Abstract
The rate of exclusive breastfeeding among workers in Indonesia tends to be low. There has been a limited number of research examining multilevel factors that determine exclusive breastfeeding among female textile industry workers. Therefore, this study aims to develop a model of exclusive breastfeeding among textile workers in West Java, Indonesia, using multilevel analysis at the individual and company levels. A cross-sectional study was conducted from August to December 2023 in 4 textile companies in West Java, Indonesia. Data were collected from 558 female workers using a structured questionnaire that captured individual- and company-level factors. A multilevel logistic regression analysis was conducted to develop models. Exclusive breastfeeding prevalence was 8.1% (45 respondents). Coworker support (aOR = 2.787, 95% CI: 1.337-5.863), supervisor support (aOR = 0.447, 95% CI: 0.221-0.952), and higher household income (aOR = 0.327, 95% CI: 0.156-0.715) were significantly associated with exclusive breastfeeding. These findings should encourage company management and policymakers to implement workplace programs that strengthen support for and education about exclusive breastfeeding.
Background
Currently, the industrial sector is growing rapidly, absorbing a significant number of workers, including female labor. According to data from several countries, the industrial sector has the highest proportion of female workers compared to other sectors. 1 With the textile sector being a well-known option in the general population, the industry naturally employs a large number of workers. Thus, with this high demand for labor, women are not exempt as recruitment targets in this sector. 2
There are both advantages and challenges to employing women in the textile industry. Female workers tend to be meticulous, loyal, and capable of multitasking, with high adaptability, which can be a strength for running the textile industry. 3 However, female workers also face vulnerabilities due to the dual burdens they carry. In addition to working in the industry, many women still have domestic responsibilities such as caring for children, managing the household, and supporting the family. Due to the pressure to meet the family’s economic needs, many mothers are often forced to return to work shortly after giving birth. 4 This creates a significant risk that female workers who have recently given birth may stop breastfeeding their babies when they return to work after maternity leave. Ideally, the best practice is to provide exclusive breastfeeding from birth until the baby is 6 months old. It is unfortunate that this care has to be neglected. 5
The lack of support results in a low prevalence of exclusive breastfeeding among working women. A study conducted in the workplace found that the prevalence of exclusive breastfeeding in the workplace was very low, at only around 11%. 6 Previous studies in Indonesia have also shown that the breastfeeding coverage rate among female workers in the industrial sector is only 32.3%-37.3%.7,8 This indicates that maintaining breastfeeding among working mothers in the industrial sector remains a significant challenge.
Based on the socio-ecological model framework, there are 5 levels (individual, interpersonal, community, organizational, and policy) that influence breastfeeding behavior. 9 Structural barriers primarily emerge at the individual, household, and workplace levels as critical factors. 10 For working mothers, difficulties in breastfeeding or expressing breast milk at work are closely related to lactation support at the workplace.11,12 The workplace can hold a significant role in providing breastfeeding facilities for female workers, such as maternity leave,13 -15 establishing breastfeeding policies at work,16 -18 flexible working hours,19,20 providing facilities and infrastructure for lactation rooms.21 -23
Unfortunately, the implementation of lactation policies as aforementioned, based on previous research, may not have been effective. While physical facilities and infrastructure have been provided, many aspects related to supporting lactation at the workplace remain unaddressed, particularly psychosocial aspects such as support from coworkers, supervisors, and a lactation-friendly workplace culture. 24 However, based on the researchers’ knowledge, few previous studies have examined interactions across levels in determining the determinants of exclusive breastfeeding among female textile industry workers in Indonesia. In consideration that a multilevel analysis approach can reveal differences in influence and interactions at each level—especially at the individual and workplace levels—it provides a comprehensive picture for policy-making and workplace lactation support programs. Therefore, this study aims to not only fill a gap in scientific knowledge, but also contribute to the design of comprehensive, sustainable models of workplace breastfeeding support. In addition, the findings of this study can serve as a basis for formulating company policies and strategies to create a breastfeeding-friendly work environment by providing facilities, flexible policies, and social support.
Method
Study Design, Data Collection, and Sampling
This study used a cross-sectional design and collected primary data through questionnaires. Sample selection was conducted in 2 stages: first, the selection of company clusters, in which 4 textile companies in West Java, Indonesia, are identified. The companies were selected purposely based on its identity as a textile industry, inclusion of female workers, and permission to conduct the study. Not all textile companies in West Java granted permission, so 4 companies were identified that varied in the implementation of lactation support programs based on West Java government data, and in the scale of production, which would affect working hours and workload. Among the 4 companies, Company A is a large-scale textile manufacturer, so it has more female workers than the other companies. Company B is a medium-sized company, so it has few female workers.
Second, after obtaining permission from the companies, the researchers recruited respondents from each company who were willing to participate in the study, using purposive sampling. Female workers with the following inclusion criteria are recruited: working as laborers, having children aged 6 months to 5 years, and having experience with breastfeeding while working at the company.
