An assessment of risk behaviours for HIV/AIDS among young female garment workers in Bangladesh

Abstract

This study attempted to assess the risk behaviours for HIV/AIDS among female garment workers aged 15–24 years. A total of 300 garment workers in Dhaka city were assessed with a semi-structured interview. Their knowledge of HIV/AIDS was moderate with high rates of misperception regarding modes of transmission. Further, symptoms of possible sexually transmitted infections (STIs) such as genital ulcer disease and vaginal discharge were prevalent, and risk behaviours such as low use of condoms, multiple sex partners and drug abuse were also found. Logistic regression identified that women who sourced information through radio/television, health service providers or friends had higher age at first intercourse and higher HIV/AIDS knowledge scores and were more likely to use condoms. Those who gained information through radio/televison or health service providers and those who abused drugs were more likely to have sex with multiple sex partners, while information gained through health service providers, and higher HIV/AIDS knowledge were protective against drug abuse, whereas sex with multiple partners was a predictor of drug abuse. We conclude that in Bangladesh, female garment workers are at risk of HIV/AIDS due to low education, lack of knowledge, STIs and risky behaviour.

Keywords

assessment HIV/AIDS risk behaviours female garment workers Dhaka

INTRODUCTION

The first case of HIV/AIDS in Bangladesh was detected in 1989. In 2007, the total number of reported HIV diagnoses was 1207.¹ HIV prevalence rates have remained less than 1% in all groups except injecting drug users (IDUs) in central Bangladesh, where it reached 4% among IDUs in 2003–2004.² Although overall HIV prevalence is low, significant under-reporting of cases occurs because of the country's limited voluntary testing and counselling capacity and the social stigma of HIV, which leads to fear of being identified and detected as HIV positive. Furthermore, the behavioural patterns and extensive risk factors that facilitate the rapid spread of the infection are making Bangladesh highly vulnerable to an HIV/AIDS epidemic. These risk factors include: low levels of HIV awareness, a large commercial sex industry, high numbers of transport workers, a low level of consistent condom use, significant prevalence of sexually transmitted infections (STIs), gender inequality and violence, stigma and discrimination related to HIV/AIDS, the routine needle sharing among IDUs and frequent interaction of risk groups with the general population. Prevalent STIs and HIV already exist among vulnerable populations in Bangladesh.³ Without a rapid and effective prevention programme, the spread of HIV/AIDS in Bangladesh could escalate to mirror that in surrounding countries such as India and Myanmar.⁴

Knowledge about symptoms, transmission and prevention of STIs and HIV/AIDS is inadequate in Bangladesh.⁵ Considering the low level of awareness especially among vulnerable groups as well as the general population regarding HIV/AIDS, the government of Bangladesh established a National Strategic Plan for HIV/AIDS 2004–2010, which includes one target ‘to create nationwide awareness through traditional and modern means of interpersonal communication and mass media’ among many other proposed strategies.⁶ In addition, one of the eight United Nations' Millennium Development Goals aims ‘to combat HIV/AIDS, Malaria and Tuberculosis’.⁷ However, there have been few extensive studies related to HIV/AIDS among garment workers in Bangladesh, although some studies on garment workers have been undertaken in India. In one study conducted on garment workers in Bangladesh, it was found that some level of sexual activity exists among the peers of female garment workers.⁸ The garment workers also encounter rape and sexual harassment. In another study it was found that physical and sexual abuse is common among garment workers and that the majority have low levels of knowledge about HIV/AIDS in Bangladesh.⁹ The objective of this study was to identify the level of knowledge on HIV/AIDS and to assess the risk behaviours for HIV/AIDS among female garment workers.

METHODS

Study area and population

This study was a cross-sectional, descriptive study. The study was conducted in November 2007 in Dhaka city because it has the highest concentration of garment industries, a highly mobile population and the highest number of individuals at high risk of HIV/AIDS. Out of around 1500 factories situated in Dhaka Metropolitan City, five garment factories were selected at random. The selected factories were situated in different parts of the City. A total of 350 female workers were randomly selected; there were 50 non-responders but 300 workers were interviewed successfully with the use of a semi-structured interview.

Data collection instrument

The data collection instrument was a semi-structured interview. Before collecting the final data, a pilot of the questionnaire was conducted among 10 workers in order to refine the language and sequence of the questions. The interview was carried out by trained female interviewers who obtained oral informed consent from interviewees, all of whom were allocated a unique identification number as workers' names were not used for the study.

