Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Urban–rural and socioeconomic variations in lifetime prevalence of symptoms of sexually transmitted infections among Bangladeshi adolescents

Urban–rural and socioeconomic variations in lifetime prevalence of symptoms of sexually... Aim: To identify socioeconomic and urban–rural variations in self-reported lifetime prevalence of symptoms of sexually transmitted infections (STI). Methods: This cross-sectional study used data from the Bangladesh Adolescents Survey 2005 conducted on 11,986 adolescents, using a cluster sampling methods. Data were analysed using SPSS applying principle components analysis, multivariate logistic regression analysis, and prevalence ratio (PR) with 95% confidence interval (CI). Results: Self-reported lifetime prevalence of STI symptoms was 11.6%. Urban adolescents had 11% lower prevalence than their rural counterparts (PR(U/R) = 0.89; 95% CI = 0.79-1.00). Probability of self-reported lifetime symptoms of STI was highest among 20–24 years old income-generating male educated workers of mid-socioeconomic status living in rural areas (0.31). Conclusions: The residence (urban–rural) factor is more influential than the socioeconomic factor. Simpler and cheaper mode of screening and case finding tools for STIs would greatly help. Health promotion and education programs can decrease the adolescents’ vulnerability to sexually transmitted diseases. Keywords: STI, Lifetime prevalence, Urban–rural, Adolescents, Bangladesh Introduction usually do not have access to basic information on sex- Sexually transmitted infections (STI), including acquired ual and reproductive health (SRH), skills in negotiating immunodeficiency syndrome, are influenced by a number sexual relationships, and access to affordable SRH of biological and medical factors on the one hand and services [5,6]. Although PIACT Bangladesh, a non gov- geographical, sociocultural or political factors on the other ernment organization, has worked with the National [1,2]. Globalization, involving development of trade and Curriculum and Textbook Board (NCTB) of Bangladesh movement of both goods and humans across countries to incorporate HIV modules in the curriculum of grades and territories, is also being postulated as a major factor 6 to 10, but unfortunately this issue has not yet been in- in the spread of infectious diseases worldwide [3]. In cluded in the curriculum [7,8]. Several studies revealed Bangladesh, nearly one-third of the population is in the 10 that a section of adolescents tend to engage in high-risk to 24-year age group [4]. The vast majority of this section activities such as visiting commercial sex workers with- is unaware of the risk of STIs and human immunodefi- out using condoms, and thus suffer from STIs as a ciency virus (HIV). The situation in Bangladesh, as consequence [9,10]. It has been reported that STIs sub- elsewhere in the world, is getting critical as adolescents stantially facilitated the rapid and extensive transmis- sion of HIV infections [11]. However, this issue is being given high priority in Bangladesh [12,13]. * Correspondence: showkat.gani@brac.net The World Health Organization has estimated that in Health and Population Research Unit, Research and Evaluation Division, BRAC, 15th Floor, BRAC Centre, 75 Mohakhali, Dhaka 1212, Bangladesh 1999 there were 340 million incidences of STIs (gonorrhoea, Full list of author information is available at the end of the article © 2014 Gani et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 2 of 8 http://www.apfmj.com/content/13/1/7 chlamydial infection, syphilis, and trichomoniasis) among 64 districts, 507 sub-districts, 87,928 villages and 8048 the 15-49-year-age group in the world, the highest concen- mahallas (the smallest identifiable administrative unit tration in Asia [14]. STIs are currently recognized as the in urban areas) [4]. major health and economic burdens for many developing and developed countries [14,15]. Thus, the control of STIs Sampling procedure is important not only to prevent complications from these A two-stage cluster sample survey was done in all the six infections but also to prevent HIV transmission. In divisions. The sample size for each division was planned Bangladesh, several studies have assessed adolescents know- to get sufficient precision in the estimates by sex and resi- ledge [9,10,16,17] and estimated the seroprevalence of STIs dence (urban/rural areas). and reproductive tract infections (RTIs) in both the general The Bangladesh Bureau of Statistics (BBS) estimated that population and the high-risk groups [18-21]. However, there a total of 1,000 primary sampling units (PSU) would be suf- is no nationwide study on the prevalence of lifetime symp- ficient to estimate the demographic indicators for Sample toms (persons known to have had the disease for at least Vital Registration System at district level, and thus formed part of their life) [22] of STIs. Therefore, this study aims to asamplingframe known as the Integrated Multi-Purpose estimate the prevalence of self-reported lifetime symptoms Master Sample [24]. The survey technique mainly followed of STIs, and the impact of socioeconomic, demographic and the Bangladesh Demographic and Health Survey (BDHS) urban–rural factors on the risk of STIs among the 12 to 24- [25] which uses the BBS sampling frame (Figure 1). Each year age group of population in Bangladesh. PSU consists of 200 households. In rural areas, a village is used as a PSU, but in some other parts of the country, the Methods big village or mauza (a geographical boundary consisting of Study design ≥1 village) is used as a PSU. In urban areas, the PSU is at This cross-sectional study used data from the Bangladesh the mahalla level. Figure 1 shows the sampling procedure. Adolescents Survey 2005. Study area Training of interviewers Bangladesh, covering an area of 147,570 km ,isone of One hundred and twenty graduates were preliminarily the most densely populated countries (1,114 popula- selected for a 2-week training in two batches. A training tion/km ) in the world and ranked 129th in the human manual was developed. Three experienced researchers development index [23]. It is surrounded by India and facilitated the training sessions through theoretical and the Bay of Bengal, and has a tropical climate [4]. practical lessons on data collection. Finally, 110 inter- Bangladesh is divided into six administrative divisions, viewers were selected, half of whom were females. Total PSUs in Bangladesh: 131,000 Bangladesh Bureau of Statistics PSUs for Sample Vital Registration System for district estimate: 1000 BRAC PSUs for The Bangladesh Adolescents Survey 2005 for divisional estimate: 361 Rural PSUs: 277 Urban PSUs: 84 Households: 16,616 Households: 5040 Rural adolescents (10-24 years): Urban adolescents (10-24 years): 11,721 3650 Rural adolescents (10-24 years) Urban adolescents (10-24 years) agreed to response: 11,384 agreed to response: 3558 Rural adolescents (12-24 years): Urban adolescents (12-24 years): 9016 2970 Figure 1 Sampling frame of the study. Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 3 of 8 http://www.apfmj.com/content/13/1/7 Measure instrument Statistical analysis Two questionnaires were developed, one for the house- Data were analysed using SPSS. The household wealth hold and one for the individual. The household question- index was calculated according to Filmer and Pritchett naire contained information on household characteristics, [27]. Principal components analysis was used to produce household possessions and amenities, disability and death. a new set of linearly combined measurements for the The individual questionnaire contained information on household wealth scores, which were classified into adolescents’ characteristics and their health, education and quintiles. To compare the lifetime prevalence between livelihood issues. The questionnaires were pretested urban and rural areas, the prevalence ratio (PR(U/R)) The lifetime prevalence of STI symptoms was estimated and corresponding 95% confidence intervals (CI) were by asking the question: “As some people can have infec- calculated by using the formula tions that cause pain, itching, scabies, moles or unusual qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi N −A N −A 1 1 0 0 secretions transmitted to their sexual organs, have you lnðÞ PR1:96 þ ; N A N A 1 1 0 0 ever felt that you suffer from any such infections?” Aper- son was considered to have had a STI, if he/she ever suf- fered from any infection that caused pain, itching, scabies, where N = urban total, A = urban cases, N = rural 1 1 0 moles or unusual secretions in his/her sexual organ. total, A = rural cases [28]. Multivariate logistic regression [29] was used to estimate Data collection the impact of age, sex, education, marriage, occupation, The interviewers worked in 20 teams consisting of 4 to wealth index, and the inhabitants’ residential status 5 members each (2 females and 2 males) headed by a (urban–rural) on the outcome variable, lifetime symptoms supervisor. To form a cluster, listing of 200 successive of STIs