YF: study design, data collection and analysis, drafting and revising the manuscript. EAN: study design and analysis, revising the manuscript. LM: study design and analysis, revising the manuscript.

Corresponding author

Correspondence to Yu Fu .

Ethics declarations

Ethics approval and consent to participate.

Ethics approval was granted by the School of Healthcare Research Ethics Committee (HREC 18–001) at the University of Leeds, UK. All of the data were collected with the written informed consent of participants provided prior to the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Cite this article.

Fu, Y., Nelson, E.A. & McGowan, L. An evidence-based self-management package for urinary incontinence in older women: a mixed methods feasibility study. BMC Urol 20 , 43 (2020). https://doi.org/10.1186/s12894-020-00603-8

Download citation

Received : 15 January 2020

Accepted : 13 March 2020

Published : 20 April 2020

DOI : https://doi.org/10.1186/s12894-020-00603-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

  • Self-management
  • Feasibility
  • Acceptability
  • Mixed methods

BMC Urology

ISSN: 1471-2490

nursing research articles on incontinence

U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings
  • Advanced Search
  • Journal List
  • HHS Author Manuscripts

Logo of nihpa

Incidence and Predictors of Incontinence Associated Skin Damage in Nursing Home Residents with New Onset Incontinence

Donna z. bliss.

a University of Minnesota School of Nursing, Minneapolis, MN

Michelle A. Mathiason

Olga gurvich, lynn e. eberly.

b School of Public Health Division of Biostatistics, Minneapolis, MN

Jessica Fisher

Kjerstie r. wiltzen, haley akermark, amanda hildebrandt, megan jacobson, taylor funk, amanda beckman, reed larson.

The purpose of this study was to determine the incidence and predictors of incontinence associated dermatitis (IAD) in nursing home residents.

Records of a cohort of 10,713 elderly (aged 65+) newly incontinent nursing home residents in 448 nursing homes in 28 states free of IAD were followed for IAD development. Potential multi-level predictors of IAD were identified in four national datasets containing information about the characteristics of individual nursing home residents, nursing home care environment, and communities in which the nursing homes were located. A unique set of health practitioner orders provided information about IAD and the predictors of IAD prevention and pressure injuries in the extended perineal area. Analysis was based on hierarchical logistical regression.

The incidence of IAD was 5.5%. Significant predictors of IAD were not receiving preventive interventions for IAD, presence of a perineal pressure injury, having greater functional limitations in activities of daily living, more perfusion problems, and lesser cognitive deficits.

Findings highlight the importance of prevention of IAD and treatment/prevention of pressure injuries. A Wound Ostomy and Continence (WOC) nurse offers expertise in these interventions and can educate staff about IAD predictors which can improve resident outcomes. Other recommendations include implementing plans of care to improve functional status, treat perfusion problems, and provide assistance with incontinence and skin care to residents with milder as well as greater cognitive deficits.


Incontinence can lead to inflammation and damage of the skin resulting in adverse symptoms of discomfort/pain, itching, burning-like sensations, and secondary fungal infection. 1 , 2 Incontinence associated dermatitis (IAD), has been classified as a type of moisture associated skin damage (MASD) of the superficial skin barrier. 3 It afflicts patients in multiple care settings from critical care to long-term care. 1 , 4 – 8 IAD incidence has been reported to be approximately 29% – 36% in critical care patients, 5 , 6 7% in long-term acute care residents, 7 and 3% to 4% among nursing homes residents on skin damage prevention programs. 8 , 9 Numerous characteristics of a patient have been proposed as risk factors for IAD by clinical experts based on practice experience and theoretical knowledge, 2 , 10 – 12 but supportive evidence remains sparse. In a cross-sectional study of nursing home residents, Bliss and colleagues 13 reported that skin damage in the perineal area was associated not only with incontinence (mainly fecal and dual incontinence) but with functional, physical, cognitive, and mobility problems of residents. Their multivariate analysis adjusted for effects of other variables in the model. In a cross-sectional study in which data of patients from multiple care settings (hospitals, nursing homes, and home care) were combined, Kottner and associates 14 examined two-way associations between IAD and variables on the Braden Scale and Care Dependency Scale. They observed associations between incontinence associated skin damage and fecal incontinence, skin moisture or friction and shear, higher body mass index, and diabetes mellitus.

In addition to health status factors of individuals, analysis of complex conditions such as IAD in populations, requires consideration of potential predictors at multiple levels. In addition to a nursing home resident’s health status, for example, there may be organizational or community resource factors that can influence their healthcare and outcomes. 15 , 16 Some of these factors might include the proportion of individuals of a racial or ethnic group who reside in a nursing home 17 or require incontinence care, nursing staffing levels, 18 , 19 knowledge and attitudes of nursing staff, 20 , 21 deficiencies in the quality of nursing home care, socioeconomic resources of the community in which a nursing home is located, 22 and healthcare access prior to nursing home admission. 23 Limiting analyses to only resident-level characteristics has been referred to by social science researchers as an “individualistic fallacy.” 24 The purpose of this study was to determine the incidence and predictors of IAD in nursing home residents. Potential predictors at the individual, nursing home, and community levels were assessed.

Four large data sets were linked and analyzed. Minimum Dataset (MDS) v.2.0 records provided demographic and comprehensive health assessment data of nursing home residents of a national for-profit chain of nursing homes. Practitioner orders (POs) are legal health records of orders written by physicians or nurse practitioners for residents in the NHs in our data set in electronic form.. Federal regulations require that all prescription and nonprescription drugs and any care interventions (including skin and wound care, mobility and physical therapy, diet, laboratory orders, and procedures administered to a NH resident), and their changes or discontinuations be documented in an order. 25 Practitioner orders in our data set included a practitioner’s reason for a prescriptive order, dose, route, and frequency for medications and other treatments, location for applying topical treatments, and descriptions and locations of some problems. Words and descriptions in POs provided data about interventions and products for care, prevention and treatment as well as the presence of incontinence, and presence/description and location of IAD and/or pressure injury. Data about nursing home staffing and deficiencies in quality of care were in Online Survey, Certification, and Reporting (OSCAR) records. Socioeconomic and sociodemographic characteristics of the Census tract of the community in which the nursing homes were located were obtained from the 2000 U.S. Census. The Minnesota Population Center at the University of Minnesota, Minneapolis, MN provided the Census tracts of the nursing homes. Data from MDS, PO, and OSCAR records were all from the same years, 2000–2002. Data were de-identified and the study was exempt from review according to the University of Minnesota Institutional Review Board.

Design, Cohort, and Outcomes

This study used a cohort design to determine the incidence of IAD in older (aged ≥65 years) nursing home residents. The cohort comprised nursing home admissions who were continent or usually continent of urine and/or feces on the first full MDS record, were free of IAD per POs at admission, and developed incontinence (urinary, fecal, or dual) during their nursing home stay per MDS or PO records.

Subsequent development of incontinence and IAD was determined by documentation in the POs. Practitioner orders were searched for words or sets of words (including misspellings and different order of words) describing incontinence, IAD, and other outcomes we studied such as pressure injuries and IAD prevention using computerized algorithms developed by a consulting company specializing in this work (Edgeworks Technology, Inc., Chicago, IL). We included ICD9 codes for urinary incontinence (788.30–9, 625.6, 788.91) in the electronic algorithms. Search words were identified from a review of the literature and expertise of the research team and three clinical experts.

For IAD and perineal pressure injury descriptions, both an appropriate description of the skin and body area needed to be present. A list of body areas and skin descriptions for IAD and perineal pressure injuries, which was developed in an earlier study 26 was also reviewed and updated by the clinical experts and investigators. Examples of words describing IAD included redness, erythema, pink areas, skin breakdown, dermatitis, perineal skin damage, and fungal rash. Examples of words searched for IAD locations are perianal, around the anus, perirectal, peri area, perineum, buttock, groin, gluteal, inner thigh. Examples of words describing pressures injuries included decubitus, bedsore, ulcer, report of a pressure injury stage, and wound. Examples of words that were excluded from defining both IAD and perineal pressure injuries included tape tear/burn/blister, burns, cuts/lacerations, and herpes lesions. Within the context of this study, the term “perineal” was used broadly to include pressure injuries in the sacral, ischial, coccygeal, trochanter, hip, buttocks, and lower back areas that would be exposed to leaked urine or feces. Examples of alternate words for incontinence included urine/urinary leakage, dribbling, leaked stool, and bowel leakage. Stress, urge, dual, double, fecal, anal, and bowel incontinence were some other descriptors.

Initial reviews of POs by research assistants which identified other words were added to the algorithms in an iterative manner. Records that contained relevant words were then reviewed and coded for the presence of outcomes of interest by research assistants. In reviewing POs, research assistants consulted with investigators as needed when distinguishing IAD from a pressure injury or determining that IAD and a pressure injury were both present. In the rare instance when IAD could not be discerned from pressure injury based on the description in a PO, the problem was coded as a pressure injury.

Incontinence incidence was also measured using MDS records as described in more detail elsewhere. 27 Briefly, MDS records were searched forward from admission for the first report of urinary or fecal incontinence (indicated by MDS item H1a or H1b >1 meaning incontinent, frequently incontinent, or occasionally incontinent) or until a resident’s available data ended. The earliest date reporting occurrence of incontinence was used. Incontinence associated dermatitis needed to occur on or after the onset of incontinence. Because the MDS record of incontinence is based on a resident’s status during 14 days before the report, the definition of IAD included data from POs during this 14-day period in residents whose incontinence was documented by the MDS.

Inter-rater reliability of the 10 coders was periodically and randomly assessed. The lowest percent agreement of the coders was 85% for 32,000 POs. Agreement increased to 95% to 100% for 102,119 POs with additional training, use of a written log of coding decisions, and periodic discussions among the coders. All differences in coding among the research assistants were reconciled.

Predictor Variables

Potential predictors of IAD were identified based on literature review and discussion among the investigators and clinical experts. Predictors were operationally defined using individual items available in the data sets or constructing established scales using individual items with good psychometric properties as multiple items on a record may relate to the same concept. At the individual resident level, potential predictors included residents’ admission demographic characteristics, functional deficits, physical health, and cognitive and emotional status. The characteristics of the cohort were also described with these variables ( Tables 1 and ​ and2). 2 ). Examples of scales used for individual level predictors were those for deficits in activities of daily living (ADLs) which include assessments of bed mobility, transferring, locomotion, dressing, eating, toilet use, and personal hygiene; cognitive deficits (MDS-COGS) 28 which include assessments of short and long-term memory, memory and recall ability, cognitive skills for daily decision making, making oneself understood, and dressing ability. 29 In addition, the Charlson Index 30 was used to identify the number of comorbid conditions such as diabetes mellitus or atherosclerotic heart disease. Where no scale existed but was needed, composite variables of multiple items were developed. Items from the MDS used in defining the variables or composite scales are listed in Tables 1 and ​ and2. 2 . Creation of composite measures of individual resident level predictors used procedures we previously established and published. 26 Scales at all levels (individual, nursing home, and community) used in this analysis were coded so that a higher score indicated a worse condition.

Demographic, Functional, and Physical Characteristics of Older, Incontinent Nursing Home Admissions Followed for Developing Incontinent Associated Dermatitis

Cognitive and Emotional Characteristics of Older Incontinent Nursing Home Residents

The predictor, receipt of IAD prevention, was defined using POs as described above with the addition of searching for words found in skin care product catalogs. Descriptions of IAD prevention in the POs may have included use of specific products (e.g., apply petrolatum after every incontinence episode to prevent skin breakdown) or a general protocol (e.g., start protocol to prevent skin breakdown) and needed to occur before IAD.

Predictors at the nursing home level included the percentage of residents receiving Medicaid and composite variables developed for deficiencies in quality of care, types of nursing home staff, and percentages of admissions by gender, race, and ethnicity. Procedures are described in more detail elsewhere. 31 The sociodemographic and socioeconomic characteristics of the communities in the US Census tracts of the nursing homes included the proportions or percentages of the community that was female or male under < 65 years or ≥65 years old, part of a racial or ethnic minority group, completed 1–8 years or 9–16 years of education, and resided in an urban or rural area. Others community level variables included the proportion of the tract that was working class or below poverty level, as well as the median home value in the tract.

Data Analysis

Data were summarized and examined using descriptive statistics (e.g., frequencies and percentages and means and standard deviations) as appropriate. Relevant predictors were screened for inclusion in the statistical model using bivariate associations with the outcome of IAD; those associated at p < .05 were considered candidates for inclusion. Collinearity between predictor variables was also examined. If a resident level and nursing home or community level variable were highly correlated, the resident level variable was included as it was more specific.

The incidence of IAD was modeled using hierarchical logistic regression with nursing home-specific random effects that accounted for clustering of residents within nursing homes. The potential predictors included in the IAD incidence model were limitations in activities of daily living (ADLs), 29 race and ethnicity, cognitive deficits per the MDS Cognition scale (MDS-COGS), 28 oxygenation and perfusion problems, presence of a perineal pressure injury at admission, prevention of IAD program, and the percentage of nursing home residents receiving Medicaid. Data management and statistical analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA) and SPSS v. 22 (SPSS, Chicago, IL) or R software. Significance level was set at p < .05.

The cohort comprised 10,713 residents in 448 nursing homes located in 28 states and all 9 United States Census divisions. Their mean age was 82 years, most were female, and approximately half had a high school education or greater ( Table 1 ). All race and ethnic groups were represented with the largest group being Whites followed by Blacks. The level of ADL deficits and the mortality (CHESS 32 ) index of the residents were moderate while their comorbidity (Charlson Index 30 ) score was relatively low. Approximately one-third of residents had perfusion or vision problems, and about a quarter of them had an acute condition. Oxygenation and bowel problems were present in about 20% of residents. Nearly half of the cohort (47%) had poor nutrition, assessed by weight loss 5% or more in the last 30 d or 10% or more in the last 180 days and leaves 25% or more of food uneaten at most meals (per MDS items noted in Table 1 ). Use of tube feeding and restraints was low (approximately 1%).

Regarding the residents’ cognitive function, the average level of cognitive deficits was moderate, and delirium affected nearly a quarter of the residents ( Table 2 ). The levels of communication difficulties and discomfort behaviors were fairly low. Depressive symptoms were present in approximately one-third of residents.

Characteristics of the nursing homes and community

Staffing by licensed nurses was 1.1 ± 0.5 hours/resident/day (mean ± SD) while staffing by CNAs was approximately twice that at 2.2 ± 2.1 CNA hours/resident/day. The average number of deficiencies of interest for which a nursing home was cited was 3.8 ± 2.4. Nearly three quarters 73.8% ± 15.9% of the residents received Medicaid. The nursing homes were located in communities with diverse racial and ethnic populations, although the percentage of minorities in some communities was small ( Table 3 ). A majority of nursing homes tended to be in working class communities. Approximately half (49.3%) of the nursing homes were in tracts that were largely (≥ 75%) urban.

Characteristics of Nursing Homes and Their Surroundings

Incidence and Predictors of IAD

The incidence of IAD among nursing home residents with new incontinence after nursing home admission was 5.5%. Of the residents who developed IAD, 89% did so within 14 days of the occurrence of incontinence. Significant predictors of developing IAD were having greater limitations in ADLs, more perfusion problems, a perineal pressure injury at admission, fewer cognitive deficits and not receiving preventive interventions for IAD ( Table 4 ). Receiving preventive interventions for IAD was the strongest protective factor against its occurrences. The odds of developing IAD among residents receiving IAD prevention were about half (46%) the odds among those not receiving prevention (OR = 0.46; 95% CI: 0.33–0.65). Residents who had a perineal pressure injury present upon nursing home admission were much more likely to develop IAD than those without a pressure injury (OR=2.04; 95% CI: 1.32–3.15).

Predictors of Incontinence Associated Dermatitis in Older Nursing Home Residents

This study is among the few 5 – 9 to examine the incidence of IAD in a national nursing home population and the first to our knowledge to report predictors of new onset IAD in those residents. Incontinence associated dermatitis occurred in nearly 6% of nursing home residents who developed incontinence after NH admission. This rate is similar to that reported in long-term acute care 7 and greater than that in nursing homes with a skin damage prevention program in place for all residents. 8 , 9

Gray and colleagues 33 underscored the need for and importance of prevention of IAD, and our results provide evidence supporting its value. Preventive measures for IAD was a strong protective factor against its development. Others have reported a strong association of incontinence with inflammatory type damage of the skin barrier as well as pressure related injury in the perineal area in cross-sectional studies of NH residents 34 and hospitalized patients. 35 Since IAD occurred within two weeks of new incontinence in the vast majority of residents (89%), instituting prevention of skin damage at the start of incontinence seems critical.

These findings are also among the first to show that pre-existing skin damage from a perineal pressure injury is a risk factor for IAD suggesting that damaged skin is more vulnerable to subsequent IAD. Other predictors for IAD suggest that functional/ADL deficits of nursing home residents may interfere with their ability to properly cleanse soiled skin. Activity of daily living deficits in residents’ mobility or toileting abilities may result in more frequent incontinence and exposure of skin to local inflammatory irritants. Perfusion problems in this study include dehydration and edema; perfusion/hydration problems are components of conceptual models of factors associated with IAD. 2 , 36 Epidermal cells are 70% water and dehydration impairs their normal life cycle and function. 37 Edema increases interstitial fluid disrupting the normal matrix of epidermal cells resulting in increased permeability of the skin barrier and inflammation. 38 , 39 Residents with lesser cognitive deficits may be more likely to develop IAD because they may need more assistance than assumed in managing perineal cleansing after incontinence. Further studies explaining the mechanisms underlying IAD predictors are needed. The results of this study can be used to guide the focus of such investigations. Our results support and extend associations of similar variables with IAD observed in previous cross-sectional studies, 13 , 14 and proposed in a conceptual model. 2

Preventing and managing IAD is an integral part of WOC nursing practice and findings of this study affirm the value of the WOC nurse in the nursing home setting. The predictors of IAD identified in this study assist WOC nurses in developing the focus of education for nursing home staff and in planning their consultation services. For example, WOC nurses can use the research-based evidence of this study to assist nursing homes in identifying residents at risk for IAD to reduce their morbidity and improve their health outcomes. Findings also support WOC nurse efforts advocating for IAD prevention. They emphasize the need for timely healing of perineal pressure injuries which is facilitated by the expertise of the WOC nurse.

Strengths and Limitations

Our procedures for identifying IAD in the POs may have underreport IAD as not all cases may have been identified. The time frame of our data is a limitation as newer skin care products may have increased effectiveness of IAD prevention if used by nursing homes. All relevant predictors of incontinence prevention may not be known, available in our datasets, or perhaps possible to include in our models. For example, data about the knowledge and attitudes of nursing staff were not available in our datasets, and we were unable to differentiate the type of incontinence in our model because some residents did not have these data specified in their POs. Logistic regression analysis does not account for time to development of IAD and admissions in our study had differing times to the first occurrence of IAD. However, we observed that IAD occurred within 14 days of the onset of incontinence in 89% of the residents who developed incontinence associated skin damage. The generalizability of results is limited to residents in for-profit nursing homes. We note that 69% of all US nursing homes are for-profit; furthermore, our admission cohort has similar characteristics as residents in all US nursing homes. 40

There are also strengths of this study. As IAD is not a standardized resident assessment item on the MDS, the POs provided a unique resource for investigating this understudied problem. Incontinence associated dermatitis has been shown to recur after healing in individuals whose incontinence persists. 1 We analyzed data of residents who newly developed incontinence after nursing home admission to reduce variability. Similarly, we studied only nursing home residents because risk factors for IAD in community-living individuals and critically-ill patients may be different than those in nursing home residents. 41 Another strength is that the four datasets used are from the same time period, using the most recent Census data available at the beginning of our study. We considered/screened potential predictors variables at multiple levels, and our model adjusted for the percentage of residents on Medicaid, a nursing home level factor indicating a financial resource of nursing homes. Use of multi-item scales and composite predictors, as done in this study, has been shown to improve predictive ability, including that of MDS data. 42


This is the first study to report incidence of IAD and predictors in NH residents who developed incontinence during their stay. Residents with functional, physical, and cognitive deficits are at risk for developing IAD. Lack of prevention of IAD and presence of pressure injury in the perineal area were among the strongest predictors of IAD. These findings guide WOC nurses in educating nursing home staff about the risk factors for IAD and in reducing its incidence, which will improve NH resident outcomes.


