Urinary Incontinence of Women in a Nationwide Study in Sri Lanka: Prevalence and Risk Factors FEMALE UROLOGY Ramya Pathiraja1, Shamini Prathapan2*, Sampatha Goonawardena2 Purpose: Urinary incontinence, be stress incontinence or urge incontinence or a mixed type incontinence affects women of all ages. The aim of this study was to describe the prevalence and risk factors of urinary incontinence in Sri Lanka. Materials and Methods: A community based cross-sectional study was performed in Sri Lanka. The age group of the women in Sri Lanka was categorized into 3 age groups: Less than or equal to 35 years, 36 to 50 years of age and more than or equal to 51 years of age. A sample size of 675 women was obtained from each age category obtaining a total sample of 2025 from Sri Lanka. An interviewer-administered questionnaire consisting of two parts; Socio demographic factors, Medical and Obstetric History, and the King’s Health Questionnaire (KHQ), was used for data collection. Stepwise logistic regression analysis was performed. Results: The Prevalence of women with only stress incontinence was 10%, with urge incontinence was 15.6% and with stress and urge incontinence was 29.9%. Stepwise logistic regression analysis showed that the age groups of 36 – 50 years (OR = 2.03; 95% CI = 1.56 - 2.63) and 51 years and above (OR = 2.61; 95% CI= 1.95 – 3.48), Living in one of the districts in Sri Lanka (OR = 4.58; 95% CI = 3.35 – 6.27) and having given birth to multiple children (OR = 1.1; 95% CI = 1.02 – 1.21), diabetes mellitus (OR = 1.97; 95% CI = 1.19 - 3.23), and respiratory diseases (OR = 2.17; 95% CI = 1.48 - 3.19 ) showed a significant risk in the regression analysis. Conclusion: The risk factor, mostly modifiable, if prevented early, could help to reduce the symptoms of urinary incontinence. Keywords: Prevalence; Sri Lanka; Urinary incontinence. INTRODUCTION Urinary incontinence (UI) is defined as the invol-untary loss of urine, which is a social or hygienic problem and is objectively demonstrable(1). UI could be stress urinary incontinence (SUI), urge urinary inconti- nence (UUI) or mixed urinary incontinence (MUI). It is difficult to make the correct distinction between these three types of UI based on just questionnaires alone and urodynamic tests are often considered gold standard(2). UI affects people of all ages. The exact prevalence of UI is uncertain but it varies from 10 to 60% in Europe and limited data are available for women in Asian coun- tries (3,4,5). These variations are due to lack of standardi- zation of definitions, age of target population, and var- iable study designs(6,7) . The prevalence of UI increases with age, with 20.0–30.0% prevalence in young adults, a peak around middle age (prevalence 30.0–40.0%) and a steady increase in old age (prevalence 30.0–50.0%) (8). Women are much more susceptible to UI than men. Re- productive and hormonal changes associated with preg- nancy and menopause explains the differences prevail- ing between male and female(9,10,11). It is highly probable that socio-economic and cultural factors play a crucial role in UI. However the extent of the influence of these factors on women's health remains relatively unknown. In Sri Lanka, inadequate epidemiological data on the 1Department of Obstetrics and Gynecology, Faculty of Medical Sciences, University of Sri Jayewardenepura. 2Department of Community Medicine, Faculty of Medical Sciences, University of Sri Jayewardenepura. *Correspondence: Department of Community Medicine, Faculty of Medical Sciences, University of Sri Jayewardenepura. E-mail: shamini@sjp.ac.lk. Received October 2016 & Accepted April 2017 prevalence of female UI led us to conduct this study. The aims of this community-based cross-sectional study were to investigate the prevalence and associated risk factors of UI in women in three districts of Sri Lan- ka and also to increase the awareness of both health care providers and the community about this health issue. In addition, one of our aims was to refer women with incommodious symptoms to specialized health care centers for further management. MATERIALS AND METHODS This study was a community based cross-sectional study conducted in the years 2015 - 2016. Out of the 24 dis- tricts in Sri Lanka, three districts; Colombo, Ratnapura and Trincomalee, were randomly selected for this study as the women in these three districts would represent the Sri Lankan female population. The initial hypothe- sis was that through this population we could access a very large part of the female population of our country. We believe that the distinctive characteristics of these three districts allow us to a very considerable