prevalence in terms of odds ratios (OR) and 95% households started from the northwest corner of the se- CI. The probability of prevalence was calculated from the lected village and moved anticlockwise. The list was used regression coefficients by using the formula for selecting a systematic sample of 60 households. The survey teams drew the map for each cluster showing the aþ b x i i location of households in the villages/mahallas. House- p ¼  ; "# X hold and individual information were collected in aþ b x i i 1 þ e Bangla language by same-sex interviewers privately and confidentially, particularly for the sample adolescents. If more than one eligible adolescent was in the household, Kish method [26] was used for selecting one of them. where p = estimated probability of having lifetime symp- While the survey teams had completed their work in toms of STI, a = intercept, b = regression coefficient, and the villages, five monitoring teams, each consisting of x = explanatory variables. one woman and one man monitored the data reliability on a 7% sub-sample basis in two phases. First, they re- Ethical issues interviewed the randomly selected sub-samples, and Ethical approval was obtained from the research review discussed with the survey teams regarding the dissimi- committee of BRAC Research and Evaluation Division. larities of the application of methods and questionnaire Written informed consent in Bengali was sought from all parts. Secondly, they re-interviewed the sub-samples respondents before interviewing, but the illiterate respon- and collected data without sharing with survey teams, dents provided thumbprint before a witness. Confidentiality to check data quality of the main survey. Two quality of the respondents’ identity was maintained and morally controllers worked separately to communicate survey right actions with respondents and community people were instructions from the principal investigator to inter- ensured. viewers and monitors as well as to ensure logistic sup- port to the teams. Results During the 5-month survey, 16,616 rural and 5,040 urban Data generation process households were visited (Figure 1). Of these, 11,721 (71%) Data were computerised using the FoxPro. Data cleaning rural and 3,650 (72%) urban adolescents aged 10–24 years and consistency checks were performed simultaneously by were eligible for this survey. Of them, 11,384 (97%) rural the investigators and the data management team through and 3,558 (98%) urban adolescents agreed to participate. checking the frequency distribution and the range for all Of the participating adolescents, 9,016 from rural and variables, and cross tabulations of linked questions to 2,970 from urban areas were allowed by their family to be identify and correct the inconsistencies in the data set. interviewed regarding sexual issues. Thus, 5,119 male Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 4 of 8 http://www.apfmj.com/content/13/1/7 (97.9%) and 6,867 female (96.6%) adolescents participated compared to urban areas (11.9 vs. 10.6%, p = .048) in the survey. (Table 2). The lifetime prevalence of STI increased signifi- cantly with increasing age from 5.5% in the 12–14 years Basic characteristics age group to 12.4% and 15.6% respectively in the 15–19 The mean age was 17.5 years, but the female adolescents and 20–24 years age groups. Males had a significantly and the urban adolescents were significantly older (sex: higher prevalence compared to females (14.8 vs. 9.3%, p < 17.9 vs. 16.9, p < .001, and urban–rural: 17.9 vs. 17.4, 0.001). The prevalence was significantly higher for the cur- p < .001, respectively) (Table 1). One-third of the adoles- rently married than that of the single adolescents (13.9 vs. cents were married (34%). The urban adolescents spent 10.4%, p < 0.001). Among the currently married adoles- significantly more years in school than the rural adoles- cents, the prevalence also increased significantly with in- cents (6.4 vs. 5.4 years, p < 0.001), partly because they creasing age groups (from 7.9 to 11.8% to 14.9%, p < 0.05), were older. Majority of them were students (36%) or The males had significantly higher prevalence compared were involved in domestic (36%) or money-earning work to females (19.2 vs. 13.3%, p < 0.01). The prevalence was (24%). The urban adolescents were more engaged in higher in adolescents involved in income-generating work income-generating work than the rural adolescents (27 when compared with that of other occupational groups. vs. 23%, p < 0.001). The wealth index showed that a sig- There was no trend in the wealth index. To summarise, nificantly higher proportion of rich lived in urban areas the significantly higher prevalence was found for the males compared to that in rural areas (50 vs. 15%, p < 0.001,) compared to the females, when particularly considered (Table 1). those adolescents who were currently married (20–24 years) and living in rural areas (19.9 vs. 14.8%, p < 0.05). Reported lifetime prevalence of STI symptoms The lifetime prevalence of STI symptoms among adoles- Urban/rural prevalence ratio cents was 11.6%, significantly higher in rural areas The prevalence ratio for self-reported lifetime symptoms of Table 1 Socioeconomic characteristics of 12 to 24 years STI was 11% lower in the urban adolescents compared to age group by residence rural adolescents (PR = 0.89, 95% CI = 0.79-1.0) (Table 2). Characteristics Residence All The urban/rural prevalence ratio (PR(U/R) was <1) for Urban Rural almost all categories of socioeconomic and demo- graphic characteristics. The only exception was being Age in years (%) separated/divorced/widowed in rural areas (PR = 2.8, 12-14 24.1 28.3 27.3 95% CI = 1.01-7.9). 15-19 37.5 39.3 38.8 20-24 38.4 32.4 33.9 Multivariate logistic regression analysis Mean age (years) 17.9 17.4 17.5 The likelihood of having reported lifetime symptoms of Sex ratio (M/F) 0.78 0.74 0.75 STI was three times higher for males than females Married (%) 30.2 35.1 33.9 (OR = 2.9, 95% CI = 2.4-3.6), and increased by age from No schooling (%) 10.3 11.7 11.4 one in the 12–14 years age group to 2.3 (95% CI = 1.9- Mean years of education 6.4 5.4 5.7 2.8) and 2.9 (95% CI = 2.3-3.6) in the 15–19 and 20–24 Occupation (%) years age groups, respectively (Table 3). Currently mar- ried respondents had a 40% higher risk than singles Student 38.3 35.4 36.1 (OR = 1.4; 95% CI = 1.2-1.7). Household work 30.5 37.8 36.0 Income-generating work 27.1 22.6 23.7 Others 4.1 4.2 4.2 Probability estimation The probability of lifetime prevalence of STI symptoms Wealth index (%) as outcomes of the effect of various combinations of fac- Poorest 5.6 17.8 14.8 tors is shown in Figure 2. The probability was low 2 9.1 20.7 17.8 among 12–14 years old female adolescent household 3 13.9 23.9 21.4 workers with no schooling or having primary education, 4 21.4 22.7 22.4 from either poorest or rich living in urban areas Rich 50.0 15.0 24.0 (p = 0.02) (Figure 2). The highest prevalence was found for the 20–24 years old males with higher education and Total (n) 2970 9016 11,986 a b income, from the middle socioeconomic group living in Note: All kinds of wage/self employed workers including business. Beggar, disabled, and unemployed etc. rural areas (p = 0.31). Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 5 of 8 http://www.apfmj.com/content/13/1/7 Table 2 Number of subjects, prevalence and urban/rural prevalence ratio (PR(U/R)) with 95% confidence intervals (CI) of lifetime symptoms of STIs of 12 to 24 years age group by socioeconomic factors Characteristic No. of subjects Prevalence (%) PR (U/R) 95% CI Urban (U) Rural (R) All Urban (U) Rural (R) Age in years 12-14 717 2551 5.5 5.2 5.6 0.93 (0.65-1.3) 15-19 1114 3542 12.4 10.7 13.0 0.82 (0.68-1.00) 20-24 1139 2923 15.6 14.0 16.3 0.86 (0.73-1.01) p value <0.001 <0.001 <0.001 Sex Male 1298 3821 14.8 13.2 15.3 0.86 (0.73-1.01) Female 1672 5195 9.3 8.6 9.5 0.91 (0.76-1.09) p value <0.001 <0.001 <0.001 Education No schooling 297 1082 12.8 11.4 13.2 0.87 (0.61-1.2) Primary 898 3661 10.6 10.7 10.6 1.0 (0.81-1.2) Secondary or higher 1775 4273 12.1 10.4 12.8 0.82 (0.70-0.96) p value 0.024 0.86 0.006 Marital status Single 2050 5789 10.4 9.5 10.7 0.88 (0.76-1.03) Currently married 897 3162 13.9 12.8 14.2 0.90 (0.74-1.1) Separated/Divorced/Widowed 23 65 13.6 26.1 9.2 2.8 (1.01-7.9) p value <0.001 0.001 0.0013 Occupation Student 1137 3189 9.0 7.7 9.4 0.81 (0.64-1.02) Household work 905 3410 11.8 11.5 11.8 0.97 (0.79-1.2) Income-generating Work 805 2035 15.2 13.3 16.0 0.83 (0.68-1.02) Unemployed/others 123 382 12.7 13.8 12.3 1.1 (0.67-1.9) p value <0.001 <0.001 <0.001 Wealth index Poorest 165 1604 9.5 7.3 9.7 0.75 (0.43-1.3) 2 271 1866 11.8 11.1 12.0 0.93 (0.65-1.3) 3 412 2156 13.0 15.3 12.6 1.2 (0.94-1.6) 4 637 2048 13.5 13.7 13.4 1.0 (0.81-1.3) Rich 1485 1342 9.7 8.3 11.3 0.74 (0.59-0.92) p value <0.001 <0.001 0.0098 All 2970 9016 11.6 10.6 11.9 0.89 (0.79-1.00) Abbreviation: STI sexually transmitted infection. a b p value is the test of heterogeneity; p value for test of differences in the prevalence by sex. Discussion This study confirms a high lifetime prevalence of STI This study shows that