The questionnaire was classified into four categories: socioeconomic profiles of the workers, risk behaviours, reproductive issues and HIV/AIDS/STI issues. Age at first sexual intercourse (intramarital and extramarital), monthly income and sources of knowledge were asked about with open-ended questions, while questions on the other variables were closed-ended. Data on symptoms of STIs were drawn from the health service providers of the respective garment factories, who provided treatment or suggestions to the workers who had such symptoms/infections; however, it was limited only to those who were interviewed.

Measuring knowledge on HIV/AIDS and risky behaviours

In order to assess the knowledge of HIV/AIDS among the garment workers, 14 items were developed. Workers were asked to respond either ‘yes’, ‘no’ or ‘I do not know’ to statements about HIV/AIDS transmission and prevention. Each correct response (yes) was given 1, while incorrect responses (both ‘no’ and ‘I do not know’) were given 0. The individual scores were summed into a scale score by an arithmetic transformation procedure and used in the analysis. To justify the single combined score for HIV/AIDS knowledge, Cronbach's alpha was calculated, which was 0.7932, suggesting that intercorrelations among knowledge items was strong enough to use them as a single combined variable (Table 2). Risky behaviours were defined by use of condoms, sex with multiple partners and drug abuse among female garment workers.

Data processing and analysis

Data were scrutinized with necessary editing and screening. Recoding of the open-ended questions was undertaken after editing the interview questionnaire. Furthermore, data were doubly entered into the Statistical Package for Social Sciences (SPSS) in order to minimize processing error. Logistic regression was applied as the dependent variables were dichotomous while independents were nominal, ordinal and interval levels of measurement. In this regard, the following general logistic equation¹⁰ was used:

where α is the intercept, β is the vector of unknown covariates, X is the vector of covariates that affect use of condom, sex with multiple sex partners and drug abuse and p is the end number of the series.

RESULTS

Background characteristics of the garment workers

The garment workers were young with a mean age of 20.5, (Table 1). Among the respondents, 59.3% reported that they were unmarried while 40.7% were married. Age at first sexual intercourse was low with above one-third of garment workers having had first sex between 11–15 years of age. 61% had first sex between 16–20 years of age and the remainder had sex before 25 years of age. Almost one-third of the garment workers had no schooling. However, 51.7% had primary level of education while 20% of garment workers were found to have more than primary level of education. The monthly income was less than 2000 BDT for 47.7%, 2001–4000 BDT for 46.3% and >4000 for 6% of the garment workers, respectively. Most of them resided in a rented house (84%).

Table 1

Distribution of background characteristics of sample population

Variables	Categories	No.	%
Current age	20.5 years
Marital status	Unmarried	178	59.3
	Married	122	40.7
Age at first sexual exposure	11–15	102	34.0
	16–20	183	61.0
	21–25	15	5.0
Educational qualification	No education	85	28.3
	Primary	155	51.7
	More than primary	60	20.0
Monthly income	≤2000	143	47.7
	2001–4000	139	46.3
	>4000	18	6.0
Residence	Rented house	252	84.0
	Mess and other places	48	16.0

Knowledge of HIV/AIDS

The garment workers had a medium level of knowledge, with an average knowledge level score of 8.38 out of a total score of 14 (Table 2). Almost universally, the garment workers had heard about HIV/AIDS (97.0%) and 78.3% of respondents mentioned that unsafe sex can transmit HIV, but knowledge of other modes of HIV transmission was lower.

Table 2

Distribution of knowledge about HIV/AIDS, sources of knowledge, sexually transmitted diseases and risk behaviours among garment workers

Variables	No. (Yes)	%
Knowledge on HIV/AIDS*
Have you heard about HIV/AIDS	291	97.0
Unsafe sex can transmit HIV	235	78.3
Infected blood transfusion transmit HIV/AIDS	175	56.7
Use of same syringe transmit HIV/AIDS	176	58.7
Mother to child transmission	198	66.0
Infected body transplant transmit HIV/AIDS	204	68.0
Breast milk transmit HIV/AIDS	191	63.7
Avoiding unsafe sex can prevent HIV/AIDS	234	78.0
Using condom during sex can prevent HIV/AIDS	226	75.3
Having no sex with sex workers can prevent HIV/AIDS	235	78.3
Having sex only with married partner can prevent HIV/AIDS	209	69.7
Mosquito bite can transmit HIV	189	63.0
Cough of HIV infected people can transmit HIV	190	63.3
Sharing same glass can transmit HIV	187	62.3
Cronbach's alpha = 0.7932
Mean knowledge score = 8.38
Sources of knowledge
Radio/TV	285	95.0
Health service provider	164	54.7
Teachers	82	27.3
Parents	92	30.7
Friends	195	65.0
Others	52	17.3
Sexually transmitted infections/diseases
Genital ulcer	32	10.7
Lower abdominal pain	83	27.7
Vaginal discharge	64	21.3
Others	23	7.7
Risk behaviours
Use of condom	99	33.0
Sex with multiple sex partners	93	31.0
Drug abuse	65	21.7

*Multiple response set

Comparatively, knowledge of HIV/AIDS prevention was higher than that of transmission among garment workers. However, misperceptions about HIV transmission were higher among the respondents. More than 60% of garment workers reported that mosquito bites, coughing and sharing the same glass can transmit HIV/AIDS.