This study was funded by National Institute of Nursing Research, NIH, 1R01NR010731 and the Minnesota Supercomputing Institute, University of Minnesota, Minneapolis, MN.

The authors declare no conflicts of interest related to the content of the manuscript.

  • Reference Manager
  • Simple TEXT file

People also looked at

Original research article, determinants of urinary incontinence and subtypes among the elderly in nursing homes.

nursing research articles on incontinence

  • 1 Clinical Nursing Teaching and Research Section, The Second Xiangya Hospital, Central South University, Changsha, China
  • 2 Geriatric Department, The Second Xiangya Hospital, Central South University, Changsha, China
  • 3 Xiangya School of Public Health, Central South University, Changsha, China

Urinary incontinence (UI) is a common problem among older adults. This study investigated the prevalence of UI in nursing home residents aged ≥75 years in China and examined potential risk factors associated with UI and its subtypes. Data were collected during face-to-face interviews using a general questionnaire, the International Consultation Incontinence Questionnaire Short-Form, and the Barthel Index. A total of 551 participants aged ≥75 years residing in Changsha city were enrolled from June to December 2018. The UI prevalence rate among nursing home residents aged ≥75 years was 24.3%. The most frequent subtype was mixed (M) UI (38.1%), followed by urge (U) UI (35.1%), stress (S) UI (11.9%), and other types (14.9%). In terms of severity, 57.5% had moderate UI, while 35.1% had mild and 7.5% had severe UI. Constipation, immobility, wheelchair use, cardiovascular disease (CVD), and pelvic or spinal surgery were significant risk factors for UI. Participants with a history of surgery had higher risks of SUI (odds ratio [OR] = 4.87, 95% confidence interval [CI]: 1.55–15.30) and UUI (OR = 1.97, 95% CI: 1.05–3.71), those who were immobile or used a wheelchair had higher rates of MUI (OR = 11.07, 95% CI: 4.19–29.28; OR = 3.36, 95% CI: 1.16–9.78) and other UI types (OR = 7.89, 95% CI: 1.99–31.30; OR = 14.90, 95% CI: 4.88–45.50), those with CVD had a higher rate of UUI (OR = 2.25, 95% CI: 1.17–4.34), and those with diabetes had a higher risk of UUI (OR = 2.250, 95% CI: 1.14–4.44). Use of oral antithrombotic agents increased UUI risk (OR = 4.98, 95% CI: 2.10–11.85) whereas sedative hypnotic drug use was associated with a higher risk of MUI (OR = 3.62, 95% CI: 1.25–10.45). Each UI subtype has distinct risk factors, and elderly residents of nursing homes with a history of CVD and pelvic or spinal surgery who experience constipation should be closely monitored. Reducing time spent in bed and engaging in active rehabilitation including walking and muscle strengthening may aid in UI prevention and treatment.


Urinary incontinence (UI) is defined by International Continence Society (ICS) as complaint of involuntary loss of urine ( 1 ). It's common in elderly patients and usually play a major role in independent person in the community or dependent person in the nursing home ( 2 , 3 ). UI affects nearly 40% of women aged 80 years and older, 10–35% of older men, and up to 80% of long-term care residents and can severely impair an individual's quality of life (QoL) ( 4 , 5 ) because of the associated hygiene and social problems ( 6 ). Therefore, healthcare providers need to demonstrate sensitivity in evaluating and discussing UI, particularly with older adults.

There are three major subtypes of UI—urge (U), stress (S), and mixed (M)—which have different risk factors and etiologies ( 7 ). UUI and SUI are the most common subtypes in older persons, while MUI is the combination of both types ( 8 ). UI is a risk factor for mortality in the elderly ( 9 – 11 ) and is closely related to declines in cognitive function and performance of activities of daily living (ADL) as well as age, obesity, diabetes, loss of independence, depression and anxiety levels, and agitation ( 8 , 11 – 14 ). It is important to clarify the association between UI and cardiovascular risk factors through screening ( 14 ) to prevent the development of cardiovascular disease (CVD). However, there is little known about the prevalence of UI in the elderly population of China.

There are many factors that affect urinary incontinence in older adults, such as age, frailty, depression, neurologic conditions, cognitive impairment, mobility impairment, lower urinary tract symptoms ( 15 – 18 ), race, education, hypertension, smoking, diabetes, increased parity, higher body mass index, and oral hormone therapy, and radical prostatectomy for prostate cancer in men ( 15 , 16 , 19 – 21 ). However, in the very old population, whether there are more special factors of UI or not for this high-risk population, these are worthy of our consideration.

The aims of this study was to determine the prevalence of UI in the very old population (aged ≥ 75 years old) in nursing homes, examine potential risk factors associated with UI and its subtypes, which may provide evidence for further development of UI strategies for this high-risk populations.

Materials and Methods

Study design.

This nursing institution-based cross-sectional study was carried out with face-to-face surveys conducted between June and December 2018 among older adults residing in Changsha city, the capital of Hunan province, China. Changsha is located in the east of Hunan, with an area of 11,819.5 km 2 . Changsha comprises Furong, Tianxin, Yuelu, Kaifu, Yuhua, and Wangcheng districts; Changsha and Ningxiang counties; and Liuyang city. The study involved 20 of the 83 nursing homes in Changsha, with over 551 nursing home residents enrolled.

Study Population

Older adults ≥75 years old who had resided for a minimum of 1 year in the study area with normal cognition and communication ability were eligible to participate. Mentally unstable nursing home residents with life-threatening diseases were excluded. The study was approved by the Medical Ethics Committee of Central South University. Written informed consent to participate in the study was provided by the participants or their legal guardian/next of kin.

The sample size was calculated using the formula n = μα 2 π (1–π)/δ 2 , where α is 0.05; π is the prevalence rate of UI, which was taken to be 25.0% from a previous study ( 22 ); and δ is 0.15 π. Based on this calculation, the minimum sample size for this study was determined as 512.

Survey Instrument

A questionnaire for collecting data on general characteristics was used in the face-to-face interview. The data included age, nationality, marital status, occupation, source of income, education level, and medical history, among other characteristics. Height and weight were measured with the same ruler and electronic scales for each participant.

UI was assessed using the International Consultation Incontinence Questionnaire—Short Form (ICIQ-SF), which continues to be the most internationally used questionnaire and has been translated into over 60 languages ( 23 ). It comprises three scored items and an unscored self-diagnosis item to determine the prevalence, frequency, and severity of urinary leakage and its impact on QoL ( 24 ). The sum of scores for the three items ranging from 0 to 21, and higher scores indicating increased UI severity and greater impact on QoL. The scale has demonstrated high internal reliability in British patients at a urology clinic and in a community-based study (Cronbach's α = 0.95) ( 24 , 25 ). Mild UI was defined as < 7 points; moderate UI as 7–14 points; and severe UI as >14 points.

ADL performance was significantly associated with UI ( 12 ), the Barthel Index (BI) was used to assess each individual's ADL performance (Cronbach's α = 0.93) ( 12 ). This 100-point clinical rating index includes 10 items related to self-care ability (i.e., bowels, bladder, grooming, toilet use, feeding, dressing, and bathing) and mobility (i.e., transfer, mobility, and stairs), with a higher score indicating a lower level of physical dependence. The Barthel index scores are classified as follows: 0–20 points: total dependency; 21–60 points: high-level dependency; 61–90 points: mid-level dependency; 91–99 points: low-level dependency; 100 points: total independence ( 26 ).

Data Analysis

EpiData v3.1 ( https://www.epidata.dk/index.htm ) and SPSS v25.0 (IBM Corp., Armonk, NY, USA) were used for data management and analysis, respectively. Numerical variables are expressed as the mean ± standard deviation (SD) and categorical variables as frequency and percentage. Differences in frequency distributions between groups were assessed with Pearson χ 2 tests, and determinants of UI and its subtypes were assessed using binary logistic regression (LR) models. For all tests, 2-tailed p < 0.05 were considered statistically significant.

General Characteristics of Older Adults in Nursing Homes

Of the 551 study participants, 67.0% were female; 55.7% were 70–79 years old, 44.3% were >80 years old, and the mean age (±SD) was 84.16 (±4.84) years. In terms of education level, 37.2% of participants had completed middle or high school and almost 40% had >40 years of work experience. Most participants had been married; 30.5% were still married and 67.9% were widowed. In most cases, the source of income was retirement pension (89.5%), and only 14.5% of participants had a monthly income <2,000 yuan. In terms of functional status, 62.1% of participants were fully independent in ADL as measured by BI; 6.4% were impaired, and 4.5% were disabled. Based on body mass index (BMI), 22.9% of participants were overweight and 5.1% were obese (BMI ≥ 28 kg/m 2 ) ( Table 1 ).


Table 1 . General characteristics of older adults in nursing homes.

UI Prevalence in the Geriatric Population of Nursing Homes

We found 134 UI in all 551 participants, with a UI prevalence rate of 24.3%. Of which MUI accounted to 38.1%; UUI 35.1%; SUI 11.9%; and other types 14.9%; 57.5% of UI was moderate UI, 35.1 % was mild UI, and 7.5% had severe UI ( Table 2 ).


Table 2 . The constituent ratio of different UI among 134 UI patients.

General Characteristics of UI in the Geriatric Population

There was no difference in UI prevalence between males (22.5%) and females (25.2%) or between obese (23.5%) and non-obese (39.3%) participants. Anxiety and depression were associated with higher rates of UI (32.3%, χ 2 = 7.39, p = 0.007) and other types of UI (6.5%, χ 2 = 4.91, p = 0.027). In terms of functional status, immobile participants had a higher frequency of UI and MUI (75.0%, 45.0%), whereas those who could walk independently had a low rate of UI (19.9%) (χ 2 = 51.82, p < 0.001). Mobility was a strong predictor of MUI and other UI types (χ 2 = 36.68, p < 0.001, χ 2 = 41.82, p < 0.001). Participants with a history of hypertension or urinary tract infection (UTI) had a higher rates of UI than those without this medical history (hypertension: 27.8% vs. 19.4%, χ 2 = 5.11, p = 0.024; UTI: 53.8% vs. 23.6%, χ 2 = 6.31, p = 0.012). Participants with constipation had higher rates of UI and MUI than those without constipation (UI: 32.7% vs. 20.7%, χ 2 = 6.31, p = 0.012; MUI: 13.9% vs. 7.3%, χ 2 = 6.15, p = 0.013). Participants with a history of CVD had a higher rate of UI and UUI (UI: 32.9% vs. 21.3%, χ 2 = 7.67, p = 0.006; UUI: 16.1% vs. 5.9%, χ 2 = 14.12, p < 0.001) and those with a history of surgery had a higher rate of UI (30.6% vs. 20.3%, χ 2 = 7.51, p = 0.006), SUI (5.6% vs. 1.2%, χ 2 = 8.86, p = 0.003), and UUI (11.6%, χ 2 = 4.22, p = 0.040) than those without these in their medical history. Finally, participants who were taking oral antilipidemic and antithrombotic medications had higher rates of UI than those who were not taking these drugs (antilipidemics: 44.0%, χ 2 = 5.51, p = 0.019; antithrombotics: 44.1%, χ 2 = 5.60, p = 0.018) ( Table 3 ).


Table 3 . Prevalence rate of UI in different character of nursing homes ( n = 551).

Factors Associated With UI and Its Subtypes

Binary LR was carried out to evaluate the association between UI, SUI, and other types of UI (dependent variable, dichotomized into UI vs. no UI, SUI vs. no SUI, and other UI vs. no other UI) and general characteristics of UI (covariates: anxiety/depression, constipation, mobility, CVD, hypertension, history of surgery, UTI, antilipidemic and antithrombotic medications). UI-related characteristics that were significant on Pearson χ 2 tests were entered into the LR model by backward stepwise regression, with mobility as the last categorical covariate (independent walking; α In = 0.05, α Out = 0.10).

Binary LR was used to assess the independent association between UUI (dependent variable, dichotomized into UUI vs. no UUI) and general characteristics of UI and MUI (covariates: anxiety/depression, constipation, mobility, CVD, hypertension, history of surgery, UTI, antilipidemic and antithrombotic medicines, alcohol consumption, and diabetes). All UI- and UUI-related characteristics that were significant in Pearson χ 2 tests were entered into model by backward stepwise regression, with mobility as the last categorical covariate (independent walk; probability for stepwise entry = 0.05, removal = 0.10). The binary LR model was used to assess the independent association between MUI (dependent variable, dichotomized into MUI vs. no MUI) and general characteristics of MUI (covariates: anxiety/depression, constipation, mobility, CVD, hypertension, surgical history, UTI, antilipidemic and antithrombotic medicines, age, and sedative/hypnotic drugs). All UI- and MUI-related characteristics that were significant in Pearson χ 2 tests were entered into the model by backward stepwise regression, with mobility as the last categorical covariate (independent walking; probability for stepwise entry = 0.05, removal = 0.10). The multivariate-adjusted odds ratios (ORs) and their 95% confidence intervals (CIs) and p -values were calculated. All analyses met the goodness-of-fit criterion as determined with the Hosmer–Lemeshow tests.

The results showed that constipation, immobility, wheelchair use, CVD, and history of surgery were significant risk factors for UI ( Table 3 ). Participants with a history of surgery had a higher risk of SUI (OR = 4.87, 95% CI: 1.55–15.30) and UUI (OR = 1.97, 95% CI: 1.05–3.71), and immobile and wheelchair-assisted older adults had a higher frequency of MUI (OR = 11.07, 95% CI: 4.19–29.28; OR = 3.36, 95% CI: 1.16–9.78) and other types of UI (OR = 7.89, 95% CI: 1.99–31.30; OR = 14.90, 95% CI: 4.88–45.50). Compared to participants with no history of CVD, those with CVD history reported a higher frequency of UUI (OR = 2.25, 95% CI: 1.17–4.34). Participants with diabetes were more likely to experience UUI than those without diabetes (OR = 2.250, 95% CI: 1.14–4.44). Use of oral antithrombotic drugs was associated with a higher risk for UUI (OR = 4.98, 95% CI: 2.10–11.85), and a history of sedative hypnotic drug use was associated with a higher risk of MUI (OR = 3.62, 95% CI: 1.25–10.45) ( Table 4 ).


Table 4 . Multivariate logistic regression model with UI and subtypes in the geriatric population of nursing homes.

In this study, the UI prevalence of nursing home residents aged ≥75 years was 24.3%, which is lower than that reported in other studies of individuals aged ≥65 years ( 4 , 22 ). General good health, consciousness, and good cognitive ability may explain the lower rate in our cohort. Our results showed that CVD was a risk factor for UI and UUI, which was in line with previous findings that UI had a high prevalence among heart failure patients ( 27 , 28 ). As bladder function is affected by many cardiovascular risk factors, UI is a possible consequence of metabolic syndrome ( 14 ). Water–sodium retention and impaired bladder function are associated with CVD, while diuretics used in CVD treatment may lead to nocturia and increase the occurrence of UI ( 29 ). Despite patients' perception that diuretics are unpleasant and make it difficult for them to leave their home, patients in one study were generally compliant with their medication regime; nonetheless, nearly half experienced urine leakage, and most found urgency and incontinence bothersome ( 28 ). The assessment and management of UI or UUI in patients with CVD warrant further exploration.

The relationship between mobility in ADL and UI and its subtypes was evaluated based on immobility, wheelchair dependence, and assisted and independent walking. Immobility and wheelchair dependence were found to be risk factors for UI, MUI, and other UI types in the study participants, which is in accordance with earlier observations ( 29 , 30 ). The use of walking aids and activity training may reduce or prevent the occurrence of UI in the elderly ( 30 , 31 ); thus, promoting walking ability may be effective in preventing UI in nursing home residents.

Constipation has been shown to increase the risk of UI in the elderly ( 29 , 32 ), which is consistent with our findings. The anatomy and angle of the urethra may be altered with chronic constipation, leading to problems such as overactive bladder (OAB), urinary retention, and UI ( 32 , 33 ); conversely, treatment of constipation may prevent UI.

In contrast to a previous study ( 34 ), we found no relationship between diabetes and UI in older adults; however, diabetes can lead to glycosuria and has been shown to be associated with UUI risk in multiple models. In one study, diabetes was associated with increased urination frequency and urine volume and thereby exacerbated UI and OAB by osmotic diuresis ( 35 ). Thus, stabilizing blood glucose level is a potential strategy for preventing UUI.

A history of pelvic or spinal surgery was an independent risk factor for UI and two subtypes (SUI and UUI). Damage to nerves or connective tissues near or in the bladder can occur during surgery ( 29 ), and radical pelvic dissection can result in direct and indirect injury to the pelvic plexuses, resulting in SUI and UUI ( 36 ). UI prevalence was reported to be higher among patients who had undergone spinal surgery ( 37 ).

Sedative hypnotic and antithrombotic drug use was identified as a determinant of MUI risk in our cohort. Insomnia is among the most common sleep disorders in the geriatric population ( 38 ), and the use of nighttime sedatives in this group may lead to nocturnal enuresis by inducing a deep sleep from which an individual fails to awaken in order to void ( 36 ). Antithrombotic drugs that inhibit platelet or coagulation factors are commonly prescribed drugs for preventing and treating cardiovascular disorder ( 39 ). Repeated low doses of aspirin can block arachidonic acid receptors and inhibit thromboxane A2 production by acetylating a serine residue near the narrow catalytic site of the cyclooxygenase (COX)-1 channel ( 40 ); and high doses of aspirin inhibit both COX-1 and COX-2, which have anti-inflammatory and analgesic effects ( 41 ). Oral aspirin potentially inhibits COX and decreases prostaglandin (PGE)2, a regulator of inflammation and metabolism ( 42 ) that acts through the G protein-coupled PGE2 receptors (EP) 1, EP2, EP3, and EP4. EP1 and EP3 activation in the detrusor muscle of the bladder induces muscle contraction, whereas EP4 activation causes muscle relaxation ( 43 ). Aspirin may target EP1, EP2, EP3, and EP4 to reduce bladder sphincter contraction and detrusor relaxation, leading to involuntary urine leakage, although the precise underlying mechanism remains to be determined.

Obesity was not a significant risk factor for UI in our cohort in the binary LR model, which is inconsistent with previous findings ( 14 , 44 ). Obesity was previously reported as a risk factor for UI, although the observed trends are contradictory, with decreased SUI and increased MUI rates found to be associated with higher BMI ( 44 ). Older males with a reduction in strength but not increases in body or fat mass were linked to an increased frequency of UI ( 45 ). Changes in body composition including an increase and redistribution of fat mass occur in old age, and current BMI classification may not accurately reflect the associated physical risks in the elderly, who may require age-specific BMI cut-off points ( 46 ). Thus, BMI in itself should not be considered as an independent predictor of UI in the geriatric population but should be considered in the context of fat mass, muscle strength, or other indicators.

In this study, we found that distinct factors contribute to the risk of different UI subtypes. Our results indicate that care providers in nursing homes should pay particular attention to residents with a history of CVD and pelvic or spinal surgery who are at risk of UI and may benefit from the treatment of constipation, less time spent in bed, and active training in walking and muscle strengthening. Further study is needed into the relationship between the use of antithrombotic drugs and UI, and age-specific BMI cut-off points in the elderly population must be established to determine how these factors influence UI risk.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

The studies involving human participants were reviewed and approved by Medical Ethics Committee of Zhejiang Hospital. The patients/participants provided their written informed consent to participate in this study.