Data collection was conducted during working hours. Company managers gathered female workers who met the criteria based on data from the company clinic, and then conducted structured interviews using a questionnaire guide.
Study Sample, Outcome, and Predictor Variables
The minimum sample size was determined using the multivariate sample formula (Conroy, 2018), where n (sample size) = 25 × predictor variables. This study used 16 predictor variables, requiring a minimum of 400 respondents. However, in practice, 558 female workers were willing to complete the questionnaire manually in each textile industry.
The dependent variable in this study is exclusive breastfeeding, defined according to the WHO: breastfeeding begins at birth and continues until the baby is 6 months old, without giving any food or drink other than breast milk, including water, except for vitamins, minerals, and medicines. 25
The predictor variables are divided into 2 categories based on the level observed: individual level and workplace level. The selection of research focus within the socio-ecological framework at these 2 levels is based on the research objectives and the availability of data variation. The closest and most consistent determinant of mothers’ breastfeeding experiences is the workplace, which allows the research to identify logical intervention limitations at the workplace level. At the individual level, the variables were constructed based on the results of the author’s review of various related studies, including knowledge, attitudes, intentions, household income, workload, stress at work, exposure to formula milk advertisements, breastfeeding frequency, work-breastfeeding balance, travel time, socio-cultural factors, partner support, family support, supervisor support, coworker support, and work schedule.
Workplace stress variables are responses the participants experience when stress originates in the work environment, thereby affecting milk expression frequency and physiologically inhibiting breast milk production. This variable was measured using a special instrument, the Workplace Stress Scale from The Marlin Company and the American Institute of Stress. 26 The interpretation of the results is divided into 5 stress categories: potential danger if the total score is 31 to 40, severe if the total score is 26 to 30, moderate if the total score is 21 to 25, fairly low if the total score is 16 to 20, and lower if the total score is 15 or lower. However, in this study, no respondents fell into the lower category, so the lower stress level is not shown in the results table.
The workplace level includes 1 predictor variable, which also serves as the contextual variable of the study: the implementation of breastfeeding support programs. Program implementation variables refer to guidelines developed by the Indonesian government that cover breastfeeding policy components, lactation room availability, maternity leave, flexibility and break times. 27 Program implementation data is aggregated, meaning that respondents from the same company will have consistent answers for this variable. Program implementation variables are categorized as good if the company has a written lactation policy, has a lactation room, provides maternity leave, flexibility and break times at work, and vice versa. Company A has good program implementation, while companies B, C, and D have poor program implementation in the workplace.
Data Analysis
All variables were analyzed descriptively using frequency with a 95% confidence interval. For variables previously filled with numerical values, normality tests were conducted to categorize variables with mean or median values. The variables that underwent normality tests were: maternal age, length of employment, knowledge, attitude, intention, workload, work-breastfeeding balance, and sociocultural factors.
The analysis then continued to examine the bivariate relationships between variables and the outcome variable using the chi-square test.
The results of the bivariate analysis serve as the basis for determining which variables will be included in the multilevel analysis stage. Based on the principles of hierarchical well-formulated, P-value (<.05), and substance (biological plausibility), all variables are included in the multilevel logistic regression analysis modeling.
The analysis then proceeded to the multivariate stages using a multilevel logistic regression test with 3 models. Model 1 is the null model, which is an analysis that only includes the outcome variable, namely, exclusive breastfeeding. Model 2 is an analysis that only includes all individual variables, and model 3 is an analysis that includes all individual variables and contextual variables. Then, the variance coefficients were calculated for each model, and the Median Odds Ratio (MOR) was determined to determine whether there was variation in exclusive breastfeeding in each type of company by comparing the model at level 1 and the model at levels 1 and 2. The final model will be taken by considering the value of the model that has a higher MOR value.
Ethical Clearance
Before respondents agreed to participate in the study, the researchers requested their consent and explained that there was no coercion and that all information provided would be kept confidential and would not adversely affect respondents’ careers or personal lives. This study has obtained approval from the Ethics Committee of the Faculty of Public Health, University of Indonesia, with Approval Letter Number: Ket-101/UN2.F10.D11/PPM.00.02/2023. This step was conducted prior to data collection. The ethical review was conducted to ensure that the data collection process did not violate research ethical principles.
Result
Univariate analysis was conducted to examine the characteristics of the respondents (n = 558). Table 1 shows that the majority of respondents were aged 20 to 35 years (70.3%), had a high school education (3%), were from nuclear families (74.6%), and were from PT. A (94.4%), average work experience of 9.5 years, with the majority of children being male (50.4%), children aged over 24 months (59.7%), second children (52.5%), and children living with and cared for by the respondents (94.6%).
Characteristics of Respondents.