extent, to extrapolate the results of our sample to the majority of the general female population. The inclusion criteria was women in the age group of 18 and above, living in one of the above three districts for more than three years. Women with any surgeries that were related to Female Urology 3075 The research was conducted with the permission of Deputy Provincial Director of Health Services of the three districts and from the relevant MOH’s of the dis- tricts. Chi Square test and odds ratios were used to describe the risks and associated factors. Test of proportions and the prevalence were calculated for each district and for the three age groups. Those risk factors with a p val- ue less than 0.05 in bivariate analysis were selected to enter into the multivariate model and the final model was constructed by backward stepwise method with the significance level set at 5%. Analysis was performed with SPSS version 15. Ethical clearance was obtained Urinary incontinence in women in Sri Lanka- Pathiraja et al. Figure 1. Flow chart for sampling technique. Vol 14 No 03 May-June 2017 3076 defining the severity of UI. UI occurrence was consid- ered mild if it occurred once or twice a week, moderate if it was once or twice a day and severe if it was 3 or more times a day. Women who had UI less than once a week or a month or who did not have any complaints were considered continent. Medical conditions were considered to be present if there was a diagnosis made by a Consultant Physician or a Consultant Endocrinolo- gist and only if the patient had a diagnosis card. An interviewer-administered questionnaire (IAQ) was used as the data collection instrument. This IAQ con- sists of two parts; Socio demographic factors, Medical and Obstetric History, and the King’s Health Question- naire (KHQ), which consists of UI symptoms for the prevalence study. The KHQ is a validated questionnaire with high reliability(12) and has been translated into two languages Sinhala and Tamil. To ensure that the ques- tionnaire was easily understandable, a few months be- fore the onset of the study, pre testing was done in a small group of women (n=50) in Borelesgamuwa MOH area. After reviewing these preliminary results, some questions were revised resulting in an improved, final version. The KHQ's good reliability was evidenced by Cronbach's alpha coefficients of > 0.60, indicating rea- sonable consistency in all domains. Informed written consent was obtained from study participants prior to conducting the study. All subjects were told that participation in the investigation was strictly voluntary and that data collected would not be used for anything except the aims of the research. The public health midwives were trained to collect data and all subjects were told that participation in the investi- gation was strictly voluntary. The questionnaire was administered in a setting that provided privacy during the interview and the time taken to complete the ques- tionnaire was between 15 and 20 minutes per subject. Only the investigators handled the data and the confi- dentiality was maintained. All women who were identi- fied to have more than moderate UI were referred to the respective /nearest tertiary hospitals. Figure 2. Prevalence of UI in the three districts of Sri Lanka. from the Ethical Review Committee of the Faculty of Medical Sciences, University of Sri Jayewardenepura. RESULTS The total of 2354 participated in the study 896 (38.1%) from Colombo, 725 (30.8%) from Rathnapura and 733 (31.1%) from Trincomalee. The mean age of the women was 43.8 + 14.3 years (range 18 – 90 years). Most women (92.4%) were mar- ried and 66.7% reported two or more children. Near- ly 78% had completed secondary education, but with only 13% being employed. Majority in the study were Sinhala Buddhist (65.6%); the remaining participants, primarily estate workers, were Tamil Hindus. Nearly 12% had co-morbidities such as Diabetes and respira- tory disease. Among all women, 44.4% had no symptoms of UI, whereas 10% had symptoms of only stress inconti- nence, 15.6% had symptoms of only urge incontinence and 29.9% had symptoms of both stress and urge in- continence. Our results indicated that 1308 women (55.5 %) had suffered from symptoms of UI within the last 3 months. There was a statistically significant difference in the prevalence of UI in all three districts (p < 0.001). Com- pared to other two districts, Ratnapura district had a higher prevalence of SI (14.5%) and Trincomalee dis- trict had a higher prevalence of UI (36.8%). (Figure 2) It was observed that the prevalence of UI increased with age, being especially apparent in the age groups of 36– 