Author Contributions

HTan and HTai designed the study. SL and HW performed the experiments. HTai wrote the manuscript and analyzed the data. All authors contributed to the article and approved the submitted version.

This work was supported by National key research and development plan named Research on the construction and evaluation of a comprehensive demonstration base for prevention and control of elderly syndrome (2020YFC2008606) and Nursing Research Fund of the Second Xiangya Hospital in 2018 (2018-YHL-33).

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

1. D'Ancona C, Haylen B, Oelke M, Abranches-Monteiro L, Arnold E, Goldman H, et al. The International Continence Society (ICS) report on the terminology for adult male lower urinary tract and pelvic floor symptoms and dysfunction. Neurourol Urodyn. (2019) 38:433–77. doi: 10.1002/nau.23897

PubMed Abstract | CrossRef Full Text | Google Scholar

2. Chutka DS, Fleming KC, Evans MP, Evans JM, Andrews KL. Urinary incontinence in the elderly population. Mayo Clin Proc. (1996) 71:93–101. doi: 10.4065/71.1.93

3. Acharya S, Ghimire S, Jeffers EM, Shrestha N. Health care utilization and health care expenditure of Nepali older adults. Front Public Health. (2019) 7:24. doi: 10.3389/fpubh.2019.00024

4. Vaughan CP, Markland AD, Smith PP, Burgio KL, Kuchel GA. Report and research agenda of the American geriatrics society and national institute on aging bedside-to-bench conference on urinary incontinence in older adults: a translational research agenda for a complex geriatric syndrome. J Am Geriatr Soc. (2018) 66:773–82. doi: 10.1111/jgs.15157

5. Troko J, Bach F, Toozs-Hobson P. Predicting urinary incontinence in women in later life: A systematic review. Maturitas. (2016) 94:110–6. doi: 10.1016/j.maturitas.2016.09.006

6. Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U, et al. Standardisation Sub-committee of the International Continence Society. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn. (2002) 21:167–78. doi: 10.1002/nau.10052

7. Fong E, Nitti VW. Urinary incontinence. Prim Care. (2010) 37:599–612. doi: 10.1016/j.pop.2010.04.008

8. Minassian VA, Drutz HP, Al-Badr A. Urinary incontinence as a worldwide problem. Int J Gynaecol Obstet. (2003) 82:327–38. doi: 10.1016/s0020-7292(03)00220-0

9. Taufik Ramli MH, Lodz NA, Abdul Aziz FA, Maw Pin T, Alias N, Abdul Mutalip MH, et al. Relationship between urinary incontinence and quality of life of older persons in Malaysia. Geriatr Gerontol Int. (2020) 20:38–42. doi: 10.1111/ggi.14028

10. Damián J, Pastor-Barriuso R, García López FJ, de Pedro-Cuesta J. Urinary incontinence and mortality among older adults residing in care homes. J Adv Nurs. (2017) 73:688–99. doi: 10.1111/jan.13170

11. Yoshioka T, Kamitani T, Omae K, Shimizu S, Fukuhara S, Yamamoto Y. Urgency urinary incontinence, loss of independence, and increased mortality in older adults: A cohort study. PLoS ONE. (2021) 16:e0245724. doi: 10.1371/journal.pone.0245724

12. Li HC, Chen KM, Hsu HF. Modelling factors of urinary incontinence in institutional older adults with dementia. J Clin Nurs. (2019) 28:4504–12. doi: 10.1111/jocn.15039

13. Cheng S, Lin D, Hu T, Cao L, Liao H, Mou X, et al. Association of urinary incontinence and depression or anxiety: A meta-analysis. J Int Med Res. (2020) 48:300060520931348. doi: 10.1177/0300060520931348

14. John G. Urinary incontinence and cardiovascular disease: A narrative review. Int Urogynecol J. (2020) 31:857–63. doi: 10.1007/s00192-019-04058-w

15. Vaughan CP, Goode PS, Burgio KL, Markland AD. Urinary incontinence in older adults. Mt Sinai J Med. (2011) 78:558–70. doi: 10.1002/msj.20276

16. Offermans MP, Du Moulin MF, Hamers JP, Dassen T, Halfens RJ. Prevalence of urinary incontinence and associated risk factors in nursing home residents: a systematic review. Neurourol Urodyn. (2009) 28:288–94. doi: 10.1002/nau.20668

17. Skelly J, Flint AJ. Urinary incontinence associated with dementia. J Am Geriatr Soc. (1995) 43:286–94. doi: 10.1111/j.1532-5415.1995.tb07341.x

18. Bogren MA, Hvarfwén E, Fridlund B. Urinary incontinence among a 65-year old Swedish population: medical history and psychosocial consequences. Vard Nord Utveckl Forsk. (1997) 17:21–4. doi: 10.1177/010740839701700304

19. Veronese N, Soysal P, Stubbs B, Marengoni A, Demurtas J, Maggi S, et al. Association between urinary incontinence and frailty: a systematic review and meta-analysis. Eur Geriatr Med. (2018) 9:571–78. doi: 10.1007/s41999-018-0102-y

20. Batmani S, Jalali R, Mohammadi M, Bokaee S. Prevalence and factors related to urinary incontinence in older adults women worldwide: a comprehensive systematic review and meta-analysis of observational studies. BMC Geriatr. (2021) 21:212. doi: 10.1186/s12877-021-02135-8

21. Zhou HH, Shu B, Liu TZ, Wang XH, Yang ZH, Guo YL. Association between parity and the risk for urinary incontinence in women: A meta-analysis of case-control and cohort studies. Medicine. (2018) 97:e11443. doi: 10.1097/MD.0000000000011443

22. Kim MS, Lee SH. Prevalence rate and associated factors of urinary incontinence among nursing home residents. Taehan Kanho Hakhoe Chi. (2008) 38:92–100. doi: 10.4040/jkan.2008.38.1.92

23. Uren AD, Cotterill N, Pardoe M, Abrams P. The International Consultation on Incontinence Questionnaires (ICIQ): An update on status and direction. Neurourol Urodyn. (2020) 39:1889–96. doi: 10.1002/nau.24437

24. Avery K, Donovan J, Peters TJ, Shaw C, Gotoh M, Abrams P. ICIQ: A brief and robust measure for evaluating the symptoms and impact of urinary incontinence. Neurourol Urodyn. (2004) 23:322–30. doi: 10.1002/nau.20041

25. Slavin V, Gamble J, Creedy DK, Fenwick J. Perinatal incontinence: psychometric evaluation of the international consultation on incontinence questionnaire-urinary incontinence short form and wexner scale. Neurourol Urodyn. (2019) 38:2209–23. doi: 10.1002/nau.24121

26. Ozturk GZ, Egici MT, Bukhari MH, Toprak D. Association between body mass index and activities of daily living in homecare patients. Pak J Med Sci. (2017) 33:1479–84. doi: 10.12669/pjms.336.13748

27. Palmer MH, Hardin SR, Behrend C, Collins SK, Madigan CK, Carlson JR. Urinary incontinence and overactive bladder in patients with heart failure. J Urol. (2009) 182:196–202. doi: 10.1016/j.juro.2009.02.115

28. Poole K, Kerlin M, Wynne R. Prevalence and characteristics of urinary incontinence in a cohort of patients with chronic heart failure. Heart Lung. (2017) 46:67–73. doi: 10.1016/j.hrtlng.2017.01.002

29. Bardsley A. An overview of urinary incontinence. Br J Nurs. (2016) 25:S14–21. doi: 10.12968/bjon.2016.25.18.S14

30. Jachan DE, Müller-Werdan U, Lahmann NA. Impaired mobility and urinary incontinence in nursing home residents: A multicenter study. J Wound Ostomy Continence Nurs. (2019) 46:524–9. doi: 10.1097/WON.0000000000000580

31. Sackley CM, Rodriguez NA, van den Berg M, Badger F, Wright C, Besemer J, et al. A phase II exploratory cluster randomized controlled trial of a group mobility training and staff education intervention to promote urinary continence in UK care homes. Clin Rehabil. (2008) 22:714–21. doi: 10.1177/0269215508089058

32. Eshkoor SA, Hamid TA, Shahar S, Mun CY. Factors related to urinary incontinence among the Malaysian elderly. J Nutr Health Aging. (2017) 21:220–6. doi: 10.1007/s12603-016-0779-x

33. Lian WQ, Li FJ, Huang HX, Zheng YQ, Chen LH. Constipation and risk of urinary incontinence in women: A meta-analysis. Int Urogynecol J. (2019) 30:1629–34. doi: 10.1007/s00192-019-03941-w

34. Tamanini JTN, Pallone LV, Sartori MGF, Girão MJBC, Dos Santos JLF, de Oliveira Duarte YA, et al. A populational-based survey on the prevalence, incidence, and risk factors of urinary incontinence in older adults – results from the “SABE STUDY”. Neurourol Urodyn. (2018) 37:466–77. doi: 10.1002/nau.23331

35. Smith DB. Urinary incontinence and diabetes: A review. J Wound Ostomy Continence Nurs. (2006) 33:619–23. doi: 10.1097/00152192-200611000-00005

36. Bristow SE, Hilton P. Assessment and investigations for urinary incontinence. Baillieres Best Pract Res Clin Obstet Gynaecol. (2000) 14:227–49. doi: 10.1053/beog.1999.0073

37. Lieberman EG, Radoslovich S, Marshall LM, Yoo JU. Lower urinary tract symptoms and urinary bother are common in patients undergoing elective cervical spine surgery. Clin Orthop Relat Res. (2019) 477:872–8. doi: 10.1097/CORR.0000000000000666

38. Patel D, Steinberg J, Patel P. Insomnia in the elderly: A review. J Clin Sleep Med. (2018) 14:1017–24. doi: 10.5664/jcsm.7172

39. Mega JL, Simon T. Pharmacology of antithrombotic drugs: An assessment of oral antiplatelet and anticoagulant treatments. Lancet. (2015) 386:281–91. doi: 10.1016/S0140-6736(15)60243-4

40. Loll PJ, Picot D, Garavito RM. The structural basis of aspirin activity inferred from the crystal structure of inactivated prostaglandin H2 synthase. Nat Struct Biol. (1995) 2:637–43. doi: 10.1038/nsb0895-637

41. Born G, Patrono C. Antiplatelet drugs. Br J Pharmacol. (2006) 147:S241–51. doi: 10.1038/sj.bjp.0706401

42. Ratchford SM, Lavin KM, Perkins RK, Jemiolo B, Trappe SW, Trappe TA. Aspirin as a COX inhibitor and anti-inflammatory drug in human skeletal muscle. J Appl Physiol. (2017) 123:1610–6. doi: 10.1152/japplphysiol.01119.2016

43. Hou R, Yu Y, Jiang J. PGE2 receptors in detrusor muscle: Drugging the undruggable for urgency. Biochem Pharmacol. (2021) 184:114363. doi: 10.1016/j.bcp.2020.114363

44. Schreiber Pedersen L, Lose G, Høybye MT, Elsner S, Waldmann A, Rudnicki M. Prevalence of urinary incontinence among women and analysis of potential risk factors in Germany and Denmark. Acta Obstet Gynecol Scand. (2017) 96:939–48. doi: 10.1111/aogs.13149

45. Bauer SR, Grimes B, Suskind AM, Cawthon PM, Cummings S, Huang AJ, et al. Urinary incontinence and nocturia in older men: Associations with body mass, composition and strength in the Health ABC Study. J Urol. (2019) 202:1015–21. doi: 10.1097/JU.0000000000000378

46. Javed AA, Aljied R, Allison DJ, Anderson LN, Ma J, Raina P. Body mass index and all-cause mortality in older adults: A scoping review of observational studies. Obes Rev. (2020) 21:e13035. doi: 10.1111/obr.13035

Keywords: urinary incontinence, geriatric, older adult, elderly population, prevalence, risk factor, nursing home, subtype

Citation: Tai H, Liu S, Wang H and Tan H (2021) Determinants of Urinary Incontinence and Subtypes Among the Elderly in Nursing Homes. Front. Public Health 9:788642. doi: 10.3389/fpubh.2021.788642

Received: 03 October 2021; Accepted: 15 November 2021; Published: 06 December 2021.

Reviewed by:

Copyright © 2021 Tai, Liu,Wang and Tan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Haiqin Wang, wanghaiqin77@csu.edu.cn ; Hongzhuan Tan, tanhz99@qq.com

This article is part of the Research Topic

Epidemiological Characteristics and Related Risk Factors of Older Population Aged over 75 Years

Nursing research and continence care


  • 1 University of Alberta, Edmonton, Alberta, Canada.
  • PMID: 12087792

Despite 50 years of attention to continence care and pivotal nursing research in the field, incontinence continues to challenge health care professionals and consumers alike. The role of nursing research in continence is traced from the early influences of Florence Nightingale to today's clinical practice guidelines and the influence of major nursing research centers.

  • Nursing Research*
  • Urinary Incontinence / nursing*
  • Urinary Incontinence / psychology

Reimagining the nursing workload: Finding time to close the workforce gap

US healthcare organizations continue to grapple with the impacts of the nursing shortage—scaling back of health services, increasing staff burnout and mental-health challenges, and rising labor costs. While several health systems have had some success in rebuilding their nursing workforces   in recent months, estimates still suggest a potential shortage of 200,000 to 450,000 nurses in the United States, with acute-care settings likely to be most affected. 1 Gretchen Berlin, Meredith Lapointe, Mhoire Murphy, and Joanna Wexler, “ Assessing the lingering impact of COVID-19 on the nursing workforce ,” McKinsey, May 11, 2022. Identifying opportunities to close this gap remains a priority in the healthcare industry. This article highlights research conducted by McKinsey in collaboration with the ANA Enterprise on how nurses are actually spending their time during their shifts and how they would ideally distribute their time if given the chance. The research findings underpin insights that can help organizations identify new approaches to address the nursing shortage and create more sustainable and meaningful careers for nurses.

Over the past three years, McKinsey has been reporting on trends within the nursing workforce , collecting longitudinal data on nurses’ self-reported likelihood to leave their jobs and factors driving nurses’ intent to leave. 2 “ Nursing in 2023: How hospitals are confronting shortages ,” McKinsey, May 5, 2023. As of March 2023, 45 percent of inpatient nurses (who make up about 2.0 million of the 4.2 million nurses in the United States 3 Nursing fact sheet, American Association of Colleges of Nursing, updated September 2022. ) reported they are likely to leave their role in the next six months. Among those who reported an intent to leave, the top two reasons cited were not feeling valued by their organization and not having a manageable workload. In fact, nurses have consistently reported increasing workload burden as a main factor behind their intent to leave.

About the research

We conducted a survey of 310 registered nurses across the United States from February 8 to March 22, 2023. Our goal was to understand nurses’ perception of time spent throughout the course of a shift and to identify existing and desired resources to help nurses provide high-quality care. Our sample focused on nurses in roles that predominantly provide direct patient care in the intensive-care unit, step-down, general medical surgical, or emergency department settings. Insights were weighted by length of shift (the minimum shift time included was six hours).

For questions related to intent to leave nursing, all nurses from any care setting (including home care and long-term care facilities) were included. Our survey questions on intent to leave have been kept consistent to collect longitudinal data on nurses’ intent. Our last survey, of 368 frontline direct-care nurses, was conducted in September 2022.

In our new survey, nurses provided a breakdown of the average time spent during a typical shift across 69 activities (see sidebar “About the research”). They also reported their views on the ideal amount of time they would like to spend on these same activities. In looking at ways to redesign care activities, we found the potential to free up to 15 percent of nurses’ time through tech enablement, or automation, and improved delegation of tasks (Exhibit 1). Leveraging delegation and tech enablement could reduce and redistribute activities that nurses report being predominantly responsible for. The subsequent reduction in time savings could improve nursing workload and their ability to manage more complex patients. When we translate the net amount of time freed up to the projected amount of nursing time needed, we estimate the potential to close the workforce gap by up to 300,000 nurses.

Nurses report a desire to spend more time with their patients, coach fellow nurses, and participate in professional-growth activities

In our survey, we explored where nurses wanted to spend more of their time (Exhibit 2). The responses fall into the following three categories.

Direct patient care

Nurses report spending the majority of their shift—54 percent, or about seven hours of a 12-hour shift—providing direct patient care and creating personal connections with patients (direct patient care includes patient education, medication administration, and support of daily-living activities). The survey reveals that nurses wish to spend even more time in these activities.

Spending sufficient time on patient-care activities promotes both nursing satisfaction and quality of patient care. 4 Terry L. Jones, Patti Hamilton, and Nicole Murry, “Unfinished nursing care, missed care, and implicitly rationed care: State of the science review,” International Journal of Nursing Studies , June 2015, Volume 52, Issue 6. Furthermore, rushing care and not having sufficient time to meet patients’ needs can contribute to moral distress and burnout.

Teaching and training for new nurses and peers

Nurses report spending on average about 2 percent of their shift teaching peers and students (excluding shifts when nurses are in a dedicated teaching or “precepting” role), an activity they say they want to spend double the amount of time on. Peer-to-peer teaching is an important component of building workplace cohesiveness, improving patient outcomes, and preparing new generations of nurses. In our survey, nurses report that they often lack the time to engage in coaching new nurses. As a result, important informal teaching, which is critical to build confidence and to support skill development for newer nurses, is often missed.

Involvement in professional-growth activities

Similar to educating other nurses, nurses report wanting to spend more than double the amount of time on growth and development activities (about 7 percent of an ideal shift). These activities include participating in shared governance, reviewing and reading work emails, and completing annual requirements and continuing education hours.

Freeing up nursing time to support organizational initiatives and further professional development may contribute to a nursing staff that is more engaged, feels valued, and has a strong connection to their departments.

Nurses desire to spend less time on documentation, hunting and gathering, and administrative and support tasks

Charting and documentation.

Documentation continues to greatly contribute to nurses’ workloads, making up 15 percent of a nurse’s shift. The most time-consuming documentation tasks are head-to-toe assessments, admissions intakes, and vitals charting, which account for the majority of documenting time (70 percent). Nurses say that ideally, documenting should make up only about 13 percent of their shift. But without realistic and effective alternatives (for example, nursing scribes, device integration, reduction in documentation requirements, and AI to aid with documentation), it is unlikely that nurses’ documentation burden can be fully alleviated.

Hunting and gathering

For nurses, hunting and gathering means searching for individuals, equipment, supplies, medications, or information. Nurses report that they spend about 6 percent of a 12-hour shift on hunting and gathering—tasks they would spend approximately 3 percent of their shift on in an ideal shift.

Activities best delegated to support staff

Nurses report spending nearly 5 percent of their shift on tasks that do not use the fullest extent of their license and training. For example, they say they spend nearly an hour on nutrition and daily-living activities, such as toileting, bathing, and providing meals and water. In an ideal shift, nurses say they would spend about 3 percent of their time on these activities.

Redesigning care models: Adjusting how nurses spend their time

As we consider how to alleviate nursing workforce challenges, one area of intervention could be evaluating how current care models can be redesigned to better align nursing time to what has the most impact on patient care. Performing below-top-of-license or non-value-adding activities can create inefficiencies that lead to higher healthcare costs and nurse dissatisfaction. Rigorously evaluating whether tasks can be improved with technology or delegated to allow nurses to spend time on activities they find more valuable could help to reduce the time pressures felt by nurses. 5 “National guidelines for nursing delegation,” a joint statement by the NCSBN and American Nurses Association, April 1, 2019. In our analysis, we reviewed the activities nurses say they would ideally spend less time on and considered whether delegation and tech enablement of such tasks could free up nurses’ time.

Based on our analysis, we estimate that full or partial delegation of activities to roles including technicians, nursing assistants, patient-care technicians, food services, ancillary services, and other support staff, could reduce net nursing time by 5 to 10 percent during a 12-hour shift (Exhibit 3).

