Urinary symptoms and infections among female garment factory workers in Bangladesh

1 Background

The factory environment in manual-labor intensive industries in developing countries with poor health and safety regulations and practice is regarded as unhygienic for the workers. Bangladesh is the second largest garment manufacturer and supplier in the world providing employment for nearly four million workers – mostly young women with lower education and skills [1, 2–4]. Garment factories have received international scrutiny for low wages, high and variable workload, and exposure to hazardous working environments [2–5]. Such high work stress and adverse working environment have been identified as risk factors for adverse health outcomes [2–4] including urinary tract infections (UTI) and symptoms (UTS). Research on the prevalence and burden of UTI and UTS will help create awareness among the stakeholders: Western companies, consumers, donor agencies, health care providers, advocacy groups, non-governmental organizations, factory owners, the government, and worker groups on the importance. This may ensure a healthy and safe working environment for millions of young working women in the garment industry and other similar industries in developing countries.

A health problem common in these young female workers is the stress of the urinary tract system [2, 3]; this is independent of the 60% lifetime risk of UTI reported previously among women [6, 7]. UTI is among the most common infections, predominantly affecting women of the reproductive age group [6]. One etiology of these infections is possibly due to the bacteria urinary reservoir formed from chronic urine retention after experiencing an infrequent voiding schedule [8, 9]. Urinary retention also increases the risk of urine reflux into the kidney, raising the risk of hydronephrosis, pyelonephritis, and, in turn, chronic kidney or bladder injury [8]. Risk factors for UTI include female sex, pregnancy, older age, catheter use, diabetes, neurogenic bladder, stones, urine reflux, sexual activity and diaphragm use [10, 11]. In general, younger women of low socioeconomic status (SES) have been reported to be at a higher risk of UTI and UTS. The negative and adverse UT health consequences and outcome among factory and other similar working women groups have rarely been investigated and reported in Bangladesh- a developing country with 160 million people. In addition to causing pain, suffering, quality of life loss, non-communicable and infectious diseases, UTI and UTS cause undue financial burden for the lower SES population bracket by increasing their out-of-pocket spending on treatment as well as limiting their participation in income-generating activities [12].

Prolonged sitting posture at work and access to infrequent and poor toilet conditions may contribute to higher UTS among working women [13]. Data from the World Bank shows that from 1990 to 2015 there was a rise in the Bangladeshi urban population with access to improved sanitation facilities from 46.7% to 57.7%; however, the latest value falls far below that for all of South Asia (64.6%), Middle-Income nations (78.9%), and High-Income nations (97.2%) [14]. The UNICEF/WHO sanitation monitoring program reported that at least 40% of urban Bangladesh population were still beset by limited access to shared or ‘unimproved’ (e.g. hanging or bucket latrines) toilets as of 2015 [15].

Very little research has been conducted to determine the prevalence, burden and risk factors of urinary tract related health conditions in the working female population in developing countries. This cross-sectional study was conducted to assess the risk and prevalence of UTI and UTS among young female workers in Bangladesh and compare the findings with another group of comparable socioeconomic status. We hypothesize that the overall unhygienic working conditions, poor health awareness and inadequate access to clean toilets would predispose this population to adverse urinary health outcome.

2 Methods

This was a cross-sectional study to evaluate the hypothesis that female garment workers have a higher prevalence of UTI and UTS compared to a similar working population. Fieldwork and data collection were completed during February 2016 in Savar and Dhaka areas of Bangladesh. Institutional Review Board (IRB) approval was obtained from the necessary institutions.

The subjects included a convenience sample of female garment workers and a comparison group of working women who had occupations at local shops as shop owners or keepers, or at the Centre for the Rehabilitation of the Paralyzed (CRP) as guards, cleaners, secretaries or housekeepers. The sampling technique for garment workers included contacting workers through a known advocate as well as reaching out to factory workers directly. The comparison group was recruited by word of mouth around the area through snowballing. Each participant was asked if she would provide a urine sample; however, pregnant women and women undergoing menses during interview were excluded from providing a urine sample. Data on demographics, socioeconomic status, health conditions and history, UT symptoms, and toilet access patterns and practice at home and work were obtained from them.