50 years and 51 years and over (p = 0.001). In comparison of the severity of incontinence, for both SUI and UUI, there was a statistically significant as- sociation in the trend analysis between the age groups from young to old. Incontinence was mostly observed in females who were 51 years and above. Table 4 shows stepwise logistic regression analysis, formed with possible risk factors. Those in the age group 36 – 50 years and 51 years and above had two times the risk of UI than those in the 18 – 35 age group. Compared to employed women, unemployed women had a higher risk of UI. Those with Respiratory prob- lems and with Diabetes were also found to have higher risk of UI. Those living in the Trincomalee District were found to have higher risk of UI when compared to those living in Colombo District (OR = 4.58; 95% CI = 3.35 – 6.27). According to the results, no significant association was revealed between the frequency of UI and the women’s marital status and educational status. DISCUSSION UI is usually a hidden problem, either because the pa- tient considers the problem as normal or they feel em- barrassed to seek health care. Significant UI can be simply defined as bothersome urinary leakage and it is important to find out the mean number of woman with significant UI rather than the crude prevalence rate to optimize the health care re- sources(6). In the literature worldwide, there is great di- versity in the prevalence of UI reported among women, which is between 16.3 and 54.8%(2,6,13). In studies car- ried out in Sri Lanka, prevalence of 9.8% was reported by some researchers(14) and much higher prevalence rate of 23.3% was reported by others(15). We used a vali- dated questionnaire rather than the questions prepared Urinary incontinence in women in Sri Lanka- Pathiraja et al. Female Urology 3077 Table 1. The socio demographic and medical characteristics of the women studied. Variable n % Age Group (years) < 35 787 33.4 36 – 50 781 33.2 > 51 780 33.1 Total 2448* Education level No schooling 110 4.7 Primary education 396 16.8 Studied up to grade 10 990 42.1 Studied up to Advanced level 798 33.9 Post Graduate education 46 2.0 Total 2340* Religion Buddhist 1557 65.6 Others 793 34.3 Total 2350* Employment status Housewives 2270 83.1 Workers 84 13.0 Total 2354 Marital status Married/ Living together 2174 92.4 Single/ Unmarried/ Divorced 159 6.7 Total 2333* Number of births No births 74 3.1 1 615 26.1 2 795 33.8 ≥ 3 774 32.9 Total 2258* Co morbidities Diabetes Mellitus 179 7.6 Bronchial Asthma/Respiratory Diseases 102 4.3 Continence status Continent 1046 44.4 Incontinent 1308 55.6 Stress incontinence only 235 10.0 Urge Incontinence only 368 15.6 Both Stress and Urge Incontinence 705 29.9 on the basis of the literature, used by some researches and our prevalence rate is 55.5%. This is much high- er than the similar studies carried out using validated questionnaire, which was reported to be around 29% 16. The reasons for this high discrepancy may include variations in definitions of the frequency and nature of incontinence, sampling methodologies, response rates and question formats(13). There are many conflicting results in the literature, re- garding the type of UI reported. We haven’t used any urodynamic investigations and diagnosis is based main- ly on symptoms. Some suggested MUI as the common- est type(17) in agreement with the present study. In all three districts MUI is the commonest (29.9%), followed by UUI (15.6%) and SI (10%). UUI was more common in women in Trincomalee district, where as SI was com- moner in Colombo and Rathnapura district. Only 30 % of Women in Rathnapura district reported symptoms of UUI, which is much lower than the other two districts. The differences in the findings in the prevalence of dif- ferent types of UI in three districts could be due to the standardization problems in the definition of UI and the differences in the ethnical and socio demographic char- acteristics of societies which also have been reported in previous researchers(18). The severity of incontinence in cases of SUI and UUI was mild in both situations (65.2% and 63.6% respec- tivley). A similar finding was reported in South Wales, where 64% of women surveyed complained of only slight incontinence(19). A limitation in the study was that we assessed, how the patients perceive their problem and the severity was not objectively measured. How- ever, as the severity of the disease is very important in decision making pertaining to treatment, and as severity affects quality of life, perceived severity of the problem could be equalized to the true status of the problem. In accordance with some other studies 20, as