the self-reported lifetime symptoms is more prominent among males. It also con- prevalence of STI symptoms among adolescents in tradicts the misconception, found in one report, that Bangladesh is high and varies by their place of resi- STI prevalence is higher in women, which is possibly dence, and socioeconomic and demographic factors. due to lack of awareness and practice of personal hy- This information is useful for the policy-makers to de- giene [30]. Highest prevalence was found in currently velop strategies to prevent and control sexually trans- married older adolescents (20–24 years) living in rural mitted diseases. areas . Studies in Bangladesh and neighbouring countries Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 6 of 8 http://www.apfmj.com/content/13/1/7 Table 3 Multiple logistic regression analysis of risk for rural women was 39%. The higher prevalence is possibly lifetime symptoms of STI of the 12 to 24 years age partly explained by that they were older than the adoles- group, odds ratio (OR) and 95% confidence interval (CI) cents we studied, and thus had had a longer exposure Factor No. of subjects (STI) OR 95% CI time. In an urban-based cross-sectional study of resi- Yes No dents from low-income communities in Chennai, India [32], the prevalence for self-reported STI symptoms was Residence 16%, higher than that of our study. Reasons for varied Urban 315 2655 1 results could be due to differences in sampling method, Rural 1077 7939 1.13 0.98-1.3 measuring instruments, manner of queries, and way of Age in years filling questionnaires. 12-14 179 3089 1 The prevalence in high-risk groups is higher. A cross- 15-19 578 4078 2.3 1.9-2.8 sectional study of 18–30 years old street-based female sexworkers in DhakabyRahman et al. detected 33% 20-24 635 3427 2.9 2.3-3.6 prevalence of syphilis and 46% prevalence of Neisseria Sex gonorrhoea/chlamydia [33]. Another cross-sectional Female 756 4363 1 study estimated the prevalence of syphilis and Neisseria Male 636 6231 2.9 2.3-3.5 gonorrhoea/chlamydia at 8.5% and 58% respectively in Education 18–25 years old hotel-based female sex workers in No schooling 177 1202 1 Dhaka [21]. Using the Bangladesh adolescents survey 2005 dataset, Primary 485 4074 0.98 0.81-1.2 Gani and Ahmed [17] showed that the adolescents grew Secondary and higher 730 5318 1.03 0.84-1.3 up without knowing the facts of SRH, which is prominent Marital status among the low educated rural adolescents living in poor Single 815 7024 1 households. This pattern is also confirmed in this analysis. Currently married 565 3494 1.4 1.2-1.7 This might be because the cultural norms forced the par- Separated/Widowed/Divorced 12 76 1.7 0.90-3.2 ents/guardians to feel embarrassed or ashamed to discuss these issues with their adolescent children, and that the Occupation media is overlooking the issue simply as it involves the Student 388 3938 1 issue of sex [8,21]. Consequently most of them have had a Household work 508 3807 1.3 0.98-1.6 limited access to reliable and complete SRH information, Income-generating work 432 2408 0.88 0.73-1.1 which is making the adolescents vulnerable. Therefore, Unemployed/others 64 441 0.91 0.67-1.2 the cultural behaviour bias might have been associated Wealth index with this disease. Limitations of this study include non-response bias (3% Poorest 168 1601 1 for all selected adolescents) as it happened against the 2 253 1884 1.2 0.99-1.5 backdrop of a huge temporary migration of 15 to 24 years 3 335 2233 1.3 1.1-1.6 old male respondents, particularly for foreign employment 4 362 2323 1.4 1.1-1.7 [34]. The interviewers found it difficult to determine the Rich 274 2553 1.0 0.79-1.3 exact age of the respondents although the events calendar Abbreviation: STI sexually transmitted infection. was used to obtain best estimates. As such some errors cannot be ruled out. The lowest rate of STI symptoms have shown varying prevalence rates of STIs in the gen- among females might be due to silent infections or any eral population as well as in high-risk groups [18-20]. other unknown reasons relating to social stigma of STI. Bogaerts et al. [20] detected a low prevalence of Another limitation is that the self-reported symptoms are Neisseria gonorrhoea and chlamydia infections (0.5% not serologically confirmed as found in some other studies and 1.9%) among urban female clients in a basic health- [31,32], and, thus non-STI genital symptoms may have care clinic in Dhaka, while Sabin et al. [18] reported a been included in the results. higher sero-prevalence of current STIs (10.4% in 15 to Strengths of this first nationwide adolescents survey is that it permits to draw conclusions regarding the whole 54-year old men and 6.9% in 15 to 40-year old women) among slum dwellers in Dhaka, Bangladesh. Similar country by age, sex and place of residence. Furthermore, studies in neighbouring countries reported a higher the size of the study facilitates precise point estimates with narrow confidence intervals. As the sampling design is prevalence than what we found in this study. In Hainan [31], China, Xia et al. found that the lifetime prevalence similar to the 2004 BDHS [25], it allows comparison of of STI symptoms among the 18 to 49-year old married some background indicators for testing data reliability and Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 7 of 8 http://www.apfmj.com/content/13/1/7 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Combination of predictors Figure 2 Estimated probabilities of lifetime symptoms of STI of the 12 to 24 years age group. validity of this study. For example, the male/female ratio Competing interests The authors declared no conflict of interest with respect to the research, of this study is consistent with that of the national dataset authorship, and/or publication of this article. (0.80 in BAS 2005; 0.85 in BDHS 2004) [25]. Authors' contributions MSG participated in the proposal, design, coordinating the study, performed the statistical analysis, carried out of the study and drafted the manuscript. Conclusion AMRC participated in the proposal, and drafted the manuscript. LN participated in the proposal, design, coordinating the study, reviewed the The high self-reported lifetime prevalence of STIs statistical analysis and drafted the manuscript. All authors read and approved among adolescents reflects the varying living conditions of the final manuscript. in urban and rural areas, and demographic and socio- Acknowledgements economic characteristics. Older male adolescents with We thank the BRAC Research and Evaluation Division (BRAC-RED) for higher education from the middle class socioeconomic allowing us access to database of The Bangladesh Adolescents Survey 2005. background living in rural areas are especially at risk of This national survey, funded by BRAC Education Programme, is a collaborative initiative between BRAC-RED and the Population Council (USA). having the STIs. It thus indicates that STIs are endemic We are also grateful to Richard Moreino, executive editor of The Good in the entire country. Further research is needed to fully Morning (Bangladesh) and Hasan Shareef Ahmed, freelance science editor understand the epidemiology of STI in Bangladesh. A and communication consultant for their editorial support. We would like to thank all those who helped us during preparing this report. prerequisite is a cheap and reliable screening instrument to be able to quickly confirm symptomatic cases. Further Funding research is also needed to identify effective intervention The first two authors received no financial support for the research, to reduce the STI prevalence in Bangladesh. authorship, and/or publication of this article, but the third author, Lennarth Nyström was supported by the Umeå Centre for Global Health Research via Simpler and cheaper mode of screening and case find- the Swedish Council for Working Life and Social Research Grant #2006-1512, ing tools are urgently required for a clear understanding Umeå University, Sweden. of the epidemiology of STIs in Bangladesh. Therefore, Author details this simple, inexpensive, and replicable method could Health and Population Research Unit, Research and Evaluation Division, also be used to gauge the distribution of self-reported BRAC, 15th Floor, BRAC Centre, 75 Mohakhali, Dhaka 1212, Bangladesh. lifetime symptoms of STIs in other countries and, thus, BRAC Governing Body, BRAC, 75 Mohakhali, Dhaka 1212, Bangladesh. Population and Family Health, Mailman School of Public Health, Columbia create the basis for a thorough aetiological research on University, Columbia, NY 10032, USA. Department of Public Health and overall STIs and appropriate interventions to address Clinical Medicine, Epidemiology and Global Health, Umeå University, Umeå the problem. SE-901 85, Sweden. Probability 12-14y, female, primary level, unmarried, IGA, urban & 1st quintile 12-14y, female, primary level, unmarried, student, urban & 1st quintile 12-14y, female, primary level, unmarried, student, urban & 5th quintile 15-19y, female, no school, unmarried, IGA, urban & 5th quintile 15-19y, female, primary level, unmarried, household