Despite the prevalence of urinary incontinence (UI) being higher in older than younger men, most epidemiological studies have not systematically identified or categorized UI risk factors in older males. Considering the cost implications of UI, an understanding of risk factors can inform cost-effective prevention and treatment programs. This scoping review aimed to identify and categorize risk factors for UI in older men and identify gaps in the evidence for the first time.

The Joanna Briggs Institute (JBI) method for scoping reviews (1) guided the conduct of this scoping review and its reporting alongside the Preferred Reporting Items for Systematic Reviews and Meta-analyses extension for scoping reviews (PRISMA-ScR) checklist (2). JBI's Population, Concept, and Context (PCC) framework was used to frame inclusion criteria. These were all sources of evidence on UI risk factors that included older men (65+). Specifically, all data on UI risk factors stratified by age and sex, combined UI risk factors dataset of 65+ males and females stratified by sex, male UI risk factors stratified by age, and data on UI risk factors in men 65+ only.
We employed JBI’s three-step search strategy, which included a limited initial search in Ovid MEDLINE, a detailed search in all included databases, and a search of reference lists of included studies. Our comprehensive search included Ovid MEDLINE, Ovid Embase, CINAHL, Scopus, Web of Science Core Collection, Cochrane Library (via Wiley), and ProQuest Dissertations & Theses Global. Study type or publication date was not restricted, and there was no language restriction since approximately 10% of preliminary search results were in languages other than English (LOTE). For LOTE, we compared translations from two validated online language translators; DeepL translator and Google translator (https://www.deepl.com/en/translator.and https://translate.google.com/). Besides databases, we reviewed Google Scholar results, bibliographies of included studies and grey literature. Two independent reviewers screened, selected, and extracted eligible studies. 
Using a predetermined framework, we extracted and analyzed data deductively. The data were analyzed qualitatively and quantitatively, using qualitative content analysis and descriptive statistics respectively. A protocol detailing the methods was developed and published (3), and is registered in the Open Science Framework [Feb 07 2023; https://osf.io/xsrge/].

Forty-seven articles that met the inclusion criteria identified 98 risk factors across six categories. 
A total of four behavioural risk factors, reported by only two studies, were the least investigated of all the categories examined by these evidence sources, whereas 34 medical factors/diseases (with 111 frequency counts) were reported from 39 articles (83% of evidence sources). A total of 34 risk factors belonging to the other factors category were identified from 29 studies (62%) and were mostly medically-related entities that were not disease diagnoses. Nine physiological risk factors/age-related physiological changes were found in five studies (11%) with a frequency of 13. Four demographic factors with a frequency count of 15 were found in 14 studies (30%) and 13 environmental factors with a frequency of 18 were reported in eight studies (17%). Genetic factors were not documented in the included sources. Figure 1 shows frequency counts across categories. 
The top five risk factors were increasing age/advanced age (n=12), Benign Prostatic Hyperplasia (n=11), Diabetes Mellitus (n=11), Detrusor Overactivity (n=10), limitation in physical function/ADL disability (n=10), increased Body Mass Index (BMI)/overweight/obesity (n=8), Dementia (n=8), and Parkinson's disease (n=7).
Figure 2 shows that primary evidence sources are sparsely distributed across 12 of 195 countries. North America contributed almost half of the studies (n=17, 44.8% [n=2, 5.3% from Canada and n=15, 39.5% from the United States]). Ten articles (26.3%) came from four Asian countries (China, Japan, Taiwan, and Singapore), and 10 articles (26.3%) came from six European countries (Austria, Finland, Italy, Romania, Spain, and the United Kingdom). None were from African and South American countries.
Almost half (n=21, 45%) of the evidence sources were community-based studies, while eight (17%) and six (13%) provided samples from multiple settings and tertiary care facilities. Evidence from primary care settings contributed the least (n=1, 2%) to this study. Among the 491 eligible articles, 331 (67.4%) were excluded due to the lack of stratification of UI risk factors by age, sex, or age and sex, making them ineligible.

The majority of studies included in the review focused on medical risk factors, with scant attention given to behavioural risk factors. Increasing age/advanced age, the top risk factor for UI in older men, belongs to the demographic factors category. 
The primary evidence sources originated from only 12 (6%) of the countries in the world, making generalizations difficult. In a significant proportion of excluded studies, age-sex stratification of UI risk factors was not included, making it impossible to extract UI risk factors for older men.

More primary research on behavioral risk factors for UI in older men is necessary due to the lack of evidence on the topic. These factors may play a role in health promotion and disease prevention in this area. For international data on UI to be more inclusive, it is imperative that more UI research be conducted around the world, particularly in areas where there are no existing data on the topic. In addition, more research should be encouraged in primary care, which is the first point of care for the vast majority of patients. Future studies should also consider sex and age-stratification of UI risk factors, as well as other concepts of interest, to produce more meaningful data.

JBI Manual for Evidence Synthesis [Internet]. JBI; 2020 [cited 2022 Sep 27]. Available from: https://wiki.jbi.global/display/MANUAL/Chapter+11%3A+Scoping+reviews
Ann Intern Med. 2018 Oct 2;169(7):467–73.
BMJ Open. 2023 Feb 10;13(2):e068956.

Continence 7S1 (2023) 101010
DOI: 10.1016/j.cont.2023.101010