the edu- cation level of women decreased, the prevalence of UI increased and this could be due to the fact that wom- en with lower education levels accept UI symptoms as normal rather than a disease process and also perceive it as shameful and avoid presenting to health centers. In consistent with previous studies (21,22,23,24), a significant association of UI with increase age, being pregnant, childbirth and increased parity was found in our study. Similar to other studies(25), perineal trauma following vaginal delivery was significantly associated with UI. There was a significantly risk of incontinence among women with co morbidities such as diabetes and res- piratory problems in agreement with some studies20. This may be due to increased urine volume and detru- sor over activity associated with diabetes and increased abdominal pressure when coughing associated with res- piratory diseases. Interestingly, treatment-seeking behavior has been re- ported to be very poor(15) and only about 12%. We did not include investigation on the reasons in the current study. None of the patients were examined Gynaecologically and the presence of UI was determined by verbal in- quiry alone and this can reduce the precision of arriving at a diagnosis of UI. The prevalence of UI found in the women we studied was relatively high, with women displaying increasing severity and prevalence with increasing age. Despite the bias of this study based purely on the household female population, the sample can be considered rep- resentative of the country's female population since the Urinary incontinence in women in Sri Lanka- Pathiraja et al. Vol 14 No 03 May-June 2017 3078 Age groups in years 18-35 years (%) 36 – 50 years (%) 51 years and above (%) p value Stress Incontinence No Incontinence 583 (41.3) 426 (30.2) 403 (28.5) Mild 145 (23.7) 247 (40.4) 220 (35.9) .001 To a certain extent 35 (19.8) 59 (33.3) 83 (46.9) Frequently 24 (16.3) 49 (33.3) 74 (50.3) Urge Incontinence No Incontinence 546 (42.7) 366 (28.6) 366 (28.6) Mild 186 (27.3) 277 (40.7) 218 (32.0) .001 To a certain extent 37 (14.1) 91 (34.6) 135 (51.3) Frequently 18 (14.3) 47 (37.3) 61 (48.4) Table 3. Severity of Incontinence among different age groups Table 2. Distribution of the different types of UI in different age groups No UI Stress UI Urge UI Stress and Urge UI Age groups in years 18-35 years 481 (46.1) 65 (27.8) 102 (27.7) 139 (19.8) 36 – 50 years 296 (28.4) 70 (29.9) 130 (35.3) 285 (40.6) 51 years and above 267 (25.6) 99 (42.3) 136 (37.0) 278 (39.6) Table 4. Significant independent variables for UI by Regression analysis Variable No UI UI OR 95% CI Age Group (years) 18-35 years 481 (61.1) 306 (38.9) - 36 – 50 years 296 (37.9) 485 (62.1) 2.03 1.56 - 2.63 51 years and above 267 (34.2) 513 (65.8) 2.61 1.95 – 3.48 District Colombo 453 (50.6) 443 (49.4) - Ratnapura 450 (62.1) 275 (37.9) 0.61 0.48 – 0.77 Trincomalee 143 (19.5) 590 (80.5) 4.58 3.35 – 6.27 Number of children (mean ± SD) 2.19 (1.11) 2.63 (1.53) 1.11 1.02 – 1.21 Mode of Delivery Normal Vaginal Delivery 627 (43.8) 806 (56.2) - Cesarean section 233 (59.6) 158 (40.4) 0.61 0.46 – 0.79 Occupation Occupied 158 (51.8) 147 (48.2) - Presently Not Occupied 848 (43.4) 1108 (56.6) 1.75 1.28 – 2.39 Diseases No diseases 949 (45.8) 1124 (54.2) - Respiratory Problems 40 (39.2) 62 (60.8) 2.17 1.48 - 3.19 Diabetes Mellitus 57 (31.8) 122 (68.2) 1.97 1.19 - 3.23 data are comprehensive in terms of the current census. Although the prevalence and severity of UI was higher in older age groups, a significant proportion of women belonging to the younger age groups experienced UI as well. As UI was significantly associated with pregnan- cy and parity, lack of optimum perinatal and post-natal care may have contributed to the high prevalence rates in this age group. As many of the affected women were not aware of the preventive measures and available treatment options, it is important to provide education on lower urinary tract health both to the general public and healthcare providers. In Sri Lankan community, women have very close rela- tionship with members of the primary health care team specially, with the Public Health Midwife (PHM) for maternal care, childcare and family planning. There- fore, it would be beneficial to train PHM on raising awareness of UI among women and also they should be taught how to make basic continence assessments and informed the women that the occurrence of UI is not a normal part of aging and the availability of effective treatment options. ACKNOWLEDGEMENTS This study was funded by the Research Grant of the University of Sri Jayewardenepura. REFERENCES 1. Abrams P, Blaivas JG, Stanton SL, Andersen JT. Standardisation of terminology of lower urinary tract function. Neurourology and Urodynamics. 