work, urban & 1st quintile 15-19y, female, no school, widow/widower, IGA, urban & 5th quintile 15-19y, female, primary school, currently married, household work, urban & 5th quintile 15-19y, female, secondary+, currently married, household work, urban & 5th quintile 15-19y, male, secondary+, currently married, IGA, urban & 5th quintile 20-24y, male, no school, unmarried, IGA, urban & 5th quintile 20-24y, male, secondary+, unmarried, IGA, urban & 5th quintile 20-24y, male, no school, unmarried, IGA, urban & 1st quintile 20-24y, male, no school, unmarried, IGA, rural & 5th quintile 20-24y, male, no school, unmarried, IGA, urban & 3rd quintile 20-24y, male, secondary+, unmarried, IGA, rural & 3rd quintile 20-24y, male, secondary+, unmarried, IGA, rural & 4th quintile 20-24y, male, secondary+, currently married, IGA, rural & 3rd quintile 20-24y, male, secondary+, currently married, IGA, rural & 4th quintile 20-24y, male, secondary+, widow/widower, IGA, rural & 3rd quintile 20-24y, male, secondary+, widow/widower, IGA, rural & 4th quintile Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 8 of 8 http://www.apfmj.com/content/13/1/7 Received: 14 March 2012 Accepted: 27 June 2014 21. Nessa K, Waris SA, Sultan Z, Monira S, Hossain M, Nahar S, Rahman H, Alam Published: 16 July 2014 M, Baatsen P, Rahman M: Epidemiology and etiology of sexually transmitted infection among hotel-based sex workers in Dhaka, Bangladesh. J Clin Microbiol 2004, 42:618–21. References 22. Last JM: A dictionary of epidemiology. 2nd edition. New York, NY: Oxford 1. Wasserheit JN: Epidemiological synergy. Interrelationships between University Press; 1988. human immunodeficiency virus infection and other sexually transmitted 23. United Nations Development Programme: Human Development Report 2010, diseases. Sex Transm Dis 1992, 19:61–77. The Real Wealth of Nations: Pathways to Human Development. 20th 2. Mayaud P, Mabey D: Approached to the control of sexually transmitted Anniversary edition. New York, NY: United Nations Development Programme infections in developing countries: old problems and modern (UNDP); 2010. Available from: http://hdr.undp.org/en/media/ challenges. Sex Transm Infect 2004, 80:174–82. HDR_2010_EN_Tables_reprint.pdf. Published 2010. 3. Cameron DW: Internationalism, infectious diseases and international 24. Bangladesh Bureau of Statistics: Report of sample vital registration system. development: there is an elephant in the living room. Int J Infect Dis Dhaka, Bangladesh: Bangladesh Bureau of Statistics (BBS), Planning Division, 2007, 11:3. Ministry of Planning; 2004. 4. Bangladesh Bureau of Statistics: Population census 2001. National report 25. National Institute of Population Research and Training, Mitra & Associates, (Provisional). Dhaka, Bangladesh: Bangladesh Bureau of Statistics; 2003. ORC Macro: Bangladesh Demographic and Health Survey. Dhaka, Bangladesh, 5. Bhuiya I, Chowdhury AH: Prevention of HIV/AIDS Among Young People in and Claverton, MD: National Institute of Population Research and Training Bangladesh: Improving Access to Life Skills Based Sexual and Reproductive (NIPORT), Mitra & Associates, and Macro International; 2005. Health Education and Condom Services for Male Youth. Dhaka, Bangladesh: 26. Kish L: A procedure for objective respondent selection within the National AIDS/STD Programme, Save the Children-USA and Population household. J Am Stat Assoc 1949, 44:380–7. Council; 2007. 27. Filmer D, Pritchett L: Estimating wealth effects without expenditure data-tears: 6. Wong ML: Preventing adolescents from adopting HIV-related risky sexual An application to educational enrollments in states of India (World Bank Policy behaviours: do we have all the answers? Asia Pac J Public Health 2006, Research Working Paper No. 1994). Washington, DC: World Bank; 2000. 18(1):1–2. 28. Rothman KJ: Epidemiology: an introduction. 2nd edition. New York: Oxford 7. World Health Organization: South-East Asia Regional Office: Adolescent University Press; 2002. Sexual and Reproductive Health HIV/AIDS among Young People: Compendium 29. Menard S: Applied Logistic Regression Analysis. Thousand Oaks, CA: Sage; of Institutions in Bangladesh. New Delhi, India: World Health Organization (WHO); 2007:38. 30. National Institute of Population Research and Training, Mitra & Associates, 8. Prothom Alo (A Daily Bangla Newspaper): The roundtable discussion: and Macro International: Bangladesh Demographic and Health Survey. Dhaka adolescent’s sexual health: role of guardians. Prothom Alo 2012, and Calverton: National Institute of Population Research and Training January 17:20 (Col 2-5). (NIPORT), Mitra & Associates, and Macro International; 2009. 9. Joint United Nations Programme on HIV/AIDS and The World Bank: 20 years 31. Xia DY, Liao SS, He QY, Choi KH, Mandel JS: Self-reported symptoms of of HIV/AIDS in Bangladesh: Experiences and Way forward. Dhaka, Bangladesh: reproductive tract infections among rural women in Hainan, China: Joint United Nations Programme on HIV/AIDS and The World Bank; 2009. prevalence rates and risk factors. Sex Transm Dis 2004, 31:643–9. 10. Bhuiya I, Rob U, Chowdhury AH, Rahman L, Haque N, Adamchak S, Kahn 32. Panchanadeswaran S, Johnson SC, Mayer KH, Srikrisnan AK, Sivaram S, ME: Improving adolescent reproductive health in Bangladesh. Frontiers in Zelaya CE, Go VF, Solomon S, Bentley ME, Celentano DD: Gender reproductive health. Dhaka, Bangladesh: Population Council; 2004. differences in the prevalence of sexually transmitted infections and 11. Robinson NJ, Mulder DW, Auvert B, Hayes RJ: Proportion of HIV infections genital symptoms in an urban setting in southern India. Sex Transm Infect attributable to other sexually transmitted diseases in a rural Ugandan 2006, 82:491–5. population: simulation model estimates. Int J Epidemiol 1997, 26:180–9. 33. Rahman M, Alam A, Nessa K, Hossain A, Nahar S, Datta D, Khan SA, Mian RA, 12. Government of Bangladesh: National strategic plan for HIV/AIDS 2004-2012. Albert MJ: Etiology of sexually transmitted infections among street-based NASP, MOHFW: Dhaka, Bangladesh: Government of Bangladesh; 2004. female sex workers in Dhaka, Bangladesh. J Clin Microbiol 2000, Available from: http://aidstar-one.com/sites/default/files/prevention/ 38:1244–6. resources/national_strategic_plans/Bangladesh_04-10.pdf. 34. The Daily Star (A Daily English Newspaper): Young Bangladeshis: The 13. International Conference on Population and Development (ICPD+5) ; 1999. soldiers of fortune. Daily Star 2006, March 13:5. Available from: http:// Available from: http://www.un.org/popin/icpd5.htm. www.thedailystar.net/2006/03/13/d603131502122.htm. 14. World Health Organization: Global prevalence and incidence of selected curable sexually transmitted infections: overviews and estimates. Geneva, doi:10.1186/s12930-014-0007-y Switzerland: World Health Organization (WHO); 2001. Cite this article as: Gani et al.: Urban–rural and socioeconomic variations 15. Sangani P, Rutherford G, Wilkinson D: Population-based interventions for in lifetime prevalence of symptoms of sexually transmitted infections reducing sexually transmitted infections, including HIV infection. West Sussex, among Bangladeshi adolescents. Asia Pacific Family Medicine 2014 13:7. UK: Cochrane Database Syst Rev: John Wiley & Sons Ltd; 2004. Issue 3:2–3. 16. Nahar Q, Tuñón C, Houvras I, Gazi R, Reza M, Huq NL, Khuda BE: Reproductive Health Needs of Adolescents in Bangladesh: A Study Report (ICDDR,B Working Paper No. 130). Dhaka, Bangladesh: ICDDR,B: Centre for Health and Population Research; 1999. 17. Gani MS, Ahmed SM: Growing up and reproducing: knowledge and practices of young people in Bangladesh. In Adolescents and youths in Submit your next manuscript to BioMed Central Bangladesh: some selected issues (BRAC Research Monograph Series No. 31). Dhaka, Bangladesh: BRAC: Bangladesh Rural Advancement Committee; and take full advantage of: 2006:73–96. 18. Sabin KM, Rahman M, Hawkes S, Ahsan K, Begum L, Black RE, Baqui AH: • Convenient online submission Sexually transmitted infections prevalence rates in slum communities of • Thorough peer review Dhaka, Bangladesh. Int J STD AIDS 2003, 14:614–21. 19. Hawkes S, Morison L, Foster S, Gausia K, Chakraborty J, Peeling RW, Mabey • No space constraints or color figure charges D: Reproductive-tract infections in women in low-income, low- • Immediate publication on acceptance prevalence situations: assessment of syndromic management in Matlab, • Inclusion in PubMed, CAS, Scopus and Google Scholar Bangladesh. Lancet 1999, 354:1776–81. 20. Bogaerts J, Ahmed J, Akhter N, Begum N, Rahman M, Nahar S, Van Ranst M, • Research which is freely available for redistribution Verhaegen J: Sexually transmitted infections among married women in Dhaka, Bangladesh: unexpected high prevalence of herpes simplex type Submit your manuscript at 2 infection. Sex Transm Infect 2001, 77:114–9. www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Asia Pacific Family Medicine Springer Journals

Urban–rural and socioeconomic variations in lifetime prevalence of symptoms of sexually transmitted infections among Bangladeshi adolescents

Loading next page...