Sources of knowledge on HIV/AIDS

Mass media was the most common source of HIV/AIDS information among the garment workers' followed by friends and health service providers (Table 2). However, teachers and parents as information sources were less common among the garment workers.

Symptoms of sexually transmitted infections

The workers were found to have a significant prevalence of symptoms that could indicate underlying STIs (Table 2). The most common symptom was lower abdominal pain mentioned by 27.7% of workers and vaginal discharge in 21.3% of respondents. Genital ulcer was reported by 10.7% of respondents while 7.7% mentioned other possible STIs.

Risk behaviour of garment workers

Various risky behaviours were observed among the garment workers in Dhaka city (Table 2). Only one-third of workers reported that they used condoms during sexual intercourse, and 31.0% and 21.7% of respondents also reported sex with multiple sex partners and drug abuse, respectively.

Logistic regression analysis of risk behaviours

An attempt has been made in this section to predict the likelihood of risk behaviours among the garment workers in regard to the outcome variables such as use of condoms during sex, sex with multiple partners and drug abuse. Table 3 presents the odds ratios as well as significance level (P value) by different independent variables.

Table 3

Odds ratios of risk behaviours among female garment workers

	Use of condoms		Sex with multiple partners		Drug abuse
Variables	Odds ratio	P	Odds ratio	P	Odds ratio	P
Age at present	1.020	0.832	1.025	0.613	0.908	0.117
Marital status (0 = unmarried and 1 = married)	0.820	0.075	0.787	0.055	0.814	0.118
Respondents’ education
No education*	1.000		1.000		1.000
Primary	0.959	0.055	0.845	0.066	0.989	0.212
More than primary	1.232	0.042	0.929	0.062	1.211	0.064
Age at first sexual exposure
11–15*	1.000		1.000		1.000
16–20	1.240	0.025	0.767	0.055	0.784	0.068
21–25	1.370	0.048	0.772	0.063	0.747	0.055
Information sources
Radio/TV (0 = no and 1 = yes)	1.220	0.025	0.767	0.040	0.824	0.068
Health service provider (0 = no and 1 = yes)	1.804	0.038	0.782	0.023	0.767	0.015
Teachers (0 = no and 1 = yes)	1.293	0.729	0.655	0.230	1.312	0.249
Parents (0 = no and 1 = yes)	0.949	0.065	0.955	0.076	0.979	0.412
Friends	1.135	0.045	0.955	0.076	0.979	0.412
Use of condom (0 = no and 1 = yes)	–	–	0.955	0.086	0.979	0.412
Sex with multiple sex partners (0 = no and 1 = yes)	0.855	0.026	–	–	1.576	0.014
Drug abuse (0 = no and 1 = yes)	0.755	0.036	1.979	0.021	–	–
Knowledge score	1.482	0.010	0.832	0.036	0.766	0.034

*Reference category

–, Not used in the analysis

The results show that women with more than primary schooling were significantly more likely (OR 1.23) to use condoms during sex than women with primary education. Higher age at first sexual intercourse also predicted condom use. Workers who were informed through health service providers about HIV/AIDS were significantly more likely to use condoms during sex (OR 1.80) followed by information through radio/television (OR 1.22) and friends (OR 1.14). Garment workers with higher knowledge of HIV/AIDS were more likely to use condoms during sex (OR 1.48). However, workers who had multiple sex partners and who abused drugs were significantly less likely to use condoms.

In regard to sex with multiple partners, workers who were informed about HIV/AIDS from radio/television or their health service provider were significantly less likely to have sex with multiple partners. Workers who abused drugs were significantly more likely to have sex with multiple partners. Conversely, workers who were informed from health service providers and who had higher levels of knowledge of HIV/AIDS were significantly less likely to abuse drugs whereas workers who had sex with multiple partners were significantly more likely to abuse drugs (OR 1.58).