1988 Jan 1;7:403-27. 2. Schulman C, Claes H, Matthijs J. Urinary incontinence in Belgium: a population-based epidemiological survey. European urology. 1996 ; 32:315-20. 3. Bhatia JC, Cleland J. Self-reported symptoms of gynecological morbidity and their treatment in south India. Studies in family planning. 1995; 1:203-16. 4. Bhatia JC, Cleland J, Bhagavan L, Rao NS. Levels and determinants of gynecological morbidity in a district of south India. Studies in family planning. 1997; 1:95-103. 5. Thomas TM, Plymat KR, Blannin J, Meade TW. Prevalence of urinary incontinence. Br Med J. 1980; 8;281:1243-5. 6. El‐Azab AS, Mohamed EM, Sabra HI. The prevalence and risk factors of urinary incontinence and its influence on the quality of life among Egyptian women. Neurourology and urodynamics. 2007; 1;26:783-8. 7. Yip SK, Chung TH. Treatment-seeking behavior in Hong Kong Chinese women with urinary symptoms. International Urogynecology Journal. 2003; 14:27-30. Urinary incontinence in women in Sri Lanka- Pathiraja et al. Female Urology 3079 8. Thom D. Variation in estimates of urinary incontinence prevalence in the community: effects of differences in definition, population characteristics, and study type. Journal of the American Geriatrics Society. 1998;46:473-80. 9. Bø KA, Borgen JS. Prevalence of stress and urge urinary incontinence in elite athletes and controls. Medicine and science in Sports and Exercise. 2001; 33:1797-802. 10. Fitzgerald ST, Palmer MH, Kirkland VL, Robinson L. The impact of urinary incontinence in working women: a study in a production facility. Women & health. 2002; 35:1-6. 11. Schmidbauer JO, Temml C, Schatzl G, Haidinger G, Madersbacher S. Risk factors for urinary incontinence in both sexes. European urology. 2001; 39:565-70. 12. Kelleher CJ, Cardozo LD, Khullar V, Salvatore S. A new questionnaire to assess the quality of life of urinary incontinent women. BJOG: An International Journal of Obstetrics & Gynaecology. 1997;104:1374-9. 13. Hunskaar S, Burgio K, Diokno A, Herzog AR, Hjälmås K, Lapitan MC. Epidemiology and natural history of urinary incontinence in women. Urology. 2003;62:16-23. 14. Hemachandra NN, Rajapaksa LC, Manderson L. A “usual occurrence:” Stress incontinence among reproductive aged women in Sri Lanka. Social Science & Medicine. 2009;69:1395- 401. 15. Perera J, Kirthinanda DS, Wijeratne S, Wickramarachchi TK. Descriptive cross sectional study on prevalence, perceptions, predisposing factors and health seeking behaviour of women with stress urinary incontinence. BMC women's health. 2014;14:78. 16. Özerdoğan N, Beji NK, Yalçın Ö. Urinary incontinence: its prevalence, risk factors and effects on the quality of life of women living in a region of Turkey. Gynecologic and obstetric investigation. 2004;58:145-50. 17. Diokno AC, Brock BM, Brown MB, Herzog AR. Prevalence of urinary incontinence and other urological symptoms in the noninstitutionalized elderly. The Journal of urology. 1986;136:1022-5. 18. Ateskan U, Mas MR, Doruk H, Kutlu M. Urinary incontinence among the elderly people of Turkey: prevalance, clinical types and health-care seeking. Turk Geriatri Dergisi. 2000;3:45-50. 19. Yarnell JW, Voyle GJ, Richards CJ, Stephenson TP. The prevalence and severity of urinary incontinence in women. Journal of Epidemiology and Community Health. 1981;35:71-4. 20. Tozun M, Ayranci U, Unsal A. Prevalence of urinary incontinence among women and its impact on quality of life in a semirural area of Western Turkey. Gynecologic and obstetric investigation. 2009;67:241-9. 21. Kuh D, Cardozo L, Hardy R. Urinary incontinence in middle aged women: childhood enuresis and other lifetime risk factors in a British prospective cohort. Journal of epidemiology and community health. 1999;53:453-8. 22. Viktrup L. The risk of lower urinary tract symptoms five years after the first delivery. Neurourology and urodynamics. 2002;21:2-9. 23. Bortolotti, A., Bernardini, B., Colli, E., Di Benedetto, P., Giocoli Nacci, G., Landoni, M., et al. . Prevalence and risk factors for urinary incontinence in Italy. European Urology; 37: 30–35. 24. Thom DH, Van Den Eeden SK, Brown JS. Evaluation of parturition and other reproductive variables as risk factors for urinary incontinence in later life. Obstetrics & Gynecology. 1997;90:983-9. 25. Thom DH, Van Den Eeden SK, Brown JS. Evaluation of parturition and other reproductive variables as risk factors for urinary incontinence in later life. Obstetrics & Gynecology. 1997;90:983-9. Urinary incontinence in women in Sri Lanka- Pathiraja et al. Vol 14 No 03 May-June 2017 3080