 
/lp/springer-journals/urban-rural-and-socioeconomic-variations-in-lifetime-prevalence-of-30fjx5UN4Y
Publisher
Springer Journals
Copyright
Copyright © 2014 by Gani et al.; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; General Practice / Family Medicine; Primary Care Medicine
eISSN
1447-056X
DOI
10.1186/s12930-014-0007-y
Publisher site
See Article on Publisher Site

Abstract

Aim: To identify socioeconomic and urban–rural variations in self-reported lifetime prevalence of symptoms of sexually transmitted infections (STI). Methods: This cross-sectional study used data from the Bangladesh Adolescents Survey 2005 conducted on 11,986 adolescents, using a cluster sampling methods. Data were analysed using SPSS applying principle components analysis, multivariate logistic regression analysis, and prevalence ratio (PR) with 95% confidence interval (CI). Results: Self-reported lifetime prevalence of STI symptoms was 11.6%. Urban adolescents had 11% lower prevalence than their rural counterparts (PR(U/R) = 0.89; 95% CI = 0.79-1.00). Probability of self-reported lifetime symptoms of STI was highest among 20–24 years old income-generating male educated workers of mid-socioeconomic status living in rural areas (0.31). Conclusions: The residence (urban–rural) factor is more influential than the socioeconomic factor. Simpler and cheaper mode of screening and case finding tools for STIs would greatly help. Health promotion and education programs can decrease the adolescents’ vulnerability to sexually transmitted diseases. Keywords: STI, Lifetime prevalence, Urban–rural, Adolescents, Bangladesh Introduction usually do not have access to basic information on sex- Sexually transmitted infections (STI), including acquired ual and reproductive health (SRH), skills in negotiating immunodeficiency syndrome, are influenced by a number sexual relationships, and access to affordable SRH of biological and medical factors on the one hand and services [5,6]. Although PIACT Bangladesh, a non gov- geographical, sociocultural or political factors on the other ernment organization, has worked with the National [1,2]. Globalization, involving development of trade and Curriculum and Textbook Board (NCTB) of Bangladesh movement of both goods and humans across countries to incorporate HIV modules in the curriculum of grades and territories, is also being postulated as a major factor 6 to 10, but unfortunately this issue has not yet been in- in the spread of infectious diseases worldwide [3]. In cluded in the curriculum [7,8]. Several studies revealed Bangladesh, nearly one-third of the population is in the 10 that a section of adolescents tend to engage in high-risk to 24-year age group [4]. The vast majority of this section activities such as visiting commercial sex workers with- is unaware of the risk of STIs and human immunodefi- out using condoms, and thus suffer from STIs as a ciency virus (HIV). The situation in Bangladesh, as consequence [9,10]. It has been reported that STIs sub- elsewhere in the world, is getting critical as adolescents stantially facilitated the rapid and extensive transmis- sion of HIV infections [11]. However, this issue is being given high priority in Bangladesh [12,13]. * Correspondence: showkat.gani@brac.net The World Health Organization has estimated that in Health and Population Research Unit, Research and Evaluation Division, BRAC, 15th Floor, BRAC Centre, 75 Mohakhali, Dhaka 1212, Bangladesh 1999 there were 340 million incidences of STIs (gonorrhoea, Full list of author information is available at the end of the article © 2014 Gani et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 2 of 8 http://www.apfmj.com/content/13/1/7 chlamydial infection, syphilis, and trichomoniasis) among 64 districts, 507 sub-districts, 87,928 villages and 8048 the 15-49-year-age group in the world, the highest concen- mahallas (the smallest identifiable administrative unit tration in Asia [14]. STIs are currently recognized as the in urban areas) [4]. major health and economic burdens for many developing and developed countries [14,15]. Thus, the control of STIs Sampling procedure is important not only to prevent complications from these A two-stage cluster sample survey was done in all the six infections but also to prevent HIV transmission. In divisions. The sample size for each division was planned Bangladesh, several studies have assessed adolescents know- to get sufficient precision in the estimates by sex and resi- ledge [9,10,16,17] and estimated the seroprevalence of STIs dence (urban/rural areas). and reproductive tract infections (RTIs) in both the general The Bangladesh Bureau of Statistics (BBS) estimated that population and the high-risk groups [18-21]. However, there a total of 1,000 primary sampling units (PSU) would be suf- is no nationwide study on the prevalence of lifetime symp- ficient to estimate the demographic indicators for Sample toms (persons known to have had the disease for at least Vital Registration System at district level, and thus formed part of their life) [22] of STIs. Therefore, this study aims to asamplingframe known as the Integrated Multi-Purpose estimate the prevalence of self-reported lifetime symptoms Master Sample [24]. The survey technique mainly followed of STIs, and the impact of socioeconomic, demographic and the Bangladesh Demographic and Health Survey (BDHS) urban–rural factors on the risk of STIs among the 12 to 24- [25] which uses the BBS sampling frame (Figure 1). Each year age group of population in Bangladesh. PSU consists of 200 households. In rural areas, a village is used as a PSU, but in some other parts of the country, the Methods big village or mauza (a geographical boundary consisting of Study design ≥1 village) is used as a PSU. In urban areas, the PSU is at This cross-sectional study used data from the Bangladesh the mahalla level. Figure 1 shows the sampling procedure. Adolescents Survey 2005. Study area Training of interviewers Bangladesh, covering an area of 147,570 km ,isone of One hundred and twenty graduates were preliminarily the most densely populated countries (1,114 popula- selected for a 2-week training in two batches. A training tion/km ) in the world and ranked 129th in the human manual was developed. Three experienced researchers development index [23]. It is surrounded by India and facilitated the training sessions through theoretical and the Bay of Bengal, and has a tropical climate [4]. practical lessons on data collection. Finally, 110 inter- Bangladesh is divided into six administrative divisions, viewers were selected, half of whom were females. Total PSUs in Bangladesh: 131,000 Bangladesh Bureau of Statistics PSUs for Sample Vital Registration System for district estimate: 1000 BRAC PSUs for The Bangladesh Adolescents Survey 2005 for divisional estimate: 361 Rural PSUs: 277 Urban PSUs: 84 Households: 16,616 Households: 5040 Rural adolescents (10-24 years): Urban adolescents (10-24 years): 11,721 3650 Rural adolescents (10-24 years) Urban adolescents (10-24 years) agreed to response: 11,384 agreed to response: 3558 Rural adolescents (12-24 years): Urban adolescents (12-24 years): 9016 2970 Figure 1 Sampling frame of the study. Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 3 of 8 http://www.apfmj.com/content/13/1/7 Measure instrument Statistical analysis Two questionnaires were developed, one for the house- Data were analysed using SPSS. The household wealth hold and one for the individual. The household question- index was calculated according to Filmer and Pritchett naire contained information on household characteristics, [27]. Principal components analysis was used to produce household possessions and amenities, disability and death. a new set of linearly combined measurements for the The individual questionnaire contained information on household wealth scores, which were classified into adolescents’ characteristics and their health, education and quintiles. To compare the lifetime prevalence between livelihood issues. The questionnaires were pretested urban and rural areas, the prevalence ratio (PR(U/R)) The lifetime prevalence of STI symptoms was estimated and corresponding 95% confidence intervals (CI) were by asking the question: “As some people can have infec- calculated by using the formula tions that cause pain, itching, scabies, moles or unusual qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi N −A N −A 1 1 0 0 secretions transmitted to their sexual organs, have you lnðÞ PR1:96 þ ; N A N A 1 1 0 0 ever felt that you suffer from any such infections?” Aper- son was considered to have had a STI, if he/she ever suf- fered from any infection that caused pain, itching, scabies, where N = urban total, A = urban cases, N = rural 1 1 0 moles or unusual secretions in his/her sexual organ. total, A = rural cases [28]. Multivariate logistic regression [29] was used to estimate Data collection the impact of age, sex, education, marriage, occupation, The interviewers worked in 20 teams consisting of 4 to wealth index, and the inhabitants’ residential status 5 members each (2 females and 2 males) headed by a (urban–rural) on