While nurses report wanting to spend more time overall on direct patient care, there are specific tasks that could be delegated both vertically and horizontally to ensure that the work nurses perform is at the top of their license and promotes professional satisfaction. Appropriate delegation requires training support staff and upskilling where appropriate, as well as evaluating systemwide resources that can be used where needed. For example, within direct patient care, nearly an hour could potentially be freed up by delegating tasks such as patient ambulation, drawing labs and starting IVs, transferring patients, and supporting patient procedures.

Full or partial delegation of activities to roles such as technicians and other support staff could reduce net nursing time by 5 to 10 percent during a 12-hour shift.

Tasks that are evaluated for redistribution to other clinical and non-clinical staff can also be considered as part of broader care-model redesign. Upskilling support staff across clinical and nonclinical roles can often result in overall better use of resources already in place across a health system.

Tech enablement

Based on our assessment, we estimate that a net 10 to 20 percent of time spent during a 12-hour shift is spent on activities that could be optimized through tech enablement. Investing in digital approaches that automate tasks (either completely or partially), rather than simply redistributing workload, could potentially free up valuable time for nurses (Exhibit 4).

Examples of tech enablement and delegation in practice

To determine the amount of time that could potentially be freed up over the course of a nurse’s shift, we used estimations based on best-in-class care delivery models from practice, innovative emerging technology from industry, and how easy it would be for health systems to implement the intervention (for example, cost and technological requirements).


  • Robotic automatic-guided vehicles (AGVs) deliver equipment, food, and supplies throughout a hospital. 1 “Robots help nurses get the job done–with smiles and beeps,” Cedars Sinai, November 29, 2021.
  • Robotic pill-picker machines select and deliver medicines throughout a hospital. 2 Jay Kiew, “The digital surgery: Humber River Hospital reinvents itself with AI & robotics,” Change Leadership, June 16, 2018.
  • Virtual nurses monitor patients remotely, working alongside a bedside-care team comprising a bedside RN, bedside licensed vocational nurse, and virtual RN. 3 Giles Bruce, “Trinity Health plans to institute virtual nurses across its 88 hospitals in 26 states,” Becker’s Health IT, January 13, 2023.
  • Ambient intelligence (that is, passive, contactless sensors embedded in a clinical setting to recognize movement or speech) reduces documentation workload and can continuously monitor patients. 4 Albert Haque, Arnold Milstein, and Li Fei-Fei, “Illuminating the dark spaces of healthcare with ambient intelligence,” Nature , September 9, 2020.
  • Centralized training for roles such as transporters that can then be utilized in all areas of the hospital.
  • Upskilling employees and modifying staffing models allow nurses to work in units where they are needed most (for example, non-critical-care nurses in critical-care departments).

For example, nurses spend 3 percent of their shifts on patient turning and repositioning. This task could be optimized through innovative “smart” hospital-bed technology, including bed-exit alarms, advanced therapy for redistributing pressure, integrated scales and measurements, and remote information on patient conditions. Voice-automated devices and smart beds can also equip patients with control and autonomy over their rooms and preferences (for example, shades, television, and lighting) without nurse intervention (see sidebar “Examples of tech enablement and delegation in practice”).

These interventions, however, can be costly and may not be appropriate solutions in every system. Healthcare organizations will need to assess the specific needs of nurses and patients to determine which interventions will have the most impact.

Healthcare organizations could also consider continuously evaluating the digital approaches they have implemented to ensure that the technology itself does not create redundancies or rework, introduce delays, or adversely increase workload. For example, 37 percent of nurses report that they do not have access to vital signs or telemetry machines that are integrated with electronic medical records for automatic documentation. This could explain why nurses say they could spend less time—about 30 percent less—documenting vital signs. Technology like scanners and automated vitals machines have been an effective way to streamline documentation. But nurses still report spending nearly 10 percent of their shift scanning medications into the patient record, documenting vitals and completed patient education, and drafting progress notes.

Nurse time saved through care-model changes and innovations can benefit patients and nurses—and contribute to building sustainable careers in healthcare

The impact of care-model redesign could range from improving workload sustainability to addressing a substantial portion of the projected 200,000 to 450,000 nursing gap. Our analysis finds a potential net time savings of 15 to 30 percent of a 12-hour shift, based on estimating the possible range of time reduced through delegation 6 “ANAs principles for delegation,” American Nurses Association, 2012. or tech enablement. 7 Mari Kangasniemi, Suyen Karki, Noriyo Colley, and Ari Voutilainen, “The use of robots and other automated devices in nurses' work: An integrative review,” International Journal of Nursing Practice , August 2019, Volume 25, Issue 4.

In our conservative estimate, there would be no additional opportunity to alleviate the potential nursing shortage, as health systems would reallocate the saved time to their current nursing staff for activities they say they would spend more time on, including time with patients, teaching peers, and investing in their growth and development (Exhibit 5). However, this reallocation of time could improve the sustainability of nursing careers in acute-care practice.

In our optimistic estimate, after reallocating time back to nurses, health systems could free up a 15 percent net time savings, which could translate to closing the nursing workforce gap by up to 300,000 inpatient nurses. Achieving this may require health systems to invest heavily in technology, change management, and workflow redesign.

Realizing these changes will require bold departures from healthcare organizations’ current state of processes. It will be critical for hospitals to bring both discipline and creativity to redesigning care delivery in order to effectively scale change and see meaningful time savings. Close collaboration beyond nursing is also paramount to ensure alignment across the care team and hospital functions including administration, IT, informatics, facilities, and operations. A comprehensive evaluation of redesign requirements can enable health systems to understand what is limiting care-model change (for example, policies, skill development, education). Investment in education and additional onboarding may be needed to upskill and train staff on expectations as work is shifted across roles. Partnering with tech companies and industry vendors in areas such as electronic-health-record platforms can accelerate innovation and implementation to build off existing tools and reduce implementation risks. Although the idea of change may be daunting, incorporating innovations in healthcare delivery could be a strategy for building a sustainable workload that could attract and retain nursing talent by allowing them to do more of what matters to them most: taking care of patients and one another.

Gretchen Berlin, RN , is a senior partner in McKinsey’s Washington, DC, office; Ani Bilazarian, RN , is a consultant in the New York office; Joyce Chang, RN , is an associate partner in the Bay Area office; and Stephanie Hammer, RN , is a consultant in the Denver office.

The authors wish to thank Katie Boston-Leary, RN, and the ANA Enterprise for their contributions to this article. The authors also wish to acknowledge and thank the entire healthcare workforce, including all of those on the front line.

Explore a career with us

Related articles.

Two female surgeons confer before seeing patients in the hospital

Nursing in 2023: How hospitals are confronting shortages

Surgeon tying surgical mask

Surveyed nurses consider leaving direct patient care at elevated rates

Smiling female nurse.

Nurses and the Great Attrition

This paper is in the following e-collection/theme issue:

Published on 18.12.2023 in Vol 25 (2023)

Existing Barriers Faced by and Future Design Recommendations for Direct-to-Consumer Health Care Artificial Intelligence Apps: Scoping Review

Authors of this article:

Author Orcid Image

  • Xin He, MA   ; 
  • Xi Zheng, MA   ; 
  • Huiyuan Ding, BA  

School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan, China

Corresponding Author:

School of Mechanical Science and Engineering

Huazhong University of Science and Technology

Luoyu Road 1037

Hongshan District

Wuhan, 430074

Phone: 86 18707149470

Email: [email protected]

Background: Direct-to-consumer (DTC) health care artificial intelligence (AI) apps hold the potential to bridge the spatial and temporal disparities in health care resources, but they also come with individual and societal risks due to AI errors. Furthermore, the manner in which consumers interact directly with health care AI is reshaping traditional physician-patient relationships. However, the academic community lacks a systematic comprehension of the research overview for such apps.

Objective: This paper systematically delineated and analyzed the characteristics of included studies, identified existing barriers and design recommendations for DTC health care AI apps mentioned in the literature and also provided a reference for future design and development.

Methods: This scoping review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines and was conducted according to Arksey and O’Malley’s 5-stage framework. Peer-reviewed papers on DTC health care AI apps published until March 27, 2023, in Web of Science, Scopus, the ACM Digital Library, IEEE Xplore, PubMed, and Google Scholar were included. The papers were analyzed using Braun and Clarke’s reflective thematic analysis approach.

Results: Of the 2898 papers retrieved, 32 (1.1%) covering this emerging field were included. The included papers were recently published (2018-2023), and most (23/32, 72%) were from developed countries. The medical field was mostly general practice (8/32, 25%). In terms of users and functionalities, some apps were designed solely for single-consumer groups (24/32, 75%), offering disease diagnosis (14/32, 44%), health self-management (8/32, 25%), and health care information inquiry (4/32, 13%). Other apps connected to physicians (5/32, 16%), family members (1/32, 3%), nursing staff (1/32, 3%), and health care departments (2/32, 6%), generally to alert these groups to abnormal conditions of consumer users. In addition, 8 barriers and 6 design recommendations related to DTC health care AI apps were identified. Some more subtle obstacles that are particularly worth noting and corresponding design recommendations in consumer-facing health care AI systems, including enhancing human-centered explainability, establishing calibrated trust and addressing overtrust, demonstrating empathy in AI, improving the specialization of consumer-grade products, and expanding the diversity of the test population, were further discussed.

Conclusions: The booming DTC health care AI apps present both risks and opportunities, which highlights the need to explore their current status. This paper systematically summarized and sorted the characteristics of the included studies, identified existing barriers faced by, and made future design recommendations for such apps. To the best of our knowledge, this is the first study to systematically summarize and categorize academic research on these apps. Future studies conducting the design and development of such systems could refer to the results of this study, which is crucial to improve the health care services provided by DTC health care AI apps.


The scarcity and uneven distribution of health care resources, such as medical facilities and professionals, often impedes people’s access to timely and effective health care services and professional medical advice, which has been a significant health concern worldwide [ 1 ]. The World Health Organization (WHO) and other institutions have identified artificial intelligence (AI) as a technology that has the potential to fundamentally transform health care and help address these challenges, especially the reduction in health inequalities in low- and middle-income countries (LMICs) [ 2 , 3 ].

Among AI programs that provide health care functions, there is a significant surge in health care apps that are sold directly to consumers for personal use. Most of these apps are based on predictive or diagnostic functions, providing consumers with a purportedly inexpensive and accurate diagnosis of various conditions [ 4 ]. A well-known example is the Apple Watch for atrial fibrillation, which has been authorized as a class II (moderate-risk) device [ 5 ]. The increased emphasis on telemedicine and home health care in the era of the COVID-19 pandemic [ 6 ], as well as the current advancements in generative AI technologies, such as ChatGPT (where GPT stands for Generative Pretrained Transformer), further stimulate and drive the emergence of direct-to-consumer (DTC) health care AI apps. Large enterprises are racing to deploy research and development of DTC health care AI apps. For example, Dr Karen DeSalvo, Google’s chief health officer, argued at “Check Up 2023” that the future of health is consumer driven. As a company with advanced AI technologies, Google will drive AI-enabled insights, services, and care across a range of health care use cases, from search to symptom tracking and treatment [ 7 ].

However, on the one hand, existing DTC health care AI apps carry risks of errors at both the individual and the societal level. At the individual level, consumers may face the costs and consequences of overdiagnosis or underdiagnosis when using these apps. For example, Google announced an AI-powered dermatology assist app that, according to the company, can use deep learning to identify 288 skin, hair, and nail conditions based on user-submitted images [ 8 ]. However, the app has a significant limitation due to its lack of data diversity, which could lead to overdiagnosis or underdiagnosis in non-White patients [ 9 ]. At the societal level, DTC health care AI apps are designed for cost-effective, immediate, and repeated use, increasing the likelihood that their errors will spread rapidly and place a significant burden on the overall health care system [ 4 ].

On the other hand, the manner in which consumers interact directly with AI in DTC health care AI apps is transformative and alters the traditional physician-patient relationships. These apps can directly provide consumers with various functions, such as heart dysfunction identification [ 10 , 11 ], eye disease diagnosis [ 12 ], and emotion regulation and treatment [ 13 ], which were previously provided by human health care experts. However, in the process of consumers directly interacting with AI, failure to incorporate consumer behavior insights into AI technological development will undermine their experience with AI [ 14 ], thereby affecting their adoption of such apps [ 15 ].

In the context of a surge in DTC health care AI apps, academic research focusing on consumers in the health care AI field is relatively scarce, and there is limited understanding of consumer acceptance of AI in the health care domain [ 16 ]. Furthermore, most trials of clinical AI tools omit the evaluation of patients’ attitudes [ 17 ]. The majority of existing reviews either concentrate on health care AI systems for expert users, such as health care providers [ 18 , 19 ], or do not clearly differentiate the user categories for AI apps in health care [ 20 , 21 ]. There is a need for a deeper understanding of how consumers interact with DTC health care AI apps, beyond merely considering the system’s technical specifications [ 4 ]. Previous studies have reviewed AI apps that are patient oriented and have unique features, functionalities, or formats [ 22 - 24 ]. However, the overall landscape of DTC health care AI apps in academic research remains unclear. There is also a lack of studies that systematically summarize the potential barriers faced by these apps, as well as design recommendations for future research.

To the best of our knowledge, this is the first academic study to systematically summarize and sort out the profile of health care AI apps directly targeting consumers. The objectives of this research are twofold: first, to provide a comprehensive overview of existing studies related to DTC health care AI apps, exploring and mapping out their study characteristics, and, second, to summarize observed barriers and future design recommendations in the literature. Understanding these issues is crucial for the future research, design, development, and adoption of DTC health care AI apps.

Study Design

A scoping review was conducted in line with Arksey and O’Malley’s 5-stage framework [ 25 ]. Study results were reported according to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) checklist [ 26 ] ( Multimedia Appendix 1 ).

Stage 1: Identifying the Research Question

To address the aim of this study, 3 research questions were formulated:

  • Research question 1: What characteristics of DTC health care AI apps have been identified in existing research?
  • Research question 2: What barriers are faced by DTC health care AI apps in existing research?
  • Research question 3: What design recommendations for DTC health care AI Apps have been put forward in existing research?

Stage 2: Identifying Relevant Studies

Studies were searched from inception until March 27, 2023. We searched 5 databases (Web of Science, Scopus, the ACM Digital Library, IEEE Xplore, and PubMed) for 4 concept areas and their lexical variants and synonyms ( Textbox 1 ): AI (technical basis), health care (application domain), consumer (user), and app (carrier). In addition, we retrieved gray literature from the top 10 pages of Google Scholar search results. Gray literature encompasses the literature produced by various levels of government, academia, business, and industry in both print and electronic formats, which is not controlled by commercial publishers [ 27 ]. Its forms include academic papers, dissertations, research and committee reports, government publications, conference papers, and ongoing research, among others.

Search concepts combined using “AND”

  • Artificial intelligence (AI)
  • Health care

Search terms combined using “OR”

  • AI, artificial intelligence, ML, machine learning, DL, deep learning
  • Health care, health, medical
  • Consumer, consumers
  • Application, applications, app, apps, system, systems, service, mHealth, eHealth

We also conducted snowball sampling on the reference lists of related papers included in the full-text review. The specific database search strings combined with Boolean operators are detailed in Multimedia Appendix 2 .

Stage 3: Study Selection

Inclusion criteria for this review were (1) peer-reviewed studies, (2) research papers, (3) papers published in English, (4) research topics focused on DTC health care AI apps or systems, and (5) either consumers as target users or multistakeholder users with consumers as main users. Exclusion criteria were (1) duplicate papers not identified by bibliography software, (2) nonresearch papers (eg, editorials, commentaries, perspectives, opinion papers, or reports), (3) papers not published in English, (4) inability to obtain the full text, and (5) app only intended to be used by professionals.

Inclusion and exclusion criteria ( Table 1 ) were used to screen titles, abstracts, and full-text papers. When the 2 authors (XH and XZ) disagreed on the selection of studies, consensus was reached through discussion.

a DTC: direct to consumer.

b AI: artificial intelligence.

Stage 4: Charting the Data

Two authors (XH and XZ) extracted the following data for each paper: title, author, publication year, country, publication type, study objective, study design, medical field, app type, user, existing barriers, and design recommendations. We exclusively extracted data related to barriers and design recommendations from the results or discussions within the papers (eg, insights, such as opinions expressed by consumers after using the apps or recommendations proposed by researchers following app evaluations). Descriptions that were not validated through the empirical research section of the papers were not extracted (eg, viewpoints that appeared only in the Introduction or Background section).

Stage 5: Collating, Summarizing, and Reporting Results

The extracted data related to RQ1 were mapped and summarized. A reflexive thematic analysis [ 28 - 30 ] was conducted on the data related to RQ2 and RQ3 to summarize existing barriers faced by and design recommendations for DTC health care AI apps through inductive coding. NVivo (QSR International) was used to facilitate data management and analysis. The analysis proceeded through 6 steps: familiarizing with the data set; coding; generating initial themes; developing and reviewing themes; refining, defining. and naming themes; and writing up. The coding and data analysis for this study were performed in parallel, and we addressed differences and reached consensus by discussing uncertainties.

Search Results

The initial search resulted in the retrieval of 4055 records. After removing duplicates, 2898 (71.5%) records remained. After screening titles and abstracts, 2752 records (95%) were excluded, and the remaining 146 (5%) records were assessed for eligibility through full-text review. An additional 3 records were obtained through a snowball search of the reference lists in the included full-text papers. Of these 149 records, 115 (77.2%) were excluded for reasons shown in Figure 1 , resulting in 32 (21.5%) papers being included in the final scoping review. Figure 1 shows the PRISMA-ScR (Preferred Reporting Item for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) flow.

nursing research articles on incontinence

Research Question 1: Study Characteristics

An overview of the 32 papers included in the scoping review is provided in Tables 2 - 4 , including author, publication year, country, publication type, study objective, study design, medical field, app type, and user. We did not restrict the search year intentionally, as most health care AI review papers do [ 31 - 33 ]. However, the results indicated that the reviewed papers were fairly recent, with all the 32 (100%) included studies published between 2018 and 2023. Papers were from North America (7/32, 22%) [ 10 , 13 , 15 , 34 - 38 ], Asia (6/32, 19%) [ 39 - 44 ], Europe (6/32, 19%) [ 12 , 45 - 49 ], and Oceania (2/32, 6%) [ 17 , 50 ]. In addition, multiple regional cooperation was also prevalent (11/32, 34%) [ 51 - 61 ]. Publication types included 23 (72%) journal papers ( Tables 2 and 3 ) [ 10 , 12 , 15 , 17 , 34 , 37 , 39 , 41 , 43 , 45 - 49 , 52 - 62 ] and 9 (28%) conference papers ( Table 4 ) [ 13 , 35 , 36 , 38 , 40 , 42 , 44 , 50 , 51 ]. Study designs included quantitative research (22/32, 69%) [ 12 , 13 , 15 , 34 , 37 , 39 , 40 , 42 - 44 , 47 - 52 , 54 , 55 , 57 - 59 , 61 ], qualitative research (2/32, 6%) [ 35 , 60 ], and mixed methods studies (4/32, 12%) [ 38 , 41 , 45 , 46 ], in addition to systematic reviews (4/32, 12%) [ 17 , 36 , 53 , 56 ]. Most studies chose general practice (8/32, 25%) [ 34 , 37 , 40 , 41 , 46 , 49 , 54 , 55 ] as the target medical field. The app types mentioned in the studies included diagnosis (apps make determinations about the cause of a disease or pathology based on information provided by consumers; 14/32, 44%) [ 12 , 38 , 40 - 42 , 47 , 48 , 51 , 52 , 54 , 55 , 57 , 60 , 61 ], health self-management (apps encourage consumers to take actions to manage their continuous health status and quality of life, often in the management of chronic diseases or health problems; 8/32, 25%) [ 13 , 43 , 44 , 49 , 50 , 56 , 58 , 59 ], and health care information inquiry (apps extract relevant information from a large amount of health care information and generate answers based on consumer questions in common forms, such as conversational agents; 4/32, 13%) [ 35 , 37 , 39 , 46 ]. There were also review papers (4/32, 13%) [ 17 , 36 , 53 , 56 ] that reviewed apps involving more than 1 of the aforementioned function types. Some of these apps were aimed at the single-consumer group (24/32, 75%) [ 12 , 13 , 15 , 34 - 43 , 45 - 48 , 50 , 51 , 54 - 57 , 60 , 61 ], while other apps not only targeted consumers as the main users but also targeted user groups with other identities, including physicians (5/32, 16%) [ 17 , 49 , 52 , 53 , 59 ], health departments (2/32, 6%) [ 42 , 44 ], nursing staff (1/32, 3%) [ 58 ], and patients’ family members (1/32, 3%) [ 59 ]. Figure 2 shows an overview of the study characteristics of DTC health care AI apps, including country, year, application type, user, medical field, and study design.

a AI: artificial intelligence.

b N/S: not specified.

c XAI: explainable artificial intelligence.

d ECG: electrocardiogram.

nursing research articles on incontinence

Research Question 2: Barriers

We identified 8 barriers to designing and developing DTC health care AI apps: (1) lack of explainability and inappropriate explainability, (2) lack of empathy, (3) effect of information input method and content on usability, (4) concerns about the privacy protection ability, (5) concerns about the AI accountability system, (6) lack of trust and overtrust, (7) concerns about specialization, and (8) the unpredictable future physician-patient relationship. These 8 existing barriers faced by DTC health care AI apps, along with their related subthemes, and the number of studies mentioning them are shown in Figure 3 .

nursing research articles on incontinence


Lack of explainability.