UTI was determined by urine dipstick tests as these are cost-effective and practical [16, 17]. The research team consisted of three female medical and public health students from the University of Texas Health Science Center who drafted the basic demographic and health related questions from their direct patient care experience and extensive consulting of published research studies. Symptoms were measured by a self-reported survey consisting of questions related to genitourinary and renal function. In addition, the team used the International Consultation on Incontinence Modular Questionnaire-Female Lower Urinary Tract Symptoms (ICIQ-FLUTS) questionnaire; its validity and reliability have been published previously [18]. This survey contained a total of 12 questions, including four questions on filling, three on voiding, and five questions on incontinence. Each question is scored 1–4; thus, the ranges of overall scores are from 0 to 16, 12 and 20 for filling, voiding and incontinence scales, respectively. Higher scores indicate a greater impact of individual symptoms for the patient.

The language used (Bangla/Bengali) in the survey was appropriate for subjects with elementary level education. The lead investigator and the local team leaders are native Bengali speakers. Translated surveys were first piloted on a group of subjects with similar socio-demographic profile for 3 days to make needed modifications to make it user-friendly and then administered on a full-scale in February 2016. The research team trained four health science undergraduate student volunteers (all native Bengali speakers) who aided in obtaining subject consent and read the surveys when participants needed assistance. The written consent form was signed by all participants; for those who could not read, the consent was read to them in its entirety. After the completion of each survey which took about 30 minutes, participants were compensated 300 TK (about 5 US Dollar) for their time.

Data collection occurred in private rooms at CRP and the factory. Responses were recorded on paper. Urine dipstick tests were done by the respondents themselves. Dipsticks were interpreted by the same medical student for consistency. All data were entered and stored on password protected laptops and kept anonymous by only identification codes. SAS v9.4 (SAS Inc., Cary, NC) was used for data manipulation and analysis. To ensure the accuracy of data entry, responses from every 3rd participant were re-entered. Logistic regression model was applied to a binary outcome for the UTI (1 = yes and 0 = no) predicted by age, number of dependents, household size, work hours per day, garment worker (yes/no), education (yes/no), marital status (married/other), current employment (full time, part time and other), monthly income <10,000TK (yes/no), chronic pain≥6 months (yes/no), two or more life events experienced (yes/no), and diagnosed diseases including diabetes, hypertension, dysuria, joint pain, anxiety, asthma, vision problems and insomnia. The significance level was set at 5% for statistical tests.

3 Results

Data collection was completed from 310 garment workers and 297 comparison workers. Table 1 shows that garment workers were relatively young, had more dependents, were likely to be more sexually active, worked longer hours, earned less, worked more overtime hours per day, and had a shorter tenure of employment compared to non-garment workers. In addition, a higher proportion of garment workers had only primary level education, were married and were working full-time.

In terms of toilet access and usage patterns (Table 2), about one half of garment workers (54.52%) and comparison workers (58.92%) could take up to 3 toilet breaks in a day. However, only a few workers felt they had insufficient toilet breaks and very few reported they did not use tissue/towel or clean water after using a toilet. Comparison workers had a higher proportion of not having toilet access at home than garment workers (29.63% vs. 20.32%). Except for the toilet break sufficiency question, the other toilet usage patterns were somewhat comparable between the two groups of workers.

Table 3 shows that garment workers had a higher prevalence of many urinary symptoms and related conditions, except for kidney disease, dysuria, and sexually transmitted disease; however, these were not significantly different between the two groups. In particular, garment workers had a significantly higher prevalence of malodorous urine and vaginal discharge. They were also more sexually active, used more diaphragms, and had higher prevalence of menstruation compared to comparison workers.