Discussion

The main source of information about HIV/AIDS was the mass media (radio/television), which is similar to a case study conducted on students.¹¹ Health service providers were also a common source of information. However, communication with parents and teachers and information gained from other sources were very low in this study. Other work has shown that the stimulation of interest in parents and teachers about HIV/AIDS helped them to educate themselves as well as students.¹² Overall, the garment workers demonstrated satisfactory knowledge about HIV/AIDS transmission and prevention. Moreover, workers with a higher level of knowledge were more likely to use condoms and less likely to have sex with multiple partners and abuse drugs. However, misperceptions about HIV transmission were also found at high levels in this study. This suggests that knowledge about HIV/AIDS was incomplete among the study subjects. In another study conducted from 1997 among overseas job seekers in Bangladesh, it was found that most subjects who knew about AIDS had some inaccurate knowledge about how HIV is transmitted, such as believing that the virus can be transmitted through touch.¹³

Although symptoms of STIs were relatively low among the garment workers, this cannot be ignored. If such symptoms were indicative of STIs this could lead to a higher probability of getting infected with HIV/AIDS if exposed. Most importantly, the use of condoms during sex was low among the garment workers and sex with multiple partners as well as drug use was also found frequently in this study. This suggests that with the documented prevalence of sexual behaviour of multiple partners, lack of condom use, and the possible presence of STIs,^14,15 the prevention of risky sexual behaviour remains an essential intervention strategy for HIV/AIDS control within this population.

Garment workers who had more than primary level of education were more likely to use condoms during sexual activities than workers with no education. This may indicate that women with education are more likely to have better understanding about STIs/HIV as well as prevention compared with women with no education. In a study in Bangladesh, it was found that literate women were more likely to be knowledgeable about HIV/AIDS than illiterate women.⁹ Studies elsewhere found that women's education enhances female autonomy so that women develop greater confidence and capability to make decisions about their own health.^16,17 As a result, educated women are motivated and can motivate their partners to use condoms during sexual intercourse. As expected, women with higher HIV knowledge scores were more likely to use condoms and less likely to have multiple sex partners or abuse drugs. Knowledgeable people may be more health conscious and hence healthier behaviours are to be expected from them. However, studies conducted in Armenia found that knowledge of students had no significant contribution to use of condoms¹⁸ and in Kenya no association was found between knowledge of HIV transmission and sexual behaviour.¹⁹

Studies on the initiation of sexual lives and the use of condoms have shown that adolescents and young people tend not to use condoms when they begin their sexual lives very early.^20,21 Similarly, workers with higher age at first sexual intercourse were more likely to use condoms in this study. This may be because young people's sexual encounters are often unplanned, sporadic and are sometimes the result of social pressure or coercion.^22,23 Moreover, they tend to be insufficiently aware of the health risks associated with sexual intercourse.^24,25 In contrast, workers with higher age at first sex might be more likely to be health conscious and knowledgeable and hence conscious about negative health consequences of having sex without condoms. Information sources such as radio/television, health service providers, parents and friends had various effects on the likelihood of risk behaviour. However, among all the information sources, women who were informed from health service providers were most likely to use condoms during sex while less likely to have sex with multiple sex partners and abuse drugs. This may be because of the greater accuracy of information provided by health service providers compared to that of friends, teachers or neighbours.

Garment workers who had sex with multiple partners were more likely to abuse drugs and those who had multiple sex partners and who abused drugs were significantly less likely to use condoms. Moreover, workers who abused drugs were also significantly more likely to have sex with multiple partners. This association of one risk behaviour with other risk behaviours may indicate that garment workers could fall into a cycle of risk behaviour if one of the risk behaviours is present and ultimately this may lead them to be vulnerable to HIV/AIDS.

From this study it may be concluded that female garment workers in Bangladesh are at potential risk of HIV/AIDS due to an overall low level of knowledge about HIV/AIDS, the possible presence of STIs and risk behaviours such as low condom use, sex with multiple partners and drug abuse. Workers have low levels of communication with parents and teachers about HIV/AIDS, low levels of education and higher levels of misperception about HIV/AIDS, which are additional factors. Therefore, consideration should be given to various interventions aimed at reduction of misperceptions, screening and treatment of STIs, increased use of condoms, promotion of sexual monogamy and avoidance of drug abuse in order to reduce the risk of HIV/AIDS among garment workers.

LIMITATIONS OF THE STUDY

Due to the conservative culture among women in Bangladesh, female workers might not have answered accurately some of the culturally-sensitive questions such as use of condoms, sex with multiple partners and drug use; evidently this may raise the issue of bias.