Of the 32 studies, 10 (31%) [ 10 , 34 , 36 , 38 , 41 , 46 , 51 , 52 , 54 , 60 ] pointed out that the explanations provided by existing DTC health care AI apps are insufficient. Existing studies mostly provided explanations primarily for domain experts, paying less attention to the explainability needs of lay users, such as consumers [ 41 ]. In addition, 2 (6%) studies [ 46 , 51 ] pointed out that current DTC health care AI apps lack the explanations of relevant knowledge in the AI field (ie, the explanations of the working principle of the machine learning algorithm used by the apps, such as how AI correctly responds to consumers’ health consultations [ 46 ]). Furthermore, 4 (13%) studies [ 34 , 46 , 51 , 54 ] indicated that current DTC health care AI apps lack explanations of relevant knowledge of the medical field, such as highly specialized medical terminology [ 34 ] and rare diseases that have only been discussed in professional literatures [ 54 ], and 4 (13%) studies [ 36 , 38 , 51 , 60 ] pointed out the disadvantages of a lack of explainability, which caused consumers to doubt the usefulness, accuracy, and safety of the apps and even possibly view them as a threat. Moreover, 1 (3%) study [ 51 ] mentioned the advantages of providing explanations, which aided consumers in understanding the reasoning of the system, and this understanding was crucial for boosting the trust of lay users.

Inappropriate Explainability

Of the 32 studies, 3 (9%) [ 38 , 41 , 60 ] highlighted that current DTC health care AI apps contain inappropriate explanations. Specifically, 2 (6%) studies [ 38 , 41 ] mentioned that excessive explanations can result in information overload for users, which in turn would negatively impact the user experience and might cause users to ignore system prompts or suggestions. In addition, 1 (3%) study [ 60 ] pointed out that the poor information quality of explanations would be considered by users as “invalid, meaningless, not legit, or a bunch of crap” and even cause users to perceive it as a risk, prompting them to seek secondary confirmation of information through other channels (eg, online search or consultation with a doctor) to ensure their own safety. Furthermore, 2 (6%) studies [ 38 , 41 ] indicated that improper levels of transparency or inappropriate presentation formats in explanations can pose risks, potentially harming the interests of other stakeholders in the AI system or affecting the authenticity of users’ future performances. Specifically, inappropriate transparency of explanations might lead to the disclosure of sensitive details and intrusion of systems, harming the interests of AI service providers and violating the privacy of other consumers [ 38 ]. Explaining to users how a particular feature would accurately affect the disease diagnosis might affect their performance authenticity in the future diagnosis of related diseases, allowing them to manipulate the likelihood of being diagnosed or not diagnosed by deliberately meeting or avoiding meeting the characteristic threshold, respectively [ 41 ]. Inappropriate presentation forms of explanations, such as the function of counterfactual explanations that allowed users to freely edit data to view different diagnostic results, were popular with physicians because they met the needs of medical users to test different data and corresponding diagnostic possibilities, but they might become technical loopholes in the commercialization of DTC health care AI apps. Users could exploit this feature to input data for multiple individuals and view different results, thereby avoiding multiple payments and compromising the economic interests of the AI service provider [ 41 ].

In a total of 8 (25%) studies [ 17 , 36 , 39 , 41 , 45 , 46 , 51 , 60 ], users felt that AI lacked empathy and was impersonal. Among them, users in 2 (6%) studies [ 45 , 46 ] felt that AI was unable to understand emotion-related issues, especially mental health problems, and 2 (6%) studies [ 41 , 60 ] pointed out that the information-conveying method of AI, such as transmitting complex disease information without human presence [ 60 ] and explaining the disease from the perspective of “how bad it is” [ 41 ], could also lead users to think that AI is indifferent and inhumane. In addition, 5 (16%) studies [ 36 , 39 , 41 , 46 , 60 ] reported that the lack of empathy would lead to a series of negative consequences, including triggering users’ frustration, disappointment, anxiety, and other negative emotions [ 36 , 60 ]; impeding users’ acceptance of such apps [ 39 , 46 ]; and even affecting their subsequent treatments [ 41 ]. Furthermore, according to 2 (6%) studies [ 46 , 51 ], some users preferred to consult human physicians rather than AI because they could offer comfort and spiritual support.

Restricted Information Input Method

Of the 32 studies, 2 (6%) [ 36 , 54 ] pointed out that the restricted information input method in DTC health care AI apps (eg, a single way of typing) made users feel helpless and frustrated, which was contrary to their usage expectations, and even made them inclined to discontinue use.

Lack of Actionable Information

Of the 32 studies, 2 (6%) [ 10 , 54 ] pointed out that DTC health care AI apps lacked actionable information content, failing to inform users of the next actions to take, such as where to seek medical assistance.

In total, 4 (12%) studies [ 15 , 46 , 51 , 60 ] raised concerns about the ability of DTC health care AI apps to protect privacy, such as safeguarding users’ sensitive health-related information from data breaches. Users were concerned that their personal information (eg, habits, preferences, and health records) would be collected without their knowledge [ 46 ], that anonymous data would be re-identified through AI processes [ 15 ], that data would be sold by companies for secondary exploitation [ 51 ], and that their health data would be hacked and used against them [ 60 ].

Accountability and Supervision

In total, 4 (12%) studies [ 12 , 17 , 41 , 60 ] raised concerns about the accountability of DTC health care AI apps, and 2 (50%) of these studies [ 17 , 41 ] indicated that only few controversial studies exist on the distribution of AI responsibilities. Another study [ 12 ] exemplified the practice of some application manufactures who made general recommendations (eg, “recommend emergency care”) for almost every diagnosis, thereby transferring responsibility to users. In some countries, according to 1 (3%) study [ 17 ], there were concerns with the supervision of DTC health care AI apps. The absence of human supervision during the design, development, and deployment of AI not only failed to ensure the anticipated benefits but also posed a risk of potential injury to users.

Lack of Trust

A total of 10 (31%) studies [ 15 , 17 , 36 , 41 , 43 , 46 , 52 , 54 , 60 , 61 ] pointed out that users lacked trust in DTC health care AI apps. Among them, 5 (50%) studies [ 15 , 17 , 54 , 60 , 61 ] distrusted AI due to inadequate performance or the lack of performance explanations, 3 (30%) studies [ 41 , 43 , 46 ] found that even if the AI performed as well as or better than human physicians, users still placed more trust and reliance on humans, and 3 (30%) studies [ 15 , 36 , 52 ] indicated that users’ lack of trust might cause them to disregard AI recommendations or even stop using such apps.

Based on the calibration between trust and competence, trust can be divided into 3 levels: calibrated trust, distrust, and overtrust. Distrust refers to users being less willing to trust AI compared to similar human providers, even if AI shows superior performance; overtrust refers to the user’s trust in the system beyond its actual capabilities [ 63 ]. Of the 32 studies, 2 (6%) [ 47 , 52 ] indicated that users’ overtrust issues in DTC health care AI apps would impose a double burden on both individuals [ 47 , 52 ] and society [ 47 ]. At the individual level, 2 (6%) studies [ 47 , 52 ] pointed out that overtrusting false-positive results could result in users’ negative emotions (eg, stress [ 52 ]). Tools with a high rate of false positives might also reduce users’ trust in true-positive results [ 47 ]. In addition, 1 (3%) study [ 52 ] pointed out that overtrusting false-positive results could trigger users’ unnecessary behaviors, such as unnecessary medical treatment, while 1 (3%) study [ 47 ] pointed out that overtrusting false-negative results would provide users with a false sense of security and delay the disease diagnosis. At the societal level, 1 (3%) study [ 47 ] indicated that individuals’ overtrust in false-positive results could overwhelm the entire health care system, whereas individuals’ overtrust in false-negative results could exacerbate the social transmission of diseases (eg, COVID-19).


In total, 2 (6%) studies [ 48 , 51 ] raised concerns about the specialization of DTC health care AI apps. To be specific, users in 1 (3%) study [ 51 ] doubted the feasibility of substituting consumer-grade equipment for professional medical-grade equipment. For example, they argued that an artificial intelligence–electrocardiogram (AI-ECG) smartwatch that measured only the wrist could not replace a traditional ECG machine with 12 electrodes for detecting heart diseases. The other study [ 48 ] pointed out that the professional effect of DTC health care AI apps is influenced by the using environment. For example, an app that detects obstructive sleep apnea, which is affected by background noise, might work in tightly controlled laboratory conditions but might not be as accurate in in-home environments.

Physician-Patient Relationship

In total, 2 (6%) studies [ 17 , 53 ] believed that DTC health care AI apps would make the physician-patient relationship less predictable. As a result of AI user empowerment and the emergence of “do-it-yourself” medicine, users were less reliant on medical experts [ 17 ] and expert medical advice [ 53 ]. The effects of AI on the physician-patient relationship remains to be evaluated by more studies [ 53 ].

Research Question 3: Design Recommendations

The themes of design recommendations covered 6 types of recommendations and their specific contents mentioned by existing studies when designing and developing DTC health care AI apps: (1) enhance explainability, (2) improve empathy, (3) improve usability, (4) enhance privacy protection ability, (5) address AI accountability at both the individual and the government level, and (6) improve the diversity of participants to enhance inclusion. These 6 design recommendations for DTC health care AI apps, as well as the related subthemes and the number of studies mentioning them, are shown in Figure 4 .

nursing research articles on incontinence

Enhance Explainability

Of the 32 studies, 5 (16%) [ 41 , 43 , 46 , 54 , 60 ] suggested designing and developing explainable DTC health care AI apps from 3 perspectives: the explanations’ primary content, their presentation form, and their legislation. First, 4 (13%) studies [ 41 , 46 , 54 , 60 ] provided content recommendations for explanations: input (explanations of the input data) [ 41 , 54 ], output (explanations of the generated output) [ 41 ], the how (explanations of how the system as a whole works) [ 41 , 54 , 60 ], performance (explanations of the capabilities, limitations, and verification process of the current system) [ 41 , 46 , 54 , 60 ], the why (explanations as to why, and why not, the system made a specific decision) [ 41 ], what-if (explanations to speculate on the system’s output under a particular set of settings and to describe what the system would do) [ 41 ], responsibility (explanations of the system’s accountability) [ 41 ], ethics (explanations of information from regulatory approvals or peer-reviewed publications that validated the system) [ 41 ], the social effect (explanations of the results of other social subjects using the system) [ 41 ], and domain knowledge (explanations of specific AI or medical terms and information sources in the system) [ 41 , 54 ]. Second, based on the complex diversity of consumer groups with varying domain knowledge, cognitive styles, and urgency of symptoms, 1 (3%) study [ 41 ] provided suggestions for explanations’ presentation forms: using a progressive disclosure approach to present various levels and formats of explanations to meet the needs of a wider consumer group. Third, 1 (3%) study [ 43 ] provided legislative suggestions for explanations: future governments and regulatory agencies, particularly in the medical field, would need to further establish and improve the legal framework for transparent AI to safeguard the right of consumers to obtain explanations based on algorithmic decisions.

Improve Empathy

In total, 6 (19%) studies [ 15 , 36 , 41 , 49 , 55 , 60 ] designed and developed empathetic DTC health care AI apps. Specifically, 3 (9%) studies [ 15 , 36 , 49 ] suggested that such apps could directly incorporate conversational agents or refer to research results in this field to embed richer semantics [ 49 ] and add more social cues [ 15 ], while 2 (6%) studies [ 41 , 60 ] suggested focusing on skills for delivering stressful information.

Improve Usability

In total, 6 (19%) studies [ 34 , 38 , 41 , 49 , 54 , 60 ] enhanced the usability of DTC health care AI apps in 3 aspects: information input method, result output form, and content actionability. Concerning the information input method, 1 (3%) study [ 54 ] suggested simplifying the way consumers input data (eg, by sharing and describing information in the form of audio recordings) to save their time and effort, while 1 (3%) study [ 49 ] simplified the way consumers input data (eg, by barcode-scanning prescription data) to reduce the risk of manual data entry errors. Concerning the result output form, 1 (3%) study [ 34 ] translated or simplified highly specialized language that was difficult for consumers to understand (eg, rare diseases that were only discussed in professional literature) and also provided illustrations to summarize the output; 2 (6%) studies [ 38 , 41 ] suggested avoiding outputting too much and too detailed information at once so as to prevent consumers from information overload. Concerning content actionability, 1 (3%) study [ 54 ] suggested, at the initial stage of interaction, providing introductory materials to teach consumers the most effective way to use advanced technology (eg, introducing basic functions, limitations, and the use process); 1 (3%) study [ 41 ] suggested, during the interaction, clearly explaining the purpose of the current operation and context-related information to consumers and informing them of the results of the current operation directly on the interface; and 1 (3%) study [ 54 ] suggested, at the end of the interaction, informing consumers of the next step (eg, where to seek medical help).

Enhance Privacy

Of the 32 studies, 3 (9%) [ 15 , 38 , 51 ] suggested enhancing the privacy protection capabilities of DTC health care AI apps to prevent consumers’ privacy from being violated. Specifically, the recommended using state-of-the-art technology to encrypt and authenticate users’ health data [ 51 ], obtaining informed consent for health care purposes to prevent data from being resold and exploited [ 15 ], and avoiding explanations with inappropriate transparency (eg, leaking flaws in algorithms or detecting sensitive data sources) to prevent systems from being intruded [ 38 ].

Address Accountability

In total, 4 (12%) studies [ 43 , 45 , 48 , 56 ] addressed the accountability issues of DTC health care AI apps from both individual and government perspectives. At the individual level, 1 (3%) study [ 47 ] addressed accountability by informing consumers whether the app was officially certified and encouraging them to seek professional medical advice or clinical testing beyond the app, and 1 (3%) study [ 49 ] empowered patients and provided them with more responsibilities (eg, motivating patients to take their medications, while informing them of possible drug interactions) but still opted for human medical staff to undertake the responsibility for complete drug therapy. At the government level, 1 (3%) study [ 60 ] suggested developing policies or guidelines to regulate the use of such apps and establish accountability mechanisms through legislation for AI output, and 1 (3%) study [ 52 ] suggested that national health authorities should clarify the position of these apps in the health care system (eg, whether they were for laypersons, general practitioners, or specialists).

Improve Diversity

In total, 6 (19%) studies [ 41 , 46 , 52 , 54 , 55 , 60 ] designed and developed DTC health care AI apps by diversifying the test populations of the diseases targeted by apps in the future. Specifically, studies focused on clinical populations [ 46 ], community populations [ 46 ], marginalized populations (eg, populations with low education levels [ 60 ] and the elderly [ 54 , 60 ]), and children [ 55 ] and the cultural and social factors in these populations [ 54 ] in order to capture more diverse user needs and develop a more comprehensive solution.

Principal Findings

In the context of a surge in DTC health care AI apps, this scoping review identified 32 studies in the existing academic literature that address this topic. The review summarized the characteristics of existing studies on DTC health care AI apps, highlighted 8 categories of extant barriers, and pointed out 6 categories of design recommendations.

Study Characteristics

In terms of the developmental timeline, although AI has been extensively used across various sectors of health care, studies focusing on DTC health care AI apps are still in their nascent stages. We did not artificially restrict the time frame for our review; however, the papers included in our results were all published recently (between 2018 and 2023).

In terms of geographical origins, the studies on DTC health care AI apps predominantly came from high-income countries, particularly the United States. This aligns with other reviews in the domain of health care AI [ 21 , 31 , 64 ]. This correlation is intrinsically tied to the fact that a more advanced digital health care infrastructure (eg, electronic health records (EHRs), health information exchanges (HIEs), and telehealth platforms) is present in these countries. More geographically diverse research is needed in the future, and we particularly expect a surge in studies originating from LMICs, because AI is considered a technology that can help bridge the digital gap and reduce health inequities worldwide [ 2 , 3 , 64 ]. However, the current study outcomes from high-income countries cannot be directly transferred to low-income regions due to significant risks, such as output bias, poor performance, or erroneous results, when using AI solutions trained in contexts that differ substantially from the local populations [ 65 ]. When AI systems are applied to new populations with differing living environments or cultural backgrounds, adaptations to the local clinical settings and practices are required, and the measures and outcomes for design, development, and evaluation may vary [ 41 , 66 ].

In terms of the study design, the majority of the papers we reviewed opted for quantitative methods to evaluate the apps, such as collecting performance metrics when consumers use the apps or obtaining quantitative data on existing user experience dimensions through questionnaires. Fewer papers delved into the barriers and recommendations arising from users’ usage of DTC health care AI apps. However, given that the emergence of such apps is still a nascent phenomenon, future work requires more qualitative research to explore the effects generated by these technological systems when used in society, to dig out initially overlooked new themes or deeper insights, and to assess user experiences beyond what short-term metrics can capture, while also incorporating edge cases that large-scale studies may overlook [ 67 , 68 ].

In terms of medical fields, existing studies on DTC health care AI apps primarily focused on the field of general practice. This is understandable because general practice usually serves as the first medical contact point for patients [ 69 ], thereby having a broad spectrum of user needs. Moreover, the health issues diagnosed and treated in general practice are generally more common and less complex [ 70 ], thereby presenting relatively lower risks. Consequently, most studies chose general practice as the entry point for the medical fields of designing and developing DTC health care AI apps.

In terms of intended users and provided functionalities among studies on DTC health care AI apps, some were designed solely for single-consumer user groups, offering functions such as disease diagnosis, health self-management, and health care information inquiry. Others also connected with other user groups, including physicians, family members, nursing staff, and health care departments, generally to alert these groups to abnormal conditions of consumer users. For example, these functionalities may include alerting hospitals about consumer user falls due to stroke, notifying physicians and family members about medication adherence issues, referring users with high-risk skin cancer ratings to doctors, or informing health care departments about potential diagnoses of COVID-19 or other infectious diseases. However, it is crucial to note that although such intelligent functionalities for alerting other groups about users’ anomalies may contribute positively to users’ health and the efficient functioning of health care systems, they also pose risks related to consumers’ human rights, democracy, false positives due to erroneous data capture, and even the manipulation of users with low behavioral capacity [ 71 ]. Future DTC health care AI apps, when designing features that involve 2 or more user groups, must consider how to allocate, balance, and constrain power among various stakeholders, while simultaneously ensuring ethical and legal compliance as they seek to benefit consumer groups in need.