the outcome variable, lifetime symptoms supervisor. To form a cluster, listing of 200 successive of STIs prevalence in terms of odds ratios (OR) and 95% households started from the northwest corner of the se- CI. The probability of prevalence was calculated from the lected village and moved anticlockwise. The list was used regression coefficients by using the formula for selecting a systematic sample of 60 households. The survey teams drew the map for each cluster showing the aþ b x i i location of households in the villages/mahallas. House- p ¼  ; "# X hold and individual information were collected in aþ b x i i 1 þ e Bangla language by same-sex interviewers privately and confidentially, particularly for the sample adolescents. If more than one eligible adolescent was in the household, Kish method [26] was used for selecting one of them. where p = estimated probability of having lifetime symp- While the survey teams had completed their work in toms of STI, a = intercept, b = regression coefficient, and the villages, five monitoring teams, each consisting of x = explanatory variables. one woman and one man monitored the data reliability on a 7% sub-sample basis in two phases. First, they re- Ethical issues interviewed the randomly selected sub-samples, and Ethical approval was obtained from the research review discussed with the survey teams regarding the dissimi- committee of BRAC Research and Evaluation Division. larities of the application of methods and questionnaire Written informed consent in Bengali was sought from all parts. Secondly, they re-interviewed the sub-samples respondents before interviewing, but the illiterate respon- and collected data without sharing with survey teams, dents provided thumbprint before a witness. Confidentiality to check data quality of the main survey. Two quality of the respondents’ identity was maintained and morally controllers worked separately to communicate survey right actions with respondents and community people were instructions from the principal investigator to inter- ensured. viewers and monitors as well as to ensure logistic sup- port to the teams. Results During the 5-month survey, 16,616 rural and 5,040 urban Data generation process households were visited (Figure 1). Of these, 11,721 (71%) Data were computerised using the FoxPro. Data cleaning rural and 3,650 (72%) urban adolescents aged 10–24 years and consistency checks were performed simultaneously by were eligible for this survey. Of them, 11,384 (97%) rural the investigators and the data management team through and 3,558 (98%) urban adolescents agreed to participate. checking the frequency distribution and the range for all Of the participating adolescents, 9,016 from rural and variables, and cross tabulations of linked questions to 2,970 from urban areas were allowed by their family to be identify and correct the inconsistencies in the data set. interviewed regarding sexual issues. Thus, 5,119 male Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 4 of 8 http://www.apfmj.com/content/13/1/7 (97.9%) and 6,867 female (96.6%) adolescents participated compared to urban areas (11.9 vs. 10.6%, p = .048) in the survey. (Table 2). The lifetime prevalence of STI increased signifi- cantly with increasing age from 5.5% in the 12–14 years Basic characteristics age group to 12.4% and 15.6% respectively in the 15–19 The mean age was 17.5 years, but the female adolescents and 20–24 years age groups. Males had a significantly and the urban adolescents were significantly older (sex: higher prevalence compared to females (14.8 vs. 9.3%, p < 17.9 vs. 16.9, p < .001, and urban–rural: 17.9 vs. 17.4, 0.001). The prevalence was significantly higher for the cur- p < .001, respectively) (Table 1). One-third of the adoles- rently married than that of the single adolescents (13.9 vs. cents were married (34%). The urban adolescents spent 10.4%, p < 0.001). Among the currently married adoles- significantly more years in school than the rural adoles- cents, the prevalence also increased significantly with in- cents (6.4 vs. 5.4 years, p < 0.001), partly because they creasing age groups (from 7.9 to 11.8% to 14.9%, p < 0.05), were older. Majority of them were students (36%) or The males had significantly higher prevalence compared were involved in domestic (36%) or money-earning work to females (19.2 vs. 13.3%, p < 0.01). The prevalence was (24%). The urban adolescents were more engaged in higher in adolescents involved in income-generating work income-generating work than the rural adolescents (27 when compared with that of other occupational groups. vs. 23%, p < 0.001). The wealth index showed that a sig- There was no trend in the wealth index. To summarise, nificantly higher proportion of rich lived in urban areas the significantly higher prevalence was found for the males compared to that in rural areas (50 vs. 15%, p < 0.001,) compared to the females, when particularly considered (Table 1). those adolescents who were currently married (20–24 years) and living in rural areas (19.9 vs. 14.8%, p < 0.05). Reported lifetime prevalence of STI symptoms The lifetime prevalence of STI symptoms among adoles- Urban/rural prevalence ratio cents was 11.6%, significantly higher in rural areas The prevalence ratio for self-reported lifetime symptoms of Table 1 Socioeconomic characteristics of 12 to 24 years STI was 11% lower in the urban adolescents compared to age group by residence rural adolescents (PR = 0.89, 95% CI = 0.79-1.0) (Table 2). Characteristics Residence All The urban/rural prevalence ratio (PR(U/R) was <1) for Urban Rural almost all categories of socioeconomic and demo- graphic characteristics. The only exception was being Age in years (%) separated/divorced/widowed in rural areas (PR = 2.8, 12-14 24.1 28.3 27.3 95% CI = 1.01-7.9). 15-19 37.5 39.3 38.8 20-24 38.4 32.4 33.9 Multivariate logistic regression analysis Mean age (years) 17.9 17.4 17.5 The likelihood of having reported lifetime symptoms of Sex ratio (M/F) 0.78 0.74 0.75 STI was three times higher for males than females Married (%) 30.2 35.1 33.9 (OR = 2.9, 95% CI = 2.4-3.6), and increased by age from No schooling (%) 10.3 11.7 11.4 one in the 12–14 years age group to 2.3 (95% CI = 1.9- Mean years of education 6.4 5.4 5.7 2.8) and 2.9 (95% CI = 2.3-3.6) in the 15–19 and 20–24 Occupation (%) years age groups, respectively (Table 3). Currently mar- ried respondents had a 40% higher risk than singles Student 38.3 35.4 36.1 (OR = 1.4; 95% CI = 1.2-1.7). Household work 30.5 37.8 36.0 Income-generating work 27.1 22.6 23.7 Others 4.1 4.2 4.2 Probability estimation The probability of lifetime prevalence of STI symptoms Wealth index (%) as outcomes of the effect of various combinations of fac- Poorest 5.6 17.8 14.8 tors is shown in Figure 2. The probability was low 2 9.1 20.7 17.8 among 12–14 years old female adolescent household 3 13.9 23.9 21.4 workers with no schooling or having primary education, 4 21.4 22.7 22.4 from either poorest or rich living in urban areas Rich 50.0 15.0 24.0 (p = 0.02) (Figure 2). The highest prevalence was found for the 20–24 years old males with higher education and Total (n) 2970 9016 11,986 a b income, from the middle socioeconomic group living in Note: All kinds of wage/self employed workers including business. Beggar, disabled, and unemployed etc. rural areas (p = 0.31). Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 5 of 8 http://www.apfmj.com/content/13/1/7 Table 2 Number of subjects, prevalence and urban/rural prevalence ratio (PR(U/R)) with 95% confidence intervals (CI) of lifetime symptoms of STIs of 12 to 24 years age group by socioeconomic factors Characteristic No. of subjects Prevalence (%) PR (U/R) 95% CI Urban (U) Rural (R) All Urban (U) Rural (R) Age in years 12-14 717 2551 5.5 5.2 5.6 0.93 (0.65-1.3) 15-19 1114 3542 12.4 10.7 13.0 0.82 (0.68-1.00) 20-24 1139 2923 15.6 14.0 16.3 0.86 (0.73-1.01) p value <0.001 <0.001 <0.001 Sex Male 1298 3821 14.8 13.2 15.3 0.86 (0.73-1.01) Female 1672 5195 9.3 8.6 9.5 0.91 (0.76-1.09) p value <0.001 <0.001 <0.001 Education No schooling 297 1082 12.8 11.4 13.2 0.87 (0.61-1.2) Primary 898 3661 10.6 10.7 10.6 1.0 (0.81-1.2) Secondary or higher 1775 4273 12.1 10.4 12.8 0.82 (0.70-0.96) p value 0.024 0.86 0.006 Marital status Single 2050 5789 10.4 9.5 10.7 0.88 (0.76-1.03) Currently married 897 3162 13.9 12.8 14.2 0.90 (0.74-1.1) Separated/Divorced/Widowed 23 65 13.6 26.1 9.2 2.8 (1.01-7.9) p value <0.001 0.001 0.0013 Occupation Student 1137 3189 9.0 7.7 9.4 0.81 (0.64-1.02) Household work 905 3410 11.8 11.5 11.8 0.97 (0.79-1.2) Income-generating Work 805 2035 15.2 13.3 16.0 0.83 (0.68-1.02) Unemployed/others 123 382 12.7 13.8 12.3 1.1 (0.67-1.9) p value <0.001 <0.001 <0.001 Wealth index Poorest 165 1604 9.5 7.3 9.7 0.75 (0.43-1.3) 2 271 1866 11.8 11.1 12.0 0.93 (0.65-1.3) 3 412 2156 13.0 15.3 12.6 1.2 (0.94-1.6) 4 637 2048 13.5 13.7 13.4 1.0 (0.81-1.3) Rich 1485 1342 9.7 8.3 11.3 0.74 (0.59-0.92) p value <0.001 <0.001 0.0098 All 2970 9016 11.6 10.6 11.9 0.89 (0.79-1.00) Abbreviation: STI sexually transmitted infection. a b p value is the test of heterogeneity; p value for test of