Footnotes

ACKNOWLEDGEMENTS

This study was conducted by Bangladesh Institute of Social Research and all financial assistance was provided by this organization. I acknowledge the valuable suggestions provided by Dr Khurshed Alam, Chief Executive, Bangladesh Institute of Social Research.

References

UNGASS. Country Progress Report. National AIDS and STD Programme. Ministry of Health and Family Welfare, Bangladesh, 2008

GoB (2004, November). HIV in Bangladesh: The Present Scenario. A summary of key findings from the fifth round of serological and behavioural surveillance in Bangladesh 2002–2003. National Aids/STD Programme. DGHS, MOHFW, Dhaka, Bangladesh, 2004

Nili

, Sayem

, Faisal

. Assessment of knowledge on HIV/AIDS among vulnerable population groups in Bangladesh: A Review. Bang Privat Med Pract J 2007;101–6

UNAIDS. Bangladesh: Epidemiological Fact Sheets on HIV/AIDS and STIs. www.unaids.org , 2004 (last checked 22 October 2008)

Sharmin

, Sharmin

, Rahman

. Knowledge on HIV/AIDS among different population groups in Bangladesh: A review. Bang Privat Med Pract J 2005;11:89–92

National Strategic Plan for HIV/AIDS 2004–2010. National AIDS/STD Programme, Directorate General of Health Services. Bangladesh: Ministry of Health and Family Welfare, 2005

UNFPA. Ten Years of ICPD: Bangladesh Perspectives. Ministry of Health and Family Welfare, the People's Republic of Bangladesh, United Nations Population Fund, 2004

Amin

, Diamond

, Ruchira

, Naved

, Newby

. Transition to adulthood of female garment-factory workers in Bangladesh. Stud Fam Plann 1998;29:185–200

Mondol

, Hossain

, Rahman

. Knowledge and awareness about HIV/AIDS among garment worlers in Gazipur district, Bangladesh. Soc Sci 2008;3:528–30

10.

Afifi

. Computer Aided Multivariate Analysis. Variable Selection in Regression Analysis. 3rd edn. Boca Raton, Florida: CRC Press, 1996:166–96

11.

Sayem

, Nili

. Vulnerability to HIV: snapshot case study of students with higher socio-economic background in Dhaka city. Bang Privat Med Pract J 2007;13:114–7

12.

Maswanya

, Knowledge and risk perception of AIDS and reported sexual behaviour among students and secondary schools and colleges in Tanzania. Health Educ Res 2000;14:185–96

13.

Rahman

, Shimu

, Fukui

, Shimbo

, Yamamoto

. Knowledge, attitudes, belief and practices about HIV/AIDS among the overseas job seekers in Bangladesh. Pub Health 1999;113:35–8

14.

National Center for Health Statistics. CDC and prevention: premarital sexual experience among adolescent women-United States. 1970–1998. MMWR 1991;39:929–32

15.

CDC and prevention: trends in sexual risk behaviour among high school students – United States. 1991–1997. MMWR 1998;47:749–52

16.

Caldwell

. Maternal education as a factor in child mortality. World Heal Forum 1981;2:75–8

17.

Raghupathy

. Education and the use of maternal health care in Thailand. Soc Sci Med 1996;43:459–71

18.

Babikian

, Freir

, Hopkins

, FiClemente

, McBride

, Riggs

. An assessment of HIV/AIDS risk in higher education students in Yerevan, Armenia. AIDS Behav 2004;8:47–61

19.

Akwara

, Madise

, Hinde

. Perception of risk of HIV/AIDS and sexual behaviour in Kenya. J Biosoc Sci 2003;35:385–411

20.

Lagarde

, Caraël

, Glynn

, Educational level is associated with condom use within non-spousal partnerships in four cities of sub-Saharan Africa. AIDS 2001;15:1399–408

21.

Svare

, Kjaer

, Thomsen

, Bock

. Determinants for non-use of contraception at first intercourse: a study of 10,841 young Danish women from the general population. Contraception 2002;66:345–50

22.

Kelly

, Kalichman

. Increased attention to human sexuality can improve HIV-AIDS prevention efforts: Key research issues and directions. J Consult Clin Psychol 1995;63:907–18

23.

Lear

. Sexual communication in the age of AIDS: the construction of risk and trust among young adults. Soc Sci Med 1995;41:1311–23

24.

Buysse

. Adolescents, young adults and AIDS: a study of actual knowledge vs. perceived need for additional information. J Youth Adolesc 1996;25:259–71

25.

Minichello

, Paxton

, Cowling

, University students’ knowledge of STDs: labels, symptoms and transmission. Int J STD AIDS 1996;7:353–8

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.00 MB