The evaluation of urinary tract symptoms in Table 4 shows that garment workers had a significantly higher ICIQ-FLUTS score of voiding than comparison workers (0.44 vs. 0.27; p-value = 0.0167). However, comparison workers had a significantly higher ICIQ-FLUTS score for urinary incontinence (1.33 vs 0.97; p-value = 0.0142). Overall, garment workers and comparison workers had no significant difference in the total score of ICIQ-FLUTS. However, comparison workers had a significantly higher ICIQ-FLUTS score for urinary incontinence.

UTI was determined based on urine dip results, where 282 of 607 participants (46.46%) provided urine samples and 21 of 282 (7.45%) were found to have UTI (Table 5). Compared to participants without UTI, those with UTI were older, had more dependents, had a larger household size and were working fewer hours per day. However, the differences between the two groups were not statistically significant. Moreover, garment workers, part-time workers, more educated, married participants, and those with higher monthly incomes were found to have higher prevalence of UTI. Several diseases such as hypertension, dysuria, joint pain, anxiety and insomnia also appeared to make workers more prone to having UTI. These proportional differences were not statistically significant.

Table 6 shows that after considering all those risk factors in multivariate model, only garment work had a significant impact on the probability of having UTI. They had higher odds of having UTI as high as 5.46-fold (p-value = 0.0374; 95% CI = 1.10, 26.97) compared to other workers.

This study reports an important health issue among a group of reproductive age female workers making clothes for Western markets and consumers. In terms of toilet access and usage patterns, about half of the garment and other workers could not take more than 3 toilet breaks in a day. The finding may sound alarming; however, the overwhelming majority felt the breaks were sufficient. Inadequate toilet access at home for many these workers reflects the general poor living conditions of lower SES population in developing countries. The fact that 20–30% workers reported they do not have toilet availability at home is a very important finding itself and deserves attention and discussion among national stakeholders. Factory owners may be made cognizant of the fact that workers work for long hours and should have unlimited access to toilets and there should be provisions of tissue/towel and water. For young women who are going through menses, have pregnancy, are nursing babies, and who may have UTI and UTS, this need of going to toilets are likely be higher.

Both worker groups reported a higher prevalence of many urinary symptoms and related conditions. A significantly higher prevalence of malodorous urine and vaginal discharge among garment workers demand serious attention and further investigation. The findings highlight either their unawareness or negligence of these health issues or a lack of time or resources for them to address these. Untreated conditions like these can develop into additional complexities or are suggestive of other underlying ailments. Garment workers were more sexually active and thus were expected to have higher risk. Factory-level cost-effective health education and interventions will likely reduce much of these associated risk and burden.

Garment and comparison workers had no significant difference in the total score of ICIQ-FLUTS. Garment workers had a significantly higher ICIQ-FLUTS score for voiding problems compared to comparison workers indicating a poor coordination between the bladder muscle and the urethra [19].

UTI as determined in this study was conservative because only urine dip results were used: 7.45% were found to have UTI. Older workers with more dependents from larger households were found to be more vulnerable. Some health/medical conditions found to be correlated with increased risk of UTI were: hypertension, dysuria, joint pain, anxiety and insomnia.

A previous survey reported that UTI developed at a significantly higher level after workers were engaged in garment factory labor than prior to starting [20]. Another study found that 16.4% of female garment workers near Dhaka had confirmed UTI, the majority of who were married women between 20–29 years with reduced toilet breaks and water intake [21]. This value is similar to another study, which reported a 16.8 % incidence of UTI in adult females from Dhaka, Bangladesh with suspected UTI symptoms [22]. In this current study, expanding the detection method of UTI beyond dipsticks would likely report higher rates of UTI as reported here and make the finding more comparable to previous studies.