Barriers and Design Recommendations

In terms of barriers and design recommendations, it is noteworthy that many challenges are not confined solely to apps targeting consumers; rather, they exhibit considerable similarities with the issues encountered by health care AI systems designed for other user groups, such as health care professionals. First, privacy concerns have been widely recognized as a significant barrier to the application of AI in the health care domain [ 20 , 21 , 72 , 73 ]. Privacy protection has become a hot topic in the health care AI research field [ 74 ], with numerous studies dedicated to developing innovative privacy-preserving solutions without compromising the performance of big data–driven AI models. These include developing privacy-enhancing technologies, such as homomorphic encryption [ 75 ], securing multiparty computation and differential privacy [ 76 ], and exploring new training methods and data governance models, such as distributed federated machine learning using synthesized data from multiple organizations [ 77 ], data-sharing pools [ 78 ], data trusts [ 79 ], and data cooperatives [ 80 ]. Second, the lack of clarity in accountability and regulation has also been universally identified in prior research as a key obstacle to the application of AI in health care [ 81 - 83 ]. Despite the existence of various worldwide policies and regulations concerning AI accountability and regulation, such as WHO [ 84 ], the General Data Protection Regulation (GDPR) [ 85 ], the Food and Drug Administration (FDA) [ 86 ], Health Canada [ 87 ], and the AI Act [ 88 ], the rapid advancement of AI technology makes it difficult for existing regulatory frameworks to keep up, let alone be able to anticipate its potential risks and impacts. Taking the AI Act, which is currently being advanced in Europe, as an example, the emergence of new generative AI systems, such as ChatGPT, has already posed challenges to the universality and applicability of this legislation [ 89 ]. Furthermore, usability has also been shown in previous studies concerning physicians as an aspect that doctors wish to see improved in health care AI tools, such as clinical decision support systems [ 41 , 66 ]. Additionally, the evolution of physician-patient relationships has been identified as a key point requiring long-term tracking following the deployment of various types of health care AI systems [ 90 ].

In addition to identifying challenges similar to those faced by health care AI systems targeted at other user groups, this review further identified some more subtle obstacles that are particularly worth noting in consumer-facing systems and distilled corresponding design recommendations, including enhancing human-centered explainability, establishing calibrated trust and addressing overtrust, demonstrating empathy in AI, improving the specialization of consumer-grade products, and expanding the diversity of the test population.

Enhance Human-Centered Explainability

The review findings identified current barriers to explainability in DTC health care AI apps, which included not only providing inadequate explanations to consumers (a lack of explanations relating to both AI and medical domain knowledge) but also providing inappropriate explanations to consumers (excessive content caused information overload to consumers, low-quality content exposed consumers to risks and burdens, and improper transparency and presentation forms could adversely impact other stakeholders’ interests in the system). To address these barriers, our review offered design recommendations for improvements in the content, form, and legislative aspects of explanations, which future research can consider.

Furthermore, we believe that the review results demonstrate and re-emphasize the importance of designing, developing, and evaluating AI explainability from a human-centered perspective. As AI increasingly powers decision-making in high-risk areas, such as health care, explainable artificial intelligence (XAI), aimed at enabling humans to understand the logic and outcomes of AI systems, has become a research hotspot in recent years [ 91 - 95 ]. Within this interdisciplinary field, algorithm-centered approaches aim to enhance the transparency of AI models and to develop inherently explainable models [ 96 ], while human-centered approaches emphasize considerations such as who the users of explanations are, why explanations are needed (eg, how social and individual factors influence explainability objectives), and what the timing and context of providing explanations (eg, contextual variations in explainability across different application domains) are [ 97 , 98 ]. As shown in our findings, consumers of health care AI had various needs concerning the content and form of explanations, and their interactions with explanations could influence their adoption toward the apps and subsequent behavior. Furthermore, wrong explanation design could produce correlation effects on other stakeholders in the AI system. All these findings indicate that the challenges in explainability in DTC health care AI apps are not merely technical issues concerning algorithmic transparency but also significantly involve human factors. Future studies need to enhance the explainability of DTC health care AI apps from a human-centered perspective, focusing on the cognitive abilities, physical characteristics, and social and psychological factors of the human in the loop, as well as how these human factors interact with explanations, AI systems, and the environment. This will enable the design of DTC health care AI apps that meet user needs and enhance human performance, safety, and overall well-being.

Establish Calibrated Trust and Pay Special Attention to Overtrust

Our findings indicated that current DTC health care AI apps face challenges related to trust, including both a lack of trust and overtrust. The need to establish calibrated trust in AI systems, meaning cultivating the users’ ability to know when to trust (accept correct advice) or not trust (reject erroneous advice) AI [ 99 ], has reached a consensus in current research [ 100 ]. Under this premise, we believe that future designs of DTC AI apps should pay more attention to the issue of overtrust. There are multiple rationales for this focus. On the one hand, from an academic research perspective, most extant studies on AI trust predominantly center on enhancing users’ trust [ 101 - 104 ], with less attention given to the issue of overtrust; on the other hand, from a practical application perspective, 3 influencing factors also need to be considered:

  • First, the users’ background knowledge. Consumers often possess limited prior knowledge of both medical and AI domains related to these apps [ 4 ], affecting their receptivity to AI advice. Research has shown that domain experts are more likely to question AI suggestions, whereas nonexperts are more receptive to them [ 105 ].
  • Second, the differential risk in decision-making: Consumers and health care professionals differ in their risk assessments when facing AI advice. Typical consumers are loss averse; for them, changes for the worse (losses) loom larger than equivalent changes for the better [ 106 ]. Hence, they are more inclined to accept AI advice and take subsequent medical actions, rather than potentially missing out on timely disease diagnosis and treatment if AI advice is not adopted [ 4 ]. In contrast, the biggest concern of health care professionals when adopting new products to assist medical diagnosis may not be the pursuit of improvement in work performance but the potential risks to patients’ lives and health [ 107 ], so their adopting is relatively cautious.
  • Third, the drive for commercial interests may also prompt these apps to exaggerate their capabilities, thereby further exacerbating the issue of consumer overtrust [ 36 ].

Therefore, in summary, although both domain expert and nonexpert users may display overreliance on automation [ 108 ], physicians’ overtrust in AI diagnostic features is not commonly observed at this current stage of medical AI development; many reviews in the AI domain concerning physician users, while identifying trust issues, primarily discuss a lack of trust [ 66 , 109 ]. However, consumer overtrust in health care AI, along with the ensuing personal and societal effects, has already emerged as an issue that needs to be considered sooner rather than later.

Demonstrate Empathy in Artificial Intelligence

Our review indicated that even if AI can be more accurate and logical, its lack of empathy may hinder consumer acceptance of DTC health care AI apps. Empathy, defined as the ability to understand or feel what other individuals are experiencing from their frame of reference [ 110 ], is widely acknowledged as a fundamental value for achieving optimal health care practices. It is crucial for enhancing patient satisfaction, treatment compliance, and clinical outcomes [ 111 - 113 ]. In conventional medical settings, health care professionals act as the conveyors of empathy, while patients are the recipients [ 114 ]; in human-AI collaborative medical settings, such as physicians using AI for diagnostic assistance, AI primarily contributes to improving efficiency and decision-making quality, allowing health care professionals to have more time and energy to convey empathy and improve overall treatment satisfaction [ 115 ]; However, in DTC health care AI scenarios, the initial touchpoint no longer has a human element, necessitating AI to become the direct conveyor of empathy.

The topic of AI empathy in health care has become a research hotspot [ 116 - 118 ]. To address this challenge, our review offered several design recommendations: embedding richer semantics and social cues through conversational agents, as well as techniques for conveying stressful information. Current cutting-edge research supports these design suggestions for enhancing empathy through conversational agents. Studies indicate that the new generation of AI chatbots, such as ChatGPT, has scored higher than human doctors in terms of empathy [ 119 ]. Our review is current up to March 2023, and the research included in the review has not yet covered ChatGPT. Therefore, the future integration of ChatGPT or similar large language model chatbots could potentially help alleviate the empathy barriers in DTC health care AI apps.

Improve Specialization of Consumer-Grade Products

Concerns regarding the specialization of DTC health care AI apps are totally understandable. First, from a scientific and technological standpoint, many health care AI apps on the consumer market have scarcely undergone original research for effectiveness or are loosely based on scientific studies but lack a scientific consensus on their efficacy [ 120 ]. Furthermore, the data collection devices for these apps are often consumer-owned smartphones, personal computers, or wearables designed for portability, rather than specialized medical devices tailored for specific disease domains.

Second, in terms of regulatory frameworks, in the United States, where most companies producing DTC health care AI products are located, existing tiered regulatory systems permit the manufacture of general wellness products without adhering to regulations typically applicable to devices intended for diagnosing or treating diseases [ 86 ]. Consequently, driven by commercial interests, the current market is flooded with numerous tools that are approved as general health products but subtly imply that they can be used for diagnosis or treatment. Consumers can easily access these products, although the products may not have undergone rigorous testing and regulation, thus rendering their effectiveness uncertain [ 71 , 121 ].

Existing research is working to close the performance gap between consumer-grade products and clinical-grade medical devices through technological innovations, for example, developing high-precision flexible sensors to improve the data collection capabilities of wearable devices [ 122 , 123 ], as well as through algorithm-hardware cooptimization to ensure model quality is not compromised while achieving device miniaturization [ 124 ]. However, overcoming this barrier will require not only technological advancements but also further refinement of the approval and regulatory frameworks for consumer-grade AI products in the future.

Expand the Diversity of Test Populations

The need to expand the diversity of test populations is also a future direction in the design and development of DTC health care AI apps, as identified by our review. It is worth noting that whenever this theme is mentioned in the papers included in our review, it appears in the Limitations or Future Work section. This indirectly indicates that it is a prevalent yet unresolved issue in this field of research. In existing research, either the test population involves a small subset of patients in the specific disease area with limited demographic characteristics and health information literacy or it is not even the target population for the disease but rather comprises participants recruited through convenience sampling. However, if such apps truly enter the market, their actual consumer users constitute an extremely broad and heterogeneous group, with widely varying demographic characteristics, education levels, and health and information literacy [ 125 ]. Applying AI models trained on small sample data and user feedback obtained from these samples to a broader population could pose multiple risks, including inaccuracies in AI diagnostics and predictions, poor generalization ability to unseen patient data, and perpetuating biases and exclusions against marginalized groups [ 126 ]. These risks could consequently misguide clinical decisions, exacerbate health care inequalities, and trigger legal and ethical crises. Future studies on DTC health care AI apps indeed needs to consider the diversity of the consumer population in terms of culture, society, demographics, and knowledge accomplishment in order to develop more accurate and inclusive health care AI solutions.


This study has a few limitations. First, we retrieved papers written in English, thereby potentially overlooking influential papers published in other languages. Additionally, we only captured papers that were found in the search. Given the novelty of the field and terminology associated with DTC health care AI apps, some relevant studies may have been omitted. However, we attempted to mitigate this limitation by using Google Scholar to search for gray literature and by snowball-sampling from the reference lists of relevant papers. Due to the wide-ranging formats and scopes of gray literature, it often serves as a robust source of evidence in systematic reviews, offering extra data not found in commercial publications, thus reducing publication bias and enabling a more balanced view of evidence [ 27 ]. Google Scholar’s gray literature includes papers from databases that have not yet been formally published, such as arXiv and medRxiv, helping capture research that might be overlooked due to the novelty of the field and terminology.

Furthermore, when using qualitative thematic analysis to synthesize study findings and generate themes, the themes produced were potentially influenced by the prior research experience and personal understanding of the 3 authors. Therefore, the themes may not be entirely comprehensive or may differ when other researchers replicate the coding process. To minimize potential coding bias, we strictly adhered to the 6 key steps of qualitative thematic analysis: familiarizing oneself with the data set; coding; generating initial themes; developing and reviewing themes; refining, defining, and naming themes; and writing up. Each step underwent group discussions, triangulation, and interrater reliability checks among the 3 authors to resolve disagreements and reach a final consensus, thereby striving to maintain consistency and reduce individual differences.

To the best of our knowledge, this is the first study to systematically summarize and organize academic research targeting consumers through DTC health care AI apps. In this study, we delineated the current characteristics of studies focusing on DTC health care AI apps, identified 8 existing barriers, and offered 6 design recommendations. We believe that future research, by considering the key points raised in this study, addressing existing barriers, and referencing design recommendations, can better advance the study, design, and development of DTC health care AI apps, thus improving the health care services they provide.


This work was supported by the Teaching Research Project of the Huazhong University of Science and Technology (grant number 2023038).

Data Availability

All data generated and analyzed during this study are included in this published paper and its Multimedia Appendices.

Conflicts of Interest

None declared.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) checklist.

Database search details.