differences in the prevalence by sex. Discussion This study confirms a high lifetime prevalence of STI This study shows that the self-reported lifetime symptoms is more prominent among males. It also con- prevalence of STI symptoms among adolescents in tradicts the misconception, found in one report, that Bangladesh is high and varies by their place of resi- STI prevalence is higher in women, which is possibly dence, and socioeconomic and demographic factors. due to lack of awareness and practice of personal hy- This information is useful for the policy-makers to de- giene [30]. Highest prevalence was found in currently velop strategies to prevent and control sexually trans- married older adolescents (20–24 years) living in rural mitted diseases. areas . Studies in Bangladesh and neighbouring countries Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 6 of 8 http://www.apfmj.com/content/13/1/7 Table 3 Multiple logistic regression analysis of risk for rural women was 39%. The higher prevalence is possibly lifetime symptoms of STI of the 12 to 24 years age partly explained by that they were older than the adoles- group, odds ratio (OR) and 95% confidence interval (CI) cents we studied, and thus had had a longer exposure Factor No. of subjects (STI) OR 95% CI time. In an urban-based cross-sectional study of resi- Yes No dents from low-income communities in Chennai, India [32], the prevalence for self-reported STI symptoms was Residence 16%, higher than that of our study. Reasons for varied Urban 315 2655 1 results could be due to differences in sampling method, Rural 1077 7939 1.13 0.98-1.3 measuring instruments, manner of queries, and way of Age in years filling questionnaires. 12-14 179 3089 1 The prevalence in high-risk groups is higher. A cross- 15-19 578 4078 2.3 1.9-2.8 sectional study of 18–30 years old street-based female sexworkers in DhakabyRahman et al. detected 33% 20-24 635 3427 2.9 2.3-3.6 prevalence of syphilis and 46% prevalence of Neisseria Sex gonorrhoea/chlamydia [33]. Another cross-sectional Female 756 4363 1 study estimated the prevalence of syphilis and Neisseria Male 636 6231 2.9 2.3-3.5 gonorrhoea/chlamydia at 8.5% and 58% respectively in Education 18–25 years old hotel-based female sex workers in No schooling 177 1202 1 Dhaka [21]. Using the Bangladesh adolescents survey 2005 dataset, Primary 485 4074 0.98 0.81-1.2 Gani and Ahmed [17] showed that the adolescents grew Secondary and higher 730 5318 1.03 0.84-1.3 up without knowing the facts of SRH, which is prominent Marital status among the low educated rural adolescents living in poor Single 815 7024 1 households. This pattern is also confirmed in this analysis. Currently married 565 3494 1.4 1.2-1.7 This might be because the cultural norms forced the par- Separated/Widowed/Divorced 12 76 1.7 0.90-3.2 ents/guardians to feel embarrassed or ashamed to discuss these issues with their adolescent children, and that the Occupation media is overlooking the issue simply as it involves the Student 388 3938 1 issue of sex [8,21]. Consequently most of them have had a Household work 508 3807 1.3 0.98-1.6 limited access to reliable and complete SRH information, Income-generating work 432 2408 0.88 0.73-1.1 which is making the adolescents vulnerable. Therefore, Unemployed/others 64 441 0.91 0.67-1.2 the cultural behaviour bias might have been associated Wealth index with this disease. Limitations of this study include non-response bias (3% Poorest 168 1601 1 for all selected adolescents) as it happened against the 2 253 1884 1.2 0.99-1.5 backdrop of a huge temporary migration of 15 to 24 years 3 335 2233 1.3 1.1-1.6 old male respondents, particularly for foreign employment 4 362 2323 1.4 1.1-1.7 [34]. The interviewers found it difficult to determine the Rich 274 2553 1.0 0.79-1.3 exact age of the respondents although the events calendar Abbreviation: STI sexually transmitted infection. was used to obtain best estimates. As such some errors cannot be ruled out. The lowest rate of STI symptoms have shown varying prevalence rates of STIs in the gen- among females might be due to silent infections or any eral population as well as in high-risk groups [18-20]. other unknown reasons relating to social stigma of STI. Bogaerts et al. [20] detected a low prevalence of Another limitation is that the self-reported symptoms are Neisseria gonorrhoea and chlamydia infections (0.5% not serologically confirmed as found in some other studies and 1.9%) among urban female clients in a basic health- [31,32], and, thus non-STI genital symptoms may have care clinic in Dhaka, while Sabin et al. [18] reported a been included in the results. higher sero-prevalence of current STIs (10.4% in 15 to Strengths of this first nationwide adolescents survey is that it permits to draw conclusions regarding the whole 54-year old men and 6.9% in 15 to 40-year old women) among slum dwellers in Dhaka, Bangladesh. Similar country by age, sex and place of residence. Furthermore, studies in neighbouring countries reported a higher the size of the study facilitates precise point estimates with narrow confidence intervals. As the sampling design is prevalence than what we found in this study. In Hainan [31], China, Xia et al. found that the lifetime prevalence similar to the 2004 BDHS [25], it allows comparison of of STI symptoms among the 18 to 49-year old married some background indicators for testing data reliability and Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 7 of 8 http://www.apfmj.com/content/13/1/7 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Combination of predictors Figure 2 Estimated probabilities of lifetime symptoms of STI of the 12 to 24 years age group. validity of this study. For example, the male/female ratio Competing interests The authors declared no conflict of interest with respect to the research, of this study is consistent with that of the national dataset authorship, and/or publication of this article. (0.80 in BAS 2005; 0.85 in BDHS 2004) [25]. Authors' contributions MSG participated in the proposal, design, coordinating the study, performed the statistical analysis, carried out of the study and drafted the manuscript. Conclusion AMRC participated in the proposal, and drafted the manuscript. LN participated in the proposal, design, coordinating the study, reviewed the The high self-reported lifetime prevalence of STIs statistical analysis and drafted the manuscript. All authors read and approved among adolescents reflects the varying living conditions of the final manuscript. in urban and rural areas, and demographic and socio- Acknowledgements economic characteristics. Older male adolescents with We thank the BRAC Research and Evaluation Division (BRAC-RED) for higher education from the middle class socioeconomic allowing us access to database of The Bangladesh Adolescents Survey 2005. background living in rural areas are especially at risk of This national survey, funded by BRAC Education Programme, is a collaborative initiative between BRAC-RED and the Population Council (USA). having the STIs. It thus indicates that STIs are endemic We are also grateful to Richard Moreino, executive editor of The Good in the entire country. Further research is needed to fully Morning (Bangladesh) and Hasan Shareef Ahmed, freelance science editor understand the epidemiology of STI in Bangladesh. A and communication consultant for their editorial support. We would like to thank all those who helped us during preparing this report. prerequisite is a cheap and reliable screening instrument to be able to quickly confirm symptomatic cases. Further Funding research is also needed to identify effective intervention The first two authors received no financial support for the research, to reduce the STI prevalence in Bangladesh. authorship, and/or publication of this article, but the third author, Lennarth Nyström was supported by the Umeå Centre for Global Health Research via Simpler and cheaper mode of screening and case find- the Swedish Council for Working Life and Social Research Grant #2006-1512, ing tools are urgently required for a clear understanding Umeå University, Sweden. of the epidemiology of STIs in Bangladesh. Therefore, Author details this simple, inexpensive, and replicable method could Health and Population Research Unit, Research and Evaluation Division, also be used to gauge the distribution of self-reported BRAC, 15th Floor, BRAC Centre, 75 Mohakhali, Dhaka 1212, Bangladesh. lifetime symptoms of STIs in other countries and, thus, BRAC Governing Body, BRAC, 75 Mohakhali, Dhaka 1212, Bangladesh. Population and Family Health, Mailman School of Public Health, Columbia create the basis for a thorough aetiological research on University, Columbia, NY 10032, USA. Department of Public Health and overall STIs and appropriate interventions to address Clinical Medicine, Epidemiology and Global Health, Umeå University, Umeå the problem. SE-901 85, Sweden. Probability 12-14y, female, primary level, unmarried, IGA, urban & 1st quintile 12-14y, female, primary level, unmarried, student, urban & 1st quintile 12-14y, female, primary level, unmarried, student, urban & 5th quintile 15-19y, female, no school, unmarried, IGA, urban & 5th quintile 15-19y, female, primary level, unmarried, household work, urban & 1st quintile 15-19y, female, no school, widow/widower, IGA, urban & 5th quintile 15-19y, female, primary school, currently married, household work, urban & 5th quintile 15-19y, female, secondary+, currently married, household work, urban & 5th quintile 15-19y, male, secondary+, currently married, IGA, urban & 5th quintile 20-24y, male, no school, unmarried, IGA, urban & 5th quintile 20-24y, male, secondary+, unmarried, IGA, urban & 5th quintile 20-24y, male, no school, unmarried, IGA, urban & 1st quintile 20-24y, male, no school, unmarried, IGA, rural & 5th quintile 20-24y, male, no school, unmarried, IGA, urban & 3rd quintile 20-24y, male, secondary+, unmarried, IGA, rural & 3rd quintile 20-24y, male, secondary+, unmarried, IGA, rural & 4th quintile 20-24y, male, secondary+, currently married, IGA, rural & 3rd quintile 20-24y, male, secondary+, currently married, IGA, rural & 4th quintile 20-24y, male, secondary+, widow/widower, IGA, rural & 3rd quintile 20-24y, male, secondary+, widow/widower, IGA, rural & 4th quintile Gani et al. Asia Pacific Family Medicine 2014, 13:7 Page 8 of 8 http://www.apfmj.com/content/13/1/7 Received: 14 March 2012 Accepted: 27 June 2014 21. Nessa K, Waris SA, Sultan Z, Monira S, Hossain M, Nahar S, Rahman H, Alam Published: 16 July 2014 M, Baatsen P, Rahman M: Epidemiology and etiology of sexually transmitted infection among hotel-based sex workers in Dhaka, Bangladesh. J Clin Microbiol 2004, 42:618–21. References 22. Last JM: A dictionary of epidemiology. 2nd edition. New York, NY: Oxford 1. Wasserheit JN: Epidemiological synergy. Interrelationships between University Press; 1988. human immunodeficiency virus infection and other sexually transmitted 23. United Nations Development Programme: Human Development Report 2010, diseases. Sex Transm Dis 1992, 19:61–77. The Real Wealth of Nations: Pathways to Human Development. 20th 2. Mayaud P, Mabey D: Approached to the control of sexually transmitted Anniversary edition. New York, NY: United Nations Development Programme infections in developing countries: old problems and modern (UNDP); 2010. Available from: http://hdr.undp.org/en/media/ challenges. Sex Transm Infect 2004, 80:174–82. HDR_2010_EN_Tables_reprint.pdf. Published 2010. 3. Cameron DW: Internationalism, infectious diseases and international 24. Bangladesh Bureau of Statistics: Report of sample vital registration system. development: there is an elephant in the living room. Int J Infect Dis Dhaka, Bangladesh: Bangladesh Bureau of Statistics (BBS), Planning Division, 2007, 11:3. Ministry of Planning; 2004. 4. Bangladesh Bureau of Statistics: Population census 2001. National report 25. National Institute of Population Research and Training, Mitra & Associates, (Provisional). Dhaka, Bangladesh: Bangladesh Bureau of Statistics; 2003. ORC Macro: Bangladesh Demographic and Health Survey. Dhaka, Bangladesh, 5. Bhuiya I, Chowdhury AH: Prevention of HIV/AIDS Among Young People in and Claverton, MD: National Institute of Population Research and Training Bangladesh: Improving Access to Life Skills Based Sexual and Reproductive (NIPORT), Mitra & Associates, and Macro International; 2005. Health Education and Condom Services for Male Youth. Dhaka, Bangladesh: 26. Kish L: A procedure for objective respondent selection within the National AIDS/STD Programme, Save the Children-USA and Population household. J Am Stat Assoc 1949, 44:380–7. Council; 2007. 27. Filmer D, Pritchett L: Estimating wealth effects without expenditure data-tears: 6. Wong ML: Preventing adolescents from adopting HIV-related risky sexual An application to educational enrollments in states of India (World Bank Policy behaviours: do we have all the answers? Asia Pac J Public Health 2006, Research Working Paper No. 1994). Washington, DC: World Bank; 2000. 18(1):1–2. 28. Rothman KJ: Epidemiology: an introduction. 2nd edition. New York: Oxford 7. World Health Organization: South-East Asia Regional Office: Adolescent University Press; 2002. Sexual and Reproductive Health HIV/AIDS among Young People: Compendium 29. Menard S: Applied Logistic Regression Analysis. Thousand Oaks, CA: Sage; of Institutions in Bangladesh. New Delhi, India: World Health Organization (WHO); 2007:38. 30. National Institute of Population Research and Training, Mitra & Associates, 8. Prothom Alo (A Daily Bangla Newspaper): The roundtable discussion: and Macro International: Bangladesh Demographic and Health Survey. Dhaka adolescent’s sexual health: role of guardians. Prothom Alo 2012, and Calverton: National Institute of Population Research and Training January 17:20 (Col 2-5). (NIPORT), Mitra & Associates, and Macro International; 2009. 9. Joint United Nations Programme on HIV/AIDS and The World Bank: 20 years 31. Xia DY, Liao SS, He QY, Choi KH, Mandel JS: Self-reported symptoms of of HIV/AIDS in Bangladesh: Experiences and Way forward. Dhaka, Bangladesh: reproductive tract infections among rural women in Hainan, China: Joint United Nations Programme on HIV/AIDS and The World Bank; 2009. prevalence rates and risk factors. Sex Transm Dis 2004, 31:643–9. 10. Bhuiya I, Rob U, Chowdhury AH, Rahman L, Haque N, Adamchak S, Kahn 32. Panchanadeswaran S, Johnson SC, Mayer KH, Srikrisnan AK, Sivaram S, ME: Improving adolescent reproductive health in Bangladesh. Frontiers in Zelaya CE, Go VF, Solomon S, Bentley ME, Celentano DD: Gender reproductive health. Dhaka, Bangladesh: Population Council; 2004. differences in the prevalence of sexually transmitted infections and 11. Robinson NJ, Mulder DW, Auvert B, Hayes RJ: Proportion of HIV infections genital symptoms in an urban setting in southern India. Sex Transm Infect attributable to other sexually transmitted diseases in a rural Ugandan 2006, 82:491–5. population: simulation model estimates. Int J Epidemiol 1997, 26:180–9. 33. Rahman M, Alam A, Nessa K, Hossain A, Nahar S, Datta D, Khan SA, Mian RA, 12. Government of Bangladesh: National strategic plan for HIV/AIDS 2004-2012. Albert MJ: Etiology of sexually transmitted infections among street-based NASP, MOHFW: Dhaka, Bangladesh: Government of Bangladesh; 2004. female sex workers in Dhaka, Bangladesh. J Clin Microbiol 2000, Available from: http://aidstar-one.com/sites/default/files/prevention/ 38:1244–6. resources/national_strategic_plans/Bangladesh_04-10.pdf. 34. The Daily Star (A Daily English Newspaper): Young Bangladeshis: The 13. International Conference on Population and Development (ICPD+5) ; 1999. soldiers of fortune. Daily Star 2006, March 13:5. Available from: http:// Available from: http://www.un.org/popin/icpd5.htm. www.thedailystar.net/2006/03/13/d603131502122.htm. 14. World Health Organization: Global prevalence and incidence of selected curable sexually transmitted infections: overviews and estimates. Geneva, doi:10.1186/s12930-014-0007-y Switzerland: World Health Organization (WHO); 2001. Cite this article as: Gani et al.: Urban–rural and socioeconomic variations 15. Sangani P, Rutherford G, Wilkinson D: Population-based interventions for in lifetime prevalence of symptoms of sexually transmitted infections reducing sexually transmitted infections, including HIV infection. West Sussex, among Bangladeshi adolescents. Asia Pacific Family Medicine 2014 13:7. UK: Cochrane Database Syst Rev: John Wiley & Sons Ltd; 2004. Issue 3:2–3. 16. Nahar Q, Tuñón C, Houvras I, Gazi R, Reza M, Huq NL, Khuda BE: Reproductive Health Needs of Adolescents in Bangladesh: A Study Report (ICDDR,B Working Paper No. 130). Dhaka, Bangladesh: ICDDR,B: Centre for Health and Population Research; 1999. 17. Gani MS, Ahmed SM: Growing up and reproducing: knowledge and practices of young people in Bangladesh. In Adolescents and youths in Submit your next manuscript to BioMed Central Bangladesh: some selected issues (BRAC Research Monograph Series No. 31). Dhaka, Bangladesh: BRAC: Bangladesh Rural Advancement Committee; and take full advantage of: 2006:73–96. 18. Sabin KM, Rahman M, Hawkes S, Ahsan K, Begum L, Black RE, Baqui AH: • Convenient online submission Sexually transmitted infections prevalence rates in slum communities of • Thorough peer review Dhaka, Bangladesh. Int J STD AIDS 2003, 14:614–21. 19. Hawkes S, Morison L, Foster S, Gausia K, Chakraborty J, Peeling RW, Mabey • No space constraints or color figure charges D: Reproductive-tract infections in women in low-income, low- • Immediate publication on acceptance prevalence situations: assessment of syndromic management in Matlab, • Inclusion in PubMed, CAS, Scopus and Google Scholar Bangladesh. Lancet 1999, 354:1776–81. 20. Bogaerts J, Ahmed J, Akhter N, Begum N, Rahman M, Nahar S, Van Ranst M, • Research which is freely available for redistribution Verhaegen J: Sexually transmitted infections among married women in Dhaka, Bangladesh: unexpected high prevalence of herpes simplex type Submit your manuscript at 2 infection. Sex Transm Infect 2001, 77:114–9. www.biomedcentral.com/submit

Journal

Asia Pacific Family MedicineSpringer Journals

Published: Jul 16, 2014

References