This study reports that 20–30% workers do not have toilet availability at home. It did not ask further questions on cleanliness or how many people were using the same toilet. An exploratory study based on 100 working women in Bangladesh, which included some employed in garments, showed the following: most participants lived in slums and shared an unclean toilet with 4–6 other families; garment and construction workers found unisex toilets at the workplace as unhygienic and socially unacceptable, deterring women from using these toilets; and street vendors felt burdened by a lack of access to toilets during work [23]. The toilet access was reported to be high among factory workers in this study but did not ask if there was gender separation of the toilets. A study in India on working women showed that even though 90% of the women had access to toilets at the workplace, 17% did not use the toilet during work hours due to lack of water or because of limitations of having only unisex toilets. In that study, UTI prevalence was found to be 20.3% [24]. This study found restrictions of toilet access among comparison workers. One study showed a positive association of non-family based violence toward poor urban women in India and a lack of household toilets [25]. A long distance from shared local toilets or public toilet deters women from using toilets for fear of potential violent or sexual attacks. In comparison to this study finding, results from a study on 134 low socioeconomic status female construction workers in Dhaka showed that 66.5% suffered from dysuria and 44.8% lacked access to toilets in the workplace [26].

For those subjects in this current study who did have symptoms or infections in their urinary systems, many possible explanations may be proposed. The limitations to urinary voiding are positively associated with UTI incidence and have been previously reported as an occupational hazard extant in the ready-made garment industry [21]. A major factor that may impact UTI incidence includes hygiene practices in many developing countries. Hygiene practices of young unmarried female nursing students were found to be positively associated with UTI associated with incorrect wiping after toilet use in one study [27].

Limitations of the study included barriers related to language and culture, privacy, and sample population selection. The pilot phase functioned to correct for any mistakes or misunderstanding in the survey translation and two of the research team members are Bangla speaking. The four student translators were also native to that country; however, translating of survey instruments may always be not perfect. Cultural inhibition may exist with respect to talking about genitourinary diseases or practices among young women in this country and could have affected the responses. Attempts were made to reduce participants’ reluctance to answer certain questions by having exclusively female translators. The subjective recall bias of participants on their medical and exposure history could also result in generating possible points of error.

As for the urine analysis, there was a chance of collecting tainted urine samples in case the participants did not follow instructions for the clean-catch procedure There can be a slight chance of misreading dipstick results. This study findings can only be generalized to urban and low-income female working population.

About 9% garment workers reporting having vaginal discharge, 7% reporting having dysuria, 1% having blood in urine, 2% having malodorous urine show that these factories need to pay more attention to basic health screening and provide better health education for the vast majority of its workers. Education on UTI and UTS, regular health screening and medical check-ups on women health issues can be relatively inexpensive and effective interventions for the factory management to take. Such an implementation of healthcare education occurred in a Tainan, Taiwan factory, where findings showed a decrease in prevalence of UTI from 9.8% to 1.6% in female factory workers thereafter [28].

The UN/WHO state that about 2.4 billion people in the world are living without a toilet in 2015 and reaffirms that sanitation, or the lack of it, can impact on livelihoods [29]. The International Labor Organization (ILO) states that all sanitary conveniences shall be provided and maintained with adequate facilities including toilet tissue [30]. They also state that access levels to toilets in the workplace reflects access levels to toilets in the home. However, very little data exist on how many workers do not have access. A lack of toilets at work and at home has consequences on businesses and workforce: poor health, absenteeism, attrition, reduced concentration, exhaustion, and decreased productivity [29]. Every establishment should provide pure drinking water, sufficient light and air, and separate toilets for its male and female workers [30]. The Bangladesh Labour Acts do have recommendations on safe sanitation conditions and hygienic practices. Recommendations have been made to separate toilet facilities for men and women at employment sites [23]. The updated Bangladesh Labor Act of 2006 states that establishments should provide clean working toilets with available detergents, designated restrooms for women, and at least a half hour daily break [31]. In 2010, the United Nations General Assembly recognized safe sanitation as a human right. Through this study, these unmet needs are highlighted again to the global community. There should be continued assessment of health status in global outscored industries and checking of implementation of enacted safety standards.

Conflict of interest

The members of this project declare no conflicts of interest.

Funding

This study was supported by funding from the UT Health Science Center San Antonio’s Global Health Program and Friends of CRP, Canada.