  • World Health Organization. Global Strategy on Human Resources for Health: Workforce 2030. Geneva. World Health Organization; 2016.
  • Alami H, Rivard L, Lehoux P, Hoffman SJ, Cadeddu SBM, Savoldelli M, et al. Artificial intelligence in health care: laying the foundation for responsible, sustainable, and inclusive innovation in low- and middle-income countries. Global Health. 2020 Jun 24;16(1):52 [ https://globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-020-00584-1 ] [ CrossRef ] [ Medline ]
  • Global strategy on digital health 2020-2025. World Health Organization. 2021. URL: https://www.who.int/docs/default-source/documents/gs4dhdaa2a9f352b0445bafbc79ca799dce4d.pdf [accessed 2023-12-08]
  • Babic B, Gerke S, Evgeniou T, Cohen IG. Direct-to-consumer medical machine learning and artificial intelligence applications. Nat Mach Intell. 2021 Apr 20;3(4):283-287 [ CrossRef ]
  • De novo classification request for irregular rhythm notification feature. Food and Drug Administration. Aug 8, 2018. URL: https://www.accessdata.fda.gov/cdrh_docs/reviews/DEN180042.pdf [accessed 2023-12-08]
  • Yan L, Zhang H, Goncalves J, Xiao Y, Wang M, Guo Y, et al. An interpretable mortality prediction model for COVID-19 patients. Nat Mach Intell. 2020 May 14;2(5):283-288 [ CrossRef ]
  • The check up with Google Health. Googlez. 2023. URL: https://health.google/the-check-up/#latest-events [accessed 2023-12-08]
  • Liu Y, Jain A, Eng C, Way DH, Lee K, Bui P, et al. A deep learning system for differential diagnosis of skin diseases. Nat Med. 2020 Jun 18;26(6):900-908 [ CrossRef ] [ Medline ]
  • Feathers T. Google's new dermatology app wasn’t designed for people with darker skin. Vice Media Group. May 20, 2021. URL: https:/​/www.​vice.com/​en/​article/​m7evmy/​googles-new-dermatology-app-wasnt-designed-for-people-with-darker-skin [accessed 2023-12-08]
  • Attia ZI, Harmon DM, Dugan J, Manka L, Lopez-Jimenez F, Lerman A, et al. Prospective evaluation of smartwatch-enabled detection of left ventricular dysfunction. Nat Med. 2022 Dec 14;28(12):2497-2503 [ https://europepmc.org/abstract/MED/36376461 ] [ CrossRef ] [ Medline ]
  • Zhu H, Cheng C, Yin H, Li X, Zuo P, Ding J, et al. Automatic multilabel electrocardiogram diagnosis of heart rhythm or conduction abnormalities with deep learning: a cohort study. Lancet Digital Health. 2020 Jul;2(7):e348-e357 [ CrossRef ]
  • Ćirković A. Evaluation of four artificial intelligence-assisted self-diagnosis apps on three diagnoses: two-year follow-up study. J Med Internet Res. 2020 Dec 04;22(12):e18097 [ https://www.jmir.org/2020/12/e18097/ ] [ CrossRef ] [ Medline ]
  • Ameko M, Beltzer M, Cai L, Boukhechba M, Teachman B, Barnes L. Offline contextual multi-armed bandits for mobile health interventions: a case study on emotion regulation. 2020 Presented at: 14th ACM Conference on Recommender Systems; September 22-26, 2020; Virtual p. 249-258 [ CrossRef ]
  • Puntoni S, Reczek RW, Giesler M, Botti S. Consumers and artificial intelligence: an experiential perspective. J Mark. 2020 Oct 16;85(1):131-151 [ CrossRef ]
  • Esmaeilzadeh P. Use of AI-based tools for healthcare purposes: a survey study from consumers' perspectives. BMC Med Inform Decis Mak. 2020 Jul 22;20(1):170 [ https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-020-01191-1 ] [ CrossRef ] [ Medline ]
  • Longoni C, Bonezzi A, Morewedge C. Resistance to medical artificial intelligence. J Consum Res. 2019;46(4):629-650 [ CrossRef ]
  • Scott IA, Carter SM, Coiera E. Exploring stakeholder attitudes towards AI in clinical practice. BMJ Health Care Inform. 2021 Dec 09;28(1):e100450 [ https://informatics.bmj.com/lookup/pmidlookup?view=long&pmid=34887331 ] [ CrossRef ] [ Medline ]
  • Buchanan C, Howitt ML, Wilson R, Booth RG, Risling T, Bamford M. Predicted influences of artificial intelligence on the domains of nursing: scoping review. JMIR Nurs. 2020 Dec 17;3(1):e23939 [ https://nursing.jmir.org/2020/1/e23939/ ] [ CrossRef ] [ Medline ]
  • Garvey KV, Thomas Craig KJ, Russell R, Novak LL, Moore D, Miller BM. Considering clinician competencies for the implementation of artificial intelligence-based tools in health care: findings from a scoping review. JMIR Med Inform. 2022 Nov 16;10(11):e37478 [ https://medinform.jmir.org/2022/11/e37478/ ] [ CrossRef ] [ Medline ]
  • Chew HSJ, Achananuparp P. Perceptions and needs of artificial intelligence in health care to increase adoption: scoping review. J Med Internet Res. 2022 Jan 14;24(1):e32939 [ https://www.jmir.org/2022/1/e32939/ ] [ CrossRef ] [ Medline ]
  • Sharma M, Savage C, Nair M, Larsson I, Svedberg P, Nygren JM. Artificial intelligence applications in health care practice: scoping review. J Med Internet Res. 2022 Oct 05;24(10):e40238 [ https://www.jmir.org/2022/10/e40238/ ] [ CrossRef ] [ Medline ]
  • You Y, Ma R, Gui X. User experience of symptom checkers: a systematic review. 2022 Presented at: AMIA 2022 Annual Symposium; November 5-9, 2022; Washington, DC
  • Parmar P, Ryu J, Pandya S, Sedoc J, Agarwal S. Health-focused conversational agents in person-centered care: a review of apps. NPJ Digit Med. 2022 Feb 17;5(1):21 [ https://doi.org/10.1038/s41746-022-00560-6 ] [ CrossRef ] [ Medline ]
  • Kocaballi AB, Sezgin E, Clark L, Carroll JM, Huang Y, Huh-Yoo J, et al. Design and evaluation challenges of conversational agents in health care and well-being: selective review study. J Med Internet Res. 2022 Nov 15;24(11):e38525 [ https://www.jmir.org/2022/11/e38525/ ] [ CrossRef ] [ Medline ]
  • Arksey H, O'Malley L. Scoping studies: towards a methodological framework. Int J Soc Res Methodol. 2005 Feb;8(1):19-32 [ CrossRef ]
  • Tricco AC, Lillie E, Zarin W, O'Brien KK, Colquhoun H, Levac D, et al. PRISMA Extension for Scoping Reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med. 2018 Oct 02;169(7):467-473 [ https://www.acpjournals.org/doi/abs/10.7326/M18-0850?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub 0pubmed ] [ CrossRef ] [ Medline ]
  • Paez A. Gray literature: an important resource in systematic reviews. J Evid Based Med. 2017 Aug 31;10(3):233-240 [ CrossRef ] [ Medline ]
  • Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006 Jan;3(2):77-101 [ CrossRef ]
  • Braun V, Clarke V. Thematic Analysis: A Practical Guide. Thousand Oaks, CA. SAGE Publications; 2022.
  • Braun V, Clarke V. Thematic analysis. University of Auckland. 2022. URL: https://www.thematicanalysis.net/ [accessed 2023-12-08]
  • Yin J, Ngiam KY, Teo HH. Role of artificial intelligence applications in real-life clinical practice: systematic review. J Med Internet Res. 2021 Apr 22;23(4):e25759 [ https://www.jmir.org/2021/4/e25759/ ] [ CrossRef ] [ Medline ]
  • Crossnohere NL, Elsaid M, Paskett J, Bose-Brill S, Bridges JFP. Guidelines for artificial intelligence in medicine: literature review and content analysis of frameworks. J Med Internet Res. 2022 Aug 25;24(8):e36823 [ https://www.jmir.org/2022/8/e36823/ ] [ CrossRef ] [ Medline ]
  • Beets B, Newman TP, Howell EL, Bao L, Yang S. Surveying public perceptions of artificial intelligence in health care in the United States: systematic review. J Med Internet Res. 2023 Apr 04;25:e40337 [ https://www.jmir.org/2023//e40337/ ] [ CrossRef ] [ Medline ]
  • Demner-Fushman D, Mrabet Y, Ben Abacha A. Consumer health information and question answering: helping consumers find answers to their health-related information needs. J Am Med Inform Assoc. 2020 Feb 01;27(2):194-201 [ https://europepmc.org/abstract/MED/31592532 ] [ CrossRef ] [ Medline ]
  • Oniani D, Wang Y. A qualitative evaluation of language models on automatic question-answering for COVID-19. 2020 Presented at: BCB '20: 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics; September 21-24, 2020; Virtual p. Article 33 [ CrossRef ]
  • Su Z, Figueiredo M, Jo J, Zheng K, Chen Y. Analyzing description, user understanding and expectations of AI in mobile health applications. 2020 Presented at: AMIA 2020 Annual Symposium; November 14-18, 2020; Virtual p. 1170-1179
  • Savery M, Abacha AB, Gayen S, Demner-Fushman D. Question-driven summarization of answers to consumer health questions. Sci Data. 2020 Oct 02;7(1):322 [ https://doi.org/10.1038/s41597-020-00667-z ] [ CrossRef ] [ Medline ]
  • Tsai C, You Y, Gui X, Kou Y, Carroll J. Exploring and promoting diagnostic transparency and explainability in online symptom checkers. 2021 Presented at: 2021 CHI Conference on Human Factors in Computing Systems; May 8-13, 2021; Online p. Article 152 [ CrossRef ]
  • Almalki M. Exploring the influential factors of consumers' willingness toward using COVID-19 Related chatbots: an empirical study. Med Arch (Sarajevo, Bosnia and Herzegovina). 2021 Feb;75(1):50-55 [ https://europepmc.org/abstract/MED/34012200 ] [ CrossRef ] [ Medline ]
  • Gupta P, Suryavanshi A, Maheshwari S, Shukla A, Tiwari R. Human-machine interface system for pre-diagnosis of diseases using machine learning. 2018 Presented at: ICMVA 2018: International Conference on Machine Vision and Applications; April 23-25, 2018; Singapore p. 71-75 [ CrossRef ]
  • He X, Hong Y, Zheng X, Zhang Y. What are the users’ needs? Design of a user-centered explainable artificial intelligence diagnostic system. Int J Hum–Comput Interact. 2022 Jul 26;39(7):1519-1542 [ CrossRef ]
  • Iqbal M, Faiz M. Active surveillance for COVID-19 through artificial intelligence using real-time speech-recognition mobile application. 2020 Presented at: 2020 IEEE International Conference on Consumer Electronics (ICCE); September 28-30, 2020; Taoyuan City, Taiwan [ CrossRef ]
  • Kyung N, Kwon HE. Rationally trust, but emotionally? The roles of cognitive and affective trust in laypeople's acceptance of AI for preventive care operations. Product Oper Manag. 2022 Jul 31:1-20 [ CrossRef ]
  • Park S, Hussain I, Hong S, Kim D, Park H, Benjamin H. Real-time gait monitoring system for consumer stroke prediction service. 2020 Presented at: 2020 IEEE International Conference on Consumer Electronics (ICCE); September 28-30, 2020; Taoyuan City, Taiwan p. 4-6 [ CrossRef ]
  • van Bussel MJP, Odekerken-Schröder GJ, Ou C, Swart RR, Jacobs MJG. Analyzing the determinants to accept a virtual assistant and use cases among cancer patients: a mixed methods study. BMC Health Serv Res. 2022 Jul 09;22(1):890 [ https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-022-08189-7 ] [ CrossRef ] [ Medline ]
  • Nadarzynski T, Miles O, Cowie A, Ridge D. Acceptability of artificial intelligence (AI)-led chatbot services in healthcare: a mixed-methods study. Digit Health. 2019;5:2055207619871808 [ https://journals.sagepub.com/doi/abs/10.1177/2055207619871808?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub 0pubmed ] [ CrossRef ] [ Medline ]
  • Ponomarchuk A, Burenko I, Malkin E, Nazarov I, Kokh V, Avetisian M, et al. Project Achoo: a practical model and application for COVID-19 detection from recordings of breath, voice, and cough. IEEE J Sel Top Signal Process. 2022 Feb;16(2):175-187 [ CrossRef ]
  • Romero HE, Ma N, Brown GJ, Hill EA. Acoustic screening for obstructive sleep apnea in home environments based on deep neural networks. IEEE J Biomed Health Inform. 2022 Jul;26(7):2941-2950 [ CrossRef ]
  • Tschanz M, Dorner TL, Holm J, Denecke K. Using eMMA to manage medication. Computer. 2018 Aug;51(8):18-25 [ CrossRef ]
  • Sellak H, Grobler M. mHealth4U: designing for health and wellbeing self-management. 2021 Presented at: 35th IEEE/ACM International Conference on Automated Software Engineering Workshops; December 21-25, 2020; Virtual p. 41-46 [ CrossRef ]
  • Baldauf M, Fröehlich P, Endl R. Trust me, I’m a doctor – user perceptions of ai-driven apps for mobile health diagnosis. 2020 Presented at: MUM 2020: 19th International Conference on Mobile and Ubiquitous Multimedia; November 22-25, 2020; Essen, Germany p. 167-178 [ CrossRef ]
  • de Carvalho TM, Noels E, Wakkee M, Udrea A, Nijsten T. Development of smartphone apps for skin cancer risk assessment: progress and promise. JMIR Dermatol. 2019 Jul 11;2(1):e13376 [ CrossRef ]
  • Denecke K, Gabarron E, Grainger R, Konstantinidis ST, Lau A, Rivera-Romero O, et al. Artificial intelligence for participatory health: applications, impact, and future implications. Yearb Med Inform. 2019 Aug;28(1):165-173 [ http://www.thieme-connect.com/DOI/DOI?10.1055/s-0039-1677902 ] [ CrossRef ] [ Medline ]
  • Fan X, Chao D, Zhang Z, Wang D, Li X, Tian F. Utilization of self-diagnosis health chatbots in real-world settings: case study. J Med Internet Res. 2021 Jan 06;23(1):e19928 [ https://www.jmir.org/2021/1/e19928/ ] [ CrossRef ] [ Medline ]
  • Koren G, Souroujon D, Shaul R, Bloch A, Leventhal A, Lockett J, et al. “A patient like me” – an algorithm-based program to inform patients on the likely conditions people with symptoms like theirs have. Medicine. 2019;98(42):e17596 [ CrossRef ]
  • Lau AYS, Staccini P, Section Editors for the IMIA Yearbook Section on EducationConsumer Health Informatics. Artificial intelligence in health: new opportunities, challenges, and practical implications. Yearb Med Inform. 2019 Aug;28(1):174-178 [ http://www.thieme-connect.com/DOI/DOI?10.1055/s-0039-1677935 ] [ CrossRef ] [ Medline ]
  • Sangers T, Reeder S, van der Vet S, Jhingoer S, Mooyaart A, Siegel DM, et al. Validation of a market-approved artificial intelligence mobile health app for skin cancer screening: a prospective multicenter diagnostic accuracy study. Dermatology. 2022 Feb 4;238(4):649-656 [ https://doi.org/10.1159/000520474 ] [ CrossRef ] [ Medline ]
  • Sefa-Yeboah SM, Osei Annor K, Koomson VJ, Saalia FK, Steiner-Asiedu M, Mills GA. Development of a mobile application platform for self-management of obesity using artificial intelligence techniques. Int J Telemed Appl. 2021 Aug 27;2021:6624057 [ https://doi.org/10.1155/2021/6624057 ] [ CrossRef ] [ Medline ]
  • da Silva VJ, da Silva Souza V, da Cruz RG, Vidal Martinez de Lucena JM, Jazdi N, de Lucena Junior VF. Commercial devices-based system designed to improve the treatment adherence of hypertensive patients. Sensors (Basel). 2019 Oct 18;19(20):4539 [ https://www.mdpi.com/resolver?pii=s19204539 ] [ CrossRef ] [ Medline ]
  • Zhang Z, Citardi D, Wang D, Genc Y, Shan J, Fan X. Patients' perceptions of using artificial intelligence (AI)-based technology to comprehend radiology imaging data. Health Informatics J. 2021 Apr 29;27(2):14604582211011215 [ https://journals.sagepub.com/doi/abs/10.1177/14604582211011215?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub 0pubmed ] [ CrossRef ] [ Medline ]
  • Zhang Z, Genc Y, Wang D, Ahsen ME, Fan X. Effect of AI explanations on human perceptions of patient-facing AI-powered healthcare systems. J Med Syst. 2021 May 04;45(6):64 [ CrossRef ] [ Medline ]
  • Jaswal G, Bharadwaj R, Tiwari K, Thapar D, Goyal P, Nigam A. AI-biometric-driven smartphone app for strict post-COVID home quarantine management. IEEE Consumer Electron Mag. 2021 May 1;10(3):49-55 [ CrossRef ]
  • Ullrich D, Butz A, Diefenbach S. The development of overtrust: an empirical simulation and psychological analysis in the context of human–robot interaction. Front Robot AI. 2021 Apr 13;8:554578 [ https://europepmc.org/abstract/MED/33928129 ] [ CrossRef ] [ Medline ]
  • Wahl B, Cossy-Gantner A, Germann S, Schwalbe NR. Artificial intelligence (AI) and global health: how can AI contribute to health in resource-poor settings? BMJ Glob Health. 2018;3(4):e000798 [ https://gh.bmj.com/lookup/pmidlookup?view=long&pmid=30233828 ] [ CrossRef ] [ Medline ]
  • Ciecierski-Holmes T, Singh R, Axt M, Brenner S, Barteit S. Artificial intelligence for strengthening healthcare systems in low- and middle-income countries: a systematic scoping review. NPJ Digit Med. 2022 Oct 28;5(1):162 [ https://doi.org/10.1038/s41746-022-00700-y ] [ CrossRef ] [ Medline ]
  • Wang D, Wang L, Zhang Z, Wang D, Zhu H, Gao Y. “Brilliant AI doctor” in rural clinics: challenges in AI-powered clinical decision support system deployment. 2021 Presented at: 2021 CHI Conference on Human Factors in Computing Systems; May 8-13, 2021; Online p. 1-18 [ CrossRef ]
  • Sofaer S. Qualitative methods: what are they and why use them? Health Serv Res. 1999 Dec;34(5 Pt 2):1101-1118 [ https://europepmc.org/abstract/MED/10591275 ] [ Medline ]
  • Queirós A, Faria D, Almeida F. Strengths and limitations of qualitative and quantitative research methods. Eur J Educ Stud. 2017;3(9):369-387 [ CrossRef ]
  • The European definition of general practice/family medicine. WONCA Europe. 2023. URL: http://tinyurl.com/4muu47zc [accessed 2023-12-08]
  • DerSarkissian C. What is a general practitioner? WebMD. 2023. URL: https://www.webmd.com/a-to-z-guides/what-is-a-general-practitioner [accessed 2023-12-08]
  • Simon DA, Evans BJ, Shachar C, Cohen IG. Should Alexa diagnose Alzheimer's?: legal and ethical issues with at-home consumer devices. Cell Rep Med. 2022 Dec 20;3(12):100692 [ https://linkinghub.elsevier.com/retrieve/pii/S2666-3791(22)00228-2 ] [ CrossRef ] [ Medline ]
  • Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019 Jan 7;25(1):44-56 [ CrossRef ] [ Medline ]
  • Castagno S, Khalifa M. Perceptions of artificial intelligence among healthcare staff: a qualitative survey study. Front Artif Intell. 2020 Oct 21;3:578983 [ https://europepmc.org/abstract/MED/33733219 ] [ CrossRef ] [ Medline ]
  • Price WN, Cohen IG. Privacy in the age of medical big data. Nat Med. 2019 Jan 7;25(1):37-43 [ https://europepmc.org/abstract/MED/30617331 ] [ CrossRef ] [ Medline ]
  • Munjal K, Bhatia R. A systematic review of homomorphic encryption and its contributions in healthcare industry. Complex Intell Syst. 2022 May 03;9(4):1-28 [ https://europepmc.org/abstract/MED/35531323 ] [ CrossRef ] [ Medline ]
  • Kairouz P, Oh S, Viswanath P. Secure multi-party differential privacy. Adv NeurIPS. 2015;28:1-9
  • Yuan Y, Liu J, Jin D, Yue Z, Yang T, Chen R, et al. DeceFL: a principled fully decentralized federated learning framework. Natl Sci Open. 2023 Jan 10;2(1):20220043 [ CrossRef ]
  • Schneider G. Digital health research and health data pools. In: Health Data Pools Under European Data Protection and Competition Law: Health as a Digital Business. Switzerland. Springer Nature; 2022:7-60
  • Delacroix S, Lawrence N. Bottom-up data trusts: disturbing the ‘one size fits all’approach to data governance. Int Data Privacy Law. 2019;9(4):236-252 [ CrossRef ]
  • Luengo-Oroz M, Hoffmann Pham K, Bullock J, Kirkpatrick R, Luccioni A, Rubel S, et al. Artificial intelligence cooperation to support the global response to COVID-19. Nat Mach Intell. 2020 May 22;2(6):295-297 [ CrossRef ]
  • Choudhury A, Asan O. Impact of accountability, training, and human factors on the use of artificial intelligence in healthcare: exploring the perceptions of healthcare practitioners in the US. Hum Factors Healthc. 2022 Dec;2:100021 [ CrossRef ]
  • Smith H. Clinical AI: opacity, accountability, responsibility and liability. AI Soc. 2020 Jul 25;36(2):535-545 [ CrossRef ]
  • Habli I, Lawton T, Porter Z. Artificial intelligence in health care: accountability and safety. Bull World Health Organ. 2020 Feb 25;98(4):251-256 [ CrossRef ]
  • World Health Organization. Ethics and Governance of Artificial Intelligence for Health: WHO Guidance. Geneva. World Health Organization; 2021.
  • General Data Protection Regulation (GDPR). Intersoft Consulting. 2018. URL: https://gdpr-info.eu/ [accessed 2023-12-08]
  • Classify your medical device. Food and Drug Administration. 2020. URL: http://tinyurl.com/2n9ta6uy [accessed 2023-12-08]
  • Guidance document: software as a medical device (SaMD): definition and classification. Government of Canada. 2019. URL: http://tinyurl.com/4sc5wdkd [accessed 2023-12-08]
  • Madiega T. Artificial Intelligence Act. European Parliamentary Research Service. 2021. URL: https://www.europarl.europa.eu/RegData/etudes/BRIE/2021/698792/EPRS_BRI(2021)698792_EN.pdf [accessed 2023-12-08]
  • Helberger N, Diakopoulos N. ChatGPT and the AI Act. Internet Policy Rev. 2023;12(1):1-6 [ CrossRef ]
  • de Miguel I, Sanz B, Lazcoz G. Machine learning in the EU health care context: exploring the ethical, legal and social issues. Inf Commun Soc. 2020 Jul 13;23(8):1139-1153 [ CrossRef ]
  • Gunning D. Explainable artificial intelligence (XAI). Defense Advanced Research Projects Agency (DARPA). 2017. URL: https://asd.gsfc.nasa.gov/conferences/ai/program/003-XAIforNASA.pdf [accessed 2023-12-08]
  • Adadi A, Berrada M. Peeking inside the black-box: a survey on explainable artificial intelligence (XAI). IEEE Access. 2018;6:52138-52160 [ CrossRef ]
  • Barredo Arrieta A, Díaz-Rodríguez N, Del Ser J, Bennetot A, Tabik S, Barbado A, et al. Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf Fusion. 2020 Jun;58:82-115 [ CrossRef ]
  • Loh HW, Ooi CP, Seoni S, Barua PD, Molinari F, Acharya UR. Application of explainable artificial intelligence for healthcare: a systematic review of the last decade (2011-2022). Comput Methods Programs Biomed. 2022 Nov;226:107161 [ CrossRef ] [ Medline ]
  • Ali S, Abuhmed T, El-Sappagh S, Muhammad K, Alonso-Moral JM, Confalonieri R, et al. Explainable artificial intelligence (XAI): what we know and what is left to attain trustworthy artificial intelligence. Inf Fusion. 2023 Nov;99:101805 [ CrossRef ]
  • Guidotti R, Monreale A, Ruggieri S, Turini F, Giannotti F, Pedreschi D. A survey of methods for explaining black box models. ACM Comput Surv. 2018 Aug 22;51(5):1-42 [ CrossRef ]
  • Schoonderwoerd TA, Jorritsma W, Neerincx MA, van den Bosch K. Human-centered XAI: developing design patterns for explanations of clinical decision support systems. Int J Hum-Comput Stud. 2021 Oct;154:102684 [ CrossRef ]
  • Ehsan U, Liao Q, Muller M, Riedl M, Weisz J. Towards social transparency in ai systems. 2021 Presented at: 2021 CHI Conference on Human Factors in Computing Systems; May 8-13, 2021; Online [ CrossRef ]
  • Lee JD, See KA. Trust in automation: Designing for appropriate reliance. Hum Factors. 2004;46(1):50-80 [ CrossRef ]
  • Wischnewski M, Krämer N, Müller E. Measuring and understanding trust calibrations for automated systems: a survey of the state-of-the-art and future directions. 2023 Presented at: 2023 CHI Conference on Human Factors in Computing Systems; April 23-28, 2023; Hamburg, Germany [ CrossRef ]
  • Arnold M, Bellamy RKE, Hind M, Houde S, Mehta S, Mojsilovic A, et al. FactSheets: increasing trust in AI services through supplier's declarations of conformity. IBM J Res Dev. 2019 Jul 1;63(4/5):6:1-6:13 [ CrossRef ]
  • Bedué P, Fritzsche A. Can we trust AI? An empirical investigation of trust requirements and guide to successful AI adoption. J Enterp Inf Manag. 2021 Apr 30;35(2):530-549 [ CrossRef ]
  • Gillath O, Ai T, Branicky MS, Keshmiri S, Davison RB, Spaulding R. Attachment and trust in artificial intelligence. Comput Hum Behav. 2021 Feb;115:106607 [ CrossRef ]
  • Alboqami H. Trust me, I'm an influencer! - causal recipes for customer trust in artificial intelligence influencers in the retail industry. J Retail Consum Serv. 2023 May;72:103242 [ CrossRef ]
  • Logg JM, Minson JA, Moore DA. Algorithm appreciation: people prefer algorithmic to human judgment. Organ Behav Hum Decis Process. 2019 Mar;151:90-103 [ CrossRef ]
  • Novemsky N, Kahneman D. The boundaries of loss aversion. J Mark Res. 2018 Oct 10;42(2):119-128 [ CrossRef ]
  • Fan W, Liu J, Zhu S, Pardalos PM. Investigating the impacting factors for the healthcare professionals to adopt artificial intelligence-based medical diagnosis support system (AIMDSS). Ann Oper Res. 2018 Mar 19;294(1-2):567-592 [ CrossRef ]
  • Cabitza F, Campagner A, Ronzio L, Cameli M, Mandoli GE, Pastore MC, et al. Rams, hounds and white boxes: investigating human-AI collaboration protocols in medical diagnosis. Artif Intell Med. 2023 Apr;138:102506 [ https://linkinghub.elsevier.com/retrieve/pii/S0933-3657(23)00020-9 ] [ CrossRef ] [ Medline ]
  • Boillat T, Nawaz FA, Rivas H. Readiness to embrace artificial intelligence among medical doctors and students: questionnaire-based study. JMIR Med Educ. 2022 Apr 12;8(2):e34973 [ https://mededu.jmir.org/2022/2/e34973/ ] [ CrossRef ] [ Medline ]
  • Bellet PS. The importance of empathy as an interviewing skill in medicine. JAMA. 1991 Oct 02;266(13):1831 [ CrossRef ]
  • Compassion in practice: evidencing the impact. NHS England. May 2016. URL: https://www.england.nhs.uk/wp-content/uploads/2016/05/cip-yr-3.pdf [accessed 2023-12-08]
  • Spiro H. Commentary: the practice of empathy. Acad Med. 2009 Sep;84(9):1177-1179 [ CrossRef ] [ Medline ]
  • Tweedie J, Hordern J, Dacre J. Advancing Medical Professionalism. London, UK. Royal College of Physicians; 2018.
  • Decety J. Empathy in medicine: what it is, and how much we really need it. Am J Med. 2020 May;133(5):561-566 [ CrossRef ] [ Medline ]
  • Kerasidou A. Artificial intelligence and the ongoing need for empathy, compassion and trust in healthcare. Bull World Health Organ. 2020 Apr 01;98(4):245-250 [ https://europepmc.org/abstract/MED/32284647 ] [ CrossRef ] [ Medline ]
  • Pepito JA, Ito H, Betriana F, Tanioka T, Locsin RC. Intelligent humanoid robots expressing artificial humanlike empathy in nursing situations. Nurs Philos. 2020 Oct 20;21(4):e12318 [ CrossRef ] [ Medline ]
  • Montemayor C, Halpern J, Fairweather A. In principle obstacles for empathic AI: why we can't replace human empathy in healthcare. AI Soc. 2022 May 26;37(4):1353-1359 [ https://europepmc.org/abstract/MED/34054228 ] [ CrossRef ] [ Medline ]
  • Morrow E, Zidaru T, Ross F, Mason C, Patel KD, Ream M, et al. Artificial intelligence technologies and compassion in healthcare: a systematic scoping review. Front Psychol. 2022;13:971044 [ https://europepmc.org/abstract/MED/36733854 ] [ CrossRef ] [ Medline ]
  • Ayers JW, Poliak A, Dredze M, Leas EC, Zhu Z, Kelley JB, et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Intern Med. 2023 Jun 01;183(6):589-596 [ CrossRef ] [ Medline ]
  • Wexler A, Reiner PB. Oversight of direct-to-consumer neurotechnologies. Science. 2019 Jan 18;363(6424):234-235 [ https://europepmc.org/abstract/MED/30655433 ] [ CrossRef ] [ Medline ]
  • De Zambotti M, Cellini N, Goldstone A, Colrain I, Baker F. Wearable sleep technology in clinical and research settings. Med Sci Sports exerc. 2019;51(7):1538 [ CrossRef ]
  • Ates HC, Nguyen PQ, Gonzalez-Macia L, Morales-Narváez E, Güder F, Collins JJ, et al. End-to-end design of wearable sensors. Nat Rev Mater. 2022 Jul 22;7(11):887-907 [ https://europepmc.org/abstract/MED/35910814 ] [ CrossRef ] [ Medline ]
  • Ryu WM, Lee Y, Son Y, Park G, Park S. Thermally drawn multi-material fibers based on polymer nanocomposite for continuous temperature sensing. Adv Fiber Mater. 2023 Jun 12;5(5):1712-1724 [ CrossRef ]
  • Ran S, Yang X, Liu M, Zhang Y, Cheng C, Zhu H, et al. Homecare-oriented ECG diagnosis with large-scale deep neural network for continuous monitoring on embedded devices. IEEE Trans Instrum Meas. 2022;71:1-13 [ CrossRef ]
  • Edgren L. Health consumer diversity and its implications. J Syst Sci Syst Eng. 2006 Mar;15(1):34-47 [ CrossRef ]
  • Willemink MJ, Koszek WA, Hardell C, Wu J, Fleischmann D, Harvey H, et al. Preparing medical imaging data for machine learning. Radiology. 2020 Apr;295(1):4-15 [ https://europepmc.org/abstract/MED/32068507 ] [ CrossRef ] [ Medline ]


Edited by A Mavragani; submitted 01.07.23; peer-reviewed by Z Zhang, A Nagappan, L Weinert; comments to author 19.07.23; revised version received 20.09.23; accepted 28.11.23; published 18.12.23

©Xin He, Xi Zheng, Huiyuan Ding. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 18.12.2023.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.

Our websites may use cookies to personalize and enhance your experience. By continuing without changing your cookie settings, you agree to this collection. For more information, please see our University Websites Privacy Notice .

UConn Today

  • School and College News
  • Arts & Culture
  • Community Impact
  • Entrepreneurship
  • Health & Well-Being
  • Research & Discovery
  • UConn Health
  • University Life
  • UConn Voices
  • University News

December 19, 2023 | Danielle Faipler

“They Will Come at Me”: New Study Investigates Fear of Retaliation in America’s Nursing Homes

While highly prevalent and pervasive, the fear of retaliation has largely been overlooked in policy and research. A new study seeks to improve understanding of this phenomenon

A woman, seen from behind, stares out the window of her room in a nursing home.

(Adobe Stock)

“I try to stay isolated because of disrespectful treatment by staff. I am nervous now that I said something. They will come at me,” said one anonymous resident of a nursing home, describing how expressing concerns about the care they receive could lead to abuse, neglect, punishment, and other forms of retaliation.

Despite federal and state laws protecting residents’ rights to voice grievances, this fear of retaliation scares them into silence and has emotional, psychological, and physical consequences. While prevalent , the fear of retaliation has largely been overlooked in policy and research.

Eilon Caspi , a gerontologist and assistant research professor at UConn’s Institute for Collaboration on Health, Intervention, and Policy ( InCHIP ), has a new study published in the Journal of Applied Gerontology that aims to improve understanding of this phenomenon. The study analyzes 100 standard survey and complaint investigation reports from state survey agencies in nursing homes across 30 states.

Image of Eilon Caspi, PhD

The fear of retaliation describes a feeling of vulnerability that one’s actions may cause retaliation by another. Retaliation refers to the act of revenge by facility staff in response to a complaint.

Caspi’s study represents the most in-depth analysis of residents’ fear of retaliation in U.S. nursing homes to date. The study seeks to improve understanding of residents’ lived experiences of four aspects of this phenomenon: fear of retaliation, allegations of threats of retaliation, perceived retaliation, and confirmed retaliation. It also illuminates the emotional consequences of this phenomenon on residents.

“Given that research on this phenomenon is in its infancy, the findings can serve as the basis for development of questions for future research,” says Caspi.

According to the National Long-Term Care Ombudsman Resource Center , “Fear of retaliation is one of the most common reasons residents do not want to pursue a complaint and disclose their identity. Since residents live in a facility and rely on staff for their basic needs, their fear of retaliation cannot be overemphasized.”

Caspi found that various forms of retaliation and fear of retaliation resulted in suffering and harm to residents and were poorly addressed. When voicing care concerns or deciding to file complaints relating to rights violations and mistreatment, residents expressed fear of eviction, collective punishment, physical violence, delayed care, and aggressive confrontations, among other forms.

Part of the driving factor of this experience of fear of retaliation is the power imbalance that exists between staff in nursing homes and vulnerable residents. This imbalance may result in residents not being able to draw attention to abuse or staff not reporting abuse to protect themselves or their colleagues. The study also found that because residents worried about retaliation, they did not report mistreatment, and when it was reported, investigations were often delayed.

“Although additional research is needed to shed light on it, the study showed that many residents’ experiences of fear of or actual retaliation could be characterized as ‘learned helplessness,’ essentially giving up trying to voice care concerns as a result of repeated failure to bring about change,” says Caspi.

This study has a range of implications for policy and practice, namely education, oversight, enforcement of federal rights, and national and state awareness campaigns, which could encourage further discourse about and meaningful action to prevent and reduce the fear of retaliation.

Educational programs would help empower residents and families to resolve issues of neglect, fear of retaliation and retaliation itself, and provide clear guidance about what constitutes as or is perceived as retaliation against residents.

In Connecticut , nursing home administrators are mandated under state law to provide staff members with annual training on fear of retaliation, which encompasses residents’ rights to file complaints, examples of employee retaliation, and strategies to prevent staff retaliation against residents. Connecticut is the only state to require such training. That said, the law in Connecticut needs to be expanded to include the assisted living sector.

Given that the fear of retaliation has been shown in an early study led by UConn Center on Aging professor Julie Robison to be a reality in all types of long-term care homes, “all states should consider enacting similar laws while ensuring that it will be applicable to assisted living residences as well as nursing homes,” says Caspi.

Beyond education for residents, their families, and staff, Caspi recommends strengthening federal oversight, protections, deterrence, and enforcement related to fear of retaliation. The Centers for Medicare & Medicaid Services (CMS) does not centrally track this phenomenon in over 15,000 nursing homes nationwide. In addition, the National Ombudsman Reporting System currently only tracks complaints of “retaliation” and not residents’ fear of retaliation.

“Left untracked, the phenomenon remains invisible, important opportunities for learning and prevention are missed, and vulnerable residents remain silenced,” says Caspi.

Recent Articles

nursing research articles on incontinence

December 19, 2023

What You Can Do Now to Prevent Memory Loss

Read the article

Group of black women in exercise clothes smiling

Black Girls Run! Community Organization Creates Inclusive Space for Physical Activity

nursing research articles on incontinence

December 18, 2023

Diane Burgess Wins the Edward C. Marth Mentorship Award


  1. Urinary incontinence in resident women in nursing homes

    nursing research articles on incontinence

  2. Managing Urinary Incontinence : AJN The American Journal of Nursing

    nursing research articles on incontinence

  3. (PDF) Urinary Incontinence in the Elderly Part 3 of a Series of

    nursing research articles on incontinence

  4. Prompted Voiding Protocol for Individuals With Urinary Incontinence

    nursing research articles on incontinence

  5. Managing Incontinence Using Technology, Devices, and Products

    nursing research articles on incontinence

  6. (PDF) Diagnosis and Non-Surgical Management of Urinary Incontinence

    nursing research articles on incontinence


  1. Monroeville-Based Company Develops Product For Incontinence

  2. Incontinence Of Urine, Nursing Foundation,Gnm,Bsc(1st year)

  3. A nursing perspective: our Biomedical Research Centre

  4. Resident Care Procedure #53: Application of Incontinent Brief

  5. ND

  6. Nursing Care Plan for Urinary incontinence/Bladder incontinence/Involuntary urination/urine leakage


  1. Managing Urinary Incontinence : AJN The American Journal of Nursing

    Abstract In Brief This article is part of a series, Supporting Family Caregivers: No Longer Home Alone, published in collaboration with the AARP Public Policy Institute.

  2. Urinary incontinence and quality of life: a systematic review and meta

    Urinary incontinence (UI) and low quality of life (QoL) are two common conditions. Some recent literature proposed that these two entities can be associated. However, no attempt was made to collate this literature.

  3. Urinary Incontinence in Older Adults : AJN The American Journal of Nursing

    Incontinence related to chronic retention of urine (or overflow incontinence) refers to urinary leakage resulting from an overfilled and distended bladder. 13 This type of UI is frequently caused by bladder outlet obstruction, making it more prevalent among men who have hyperplasia of the prostate. 23

  4. Urinary Incontinence in Older Adults Takes Collaborative Nursing

    Urinary incontinence (UI) is defined as the complaint of any involuntary leakage of urine [ 1 ]. UI is a common problem in patients, especially those who are elderly, mostly above age 65, and living in nursing homes. The literature suggests that more than 70% of nursing home residents in the United States experience UI [ 2 ].

  5. Urinary Incontinence

    Urinary incontinence is the involuntary leakage of urine. [1] This medical condition is common in the elderly, especially in nursing homes, but it can affect younger adult males and females as well. Urinary incontinence can impact both patient health and quality of life.

  6. Diagnosis and Non-Surgical Management of Urinary Incontinence

    Urinary incontinence (UI) is a bothersome symptom with population studies suggesting a prevalence of 13.1% in women and 5.4% in men. While a significant cohort of patients with this complaint may ultimately require surgical management to achieve complete continence, a number of non-surgical measures exist to improve symptoms and quality of life.

  7. Nursing assessment and management of incontinence among ...

    The results demonstrate that nursing education and formalized assessment pathways in an acute setting can improve nursing compliance with the assessment and management of patients with either urinary or fecal incontinence to ensure safe, compassionate and person-centered care.

  8. The Effects of Education About Urinary Incontinence on Nurse

    The aims of this systematic review were to describe, critique, and summarize research about the effects of education about urinary incontinence on nurses' and nursing assistants' knowledge and attitudes toward urinary incontinence, their continence care practices, and patient outcomes. We searched key electronic databases (PsycINFO, MEDLINE, CINAHL, Web of Science, and Cochrane Library) for ...

  9. Prevalence and factors related to urinary incontinence in older adults

    The World Health Organization (WHO) finds citizens 65 years of age to be older adults and the United Nations deems people with 60 years or above to be older adults [1, 2].The world's population is aging rapidly, with 703 million people now over the age of 65, and this number is projected to reach 1.5 billion by 2050 [].Urinary incontinence is a common condition in the general population ...

  10. Understanding Nursing Knowledge of Continence Care and ...

    Nursing care staff's negative attitudes and lack of basic understanding of incontinence also limit engagement in continence care, with many reporting that incontinence is a normal, expected consequence of aging which could lead nursing staff to be less likely to intervene to manage UI (Dingwall, 2008; Ehlman et al., 2012; Terzoni et al., 2011 ...

  11. Urinary incontinence in women

    Research has revealed overlapping potential causes of incontinence, including dysfunction of the detrusor muscle or muscles of the pelvic floor, dysfunction of the neural controls of storage and voiding, and perturbation of the local environment within the bladder.

  12. Understanding the Impact of Urinary Incontinence in Persons with

    More than half of the carers ( n = 46, 65%) were concerned incontinence may be a principal reason for future nursing home admission. Conclusion UI can be distressing for persons with dementia. Care partners were concerned about loss of independence and early nursing home admission. Awareness of bladder and bowel services should be increased.

  13. Evaluation and current treatments for urinary incontinence

    Evaluation and current treatments for urinary incontinence. Evaluation and current treatments for. urinary incontinence. Hooper, Gwendolyn L. PhD, APRN, CUNP. Author Information. The Nurse Practitioner: June 2019 - Volume 44 - Issue 6 - p 21-28. doi: 10.1097/01.NPR.0000558153.53725.02.

  14. An evidence-based self-management package for urinary incontinence in

    6 Citations 4 Altmetric Metrics Abstract Background Urinary incontinence (UI) is a distressing condition that limits women's quality of life and places a heavy burden on health care services. Behavioural treatments are recommended as a first-line treatment.

  15. Incidence and Predictors of Incontinence Associated Skin Damage in

    In a cross-sectional study of nursing home residents, Bliss and colleagues 13 reported that skin damage in the perineal area was associated not only with incontinence (mainly fecal and dual incontinence) but with functional, physical, cognitive, and mobility problems of residents.

  16. Nursing Management of Urinary Incontinence in Geriatric Inpatients

    The management and treatment of incontinence represent two interrelated areas — providing enough providers of care and describing the types of management and treatment options available to the incontinent patient. This article has stressed that adequate staffing underlies any successful program of continence management and treatment. In addition, many options for the management and treatment ...

  17. Evidence-Based Continence Care: An Integrative Review

    Design: The newly designed 2015 Competency Model for Professional Rehabilitation Nursing serves as a conceptual framework to categorize evidence-based recommendations for continence care into the four domains of the model. Methods: A search of the evidence was completed in December 2015. Literature reviewed was limited to articles published ...

  18. Nurses' Perceptions and Management of Urinary Incontinence in

    Urinary incontinence (UI) is a geriatric syndrome affecting older adults in long-term and acute hospital care. Most research has focused on nurses' perceptions of UI in long-term care and...

  19. Frontiers

    Urinary incontinence (UI) is a common problem among older adults. This study investigated the prevalence of UI in nursing home residents aged ≥75 years in China and examined potential risk factors associated with UI and its subtypes.

  20. Nursing research and continence care

    Despite 50 years of attention to continence care and pivotal nursing research in the field, incontinence continues to challenge health care professionals and consumers alike. The role of nursing research in continence is traced from the early influences of Florence Nightingale to today's clinical practice guidelines and the influence of major ...

  21. Directions for Future Nursing Research on Fecal Incontinence

    Background: As knowledge of the prevalence and impact of fecal incontinence increases, additional research is needed to improve patient outcomes and support the practice of nurses in this area. Objectives: To outline needs for future nursing research on fecal incontinence in aging adults.

  22. Managing Incontinence Using Technology, Devices, and Products

    However, research on their clinical uses, problems requiring nursing care, and patient satisfaction is lacking. Objectives: To review the various products and devices used for incontinence, identify directions for research and development on technology, and outline the ways nurses can influence and participate in those investigations.

  23. Patient-Centered Interventions: Implications for Incontinence

    Background: Nurse researchers in incontinence have focused on testing the effects of standardized interventions; however, nurses in practice usually customize interventions with patients. Patient-centered interventions promise to bring research and practice closer together.Tailored interventions, one kind of patient-centered intervention, have ...

  24. Thank You for Your Service to Science, Your Colleagues, and

    On behalf of myself, the Associate Editors of Clinical Nursing Research (CNR), and the CNR Editorial Board, I offer my deepest gratitude to colleagues who served (and continue to serve) as peer reviewers for the Journal. The quality of research published in CNR hinges on the quality of peer review. The Journal is considered "highly selective," with an acceptance rate of 7% to 10%.

  25. Solutions to close the nursing shortage gap

    US healthcare organizations continue to grapple with the impacts of the nursing shortage—scaling back of health services, increasing staff burnout and mental-health challenges, and rising labor costs. While several health systems have had some success in rebuilding their nursing workforces in recent months, estimates still suggest a potential shortage of 200,000 to 450,000 nurses in the ...

  26. Journal of Medical Internet Research

    Background: Direct-to-consumer (DTC) health care artificial intelligence (AI) apps hold the potential to bridge the spatial and temporal disparities in health care resources, but they also come with individual and societal risks due to AI errors. Furthermore, the manner in which consumers interact directly with health care AI is reshaping traditional physician-patient relationships.

  27. "They Will Come at Me": New Study Investigates Fear of Retaliation in

    "Given that research on this phenomenon is in its infancy, the findings can serve as the basis for development of questions for future research," says Caspi. According to the National Long-Term Care Ombudsman Resource Center , "Fear of retaliation is one of the most common reasons residents do not want to pursue a complaint and disclose ...

  28. Exploring parental mental health and children's anxiety

    As a leading international medical research university with the largest medical faculty in Australia and integration with leading Australian teaching hospitals, we consistently rank in the top 50 universities worldwide for clinical, pre-clinical and health sciences. For more news, visit Medicine, Nursing and Health Sciences or Monash University.