Table 2 shows that the household income variable is the only factor with a significant relationship with exclusive breastfeeding (P-value = .007). Other variables do not have a significant relationship with exclusive breastfeeding. There is no difference in the exclusive breastfeeding in the workplace factor, so it can be said that the variable of program support implementation does not have a significant relationship.
Relationship Between Individual Factors and Company Factors on Exclusive Breastfeeding.
P-value < .05.
The multilevel model stages are presented in Table 3. Model 1 (M1) is a zero model analyzed without adding predictor variables at levels 1 and 2, and the results show that the Median Odds Ratio (MOR) is 1.033. When the analysis continued in the second model (M2), which added variables from level 1 (individual) to the null model, the MOR value changed to 1.019, indicating a decrease in the MOR effect at the individual level. Then, in the third model (M3), by adding variables at level 2 (company) to the null model that already included variables at level 1 (individual), the MOR increased to 1.066. However, for the predictor variable at level 2, namely the implementation of a lactation support program, the P-value was >.25 (.562), meaning that this variable did not have a significant relationship with the outcome variable. Therefore, the model selected at stage M2 is chosen. It can be concluded that there is no difference in exclusive breastfeeding among female textile workers between companies that have implemented a lactation support program and those that have not. Variations in exclusive breastfeeding are only influenced by individual factors.
Analysis of Factors Influencing Exclusive Breastfeeding Among Female Textile Workers.
Thus, in the final model as shown in Table 4, exclusive breastfeeding is explained by individual factors, with the following model equation: Logit (Breastfeeding Support at Work) = −2.26 to 0.781* Supervisor Support + 1.025* Coworker Support – 1.097* Household Income.
Model Fit Variables Influencing Exclusive Breastfeeding Among Female Textile Workers.
Abbreviations: OR, odds ratio; SE, standard error; 95% CI, 95% Confidence Interval.
Female workers who receive support from coworkers are 2.7 times more likely to successfully provide exclusive breastfeeding than mothers who do not receive such support, after controlling for supervisor support and household income variables.
Discussion
Exclusive Breastfeeding
This study found that breastfeeding success was still very low. For working mothers, breastfeeding is not easy. The main reason hindering breastfeeding was that respondents felt their milk supply decreased upon returning to work after leave, leading to feelings that it was insufficient to meet their baby’s needs. Many factors contribute to decreased breast milk supply, including breastfeeding factors, irregular breastfeeding schedules, poor latch-on, psychological factors, and physical factors in the mother.
Previous research conducted among female textile workers in Jakarta, Indonesia also found similar results. Of the 27 informants interviewed qualitatively, only 2 exclusively breastfed. 28 Another study conducted among working mothers in Egypt also showed a similar finding, that the prevalence of breastfeeding among mothers with limited employment was only 11% who were successful in exclusive breastfeeding. 29
Beliefs and attitudes toward maternal health play a crucial role in a mother’s decision to breastfeed.30,31 Research shows that perceived barriers and self-efficacy are key determinants of a mother’s behavior in providing food/breast milk to her baby. 32 Perceived barriers primarily relate to mothers who feel their breast milk is insufficient for their baby 33 Perceived barriers can also be in the form of maternal attitudes, knowledge, and education that hinder exclusive breastfeeding. 34 This underscores the importance of examining the factors influencing exclusive breastfeeding among female industrial workers, while accounting for the barriers and benefits perceived by the mother.
Co-Workers Support
The results of the model development show that coworker support plays a very important role in the success of exclusive breastfeeding among female textile industry workers, followed by support from supervisors and household income.
The conditions observed at the research site indicate that the majority of respondents perceive a lack of support from coworkers, despite the significant influence of peers in community decision-making. Other studies have highlighted that support from colleagues is an essential component of workplace support. 35 Previous research also showed that low coworker support is one of the workplace challenges. 36 Effective workplace lactation support strategies can be implemented by, among other things, changing workplace culture by increasing support from colleagues and improving the physical environment at work. 37 Other findings also align with the statement that when working mothers perceive good communication from their colleagues, it influences their decision to continue breastfeeding or pumping breast milk, especially upon returning to work after maternity leave. 38
There are significant benefits from the communication and supportive workplace culture that is created, where everyone feels a social responsibility to support breastfeeding workers. Mothers can be motivated to continue breastfeeding after hearing experiences, especially from female colleagues at their workplace. 39
In addition, support from coworkers can take the form of assistance in temporarily replacing tasks when workers request a break to express breast milk, because the workload in the textile industry makes it extremely difficult for workers to leave production equipment unless a colleague on the same line helps monitor the running equipment while the worker pumps breast milk during work.
Supervisor Support
However, an interesting finding from this study is that support from supervisors is negatively associated with breastfeeding success among textile industry workers. A review of previous research shows that workers’ perceptions of support from supervisors are not always positively associated with exclusive breastfeeding. 37
The variety of companies where the research was conducted is 1 reason why support from superiors alone is not enough to guarantee successful breastfeeding. Previous research shows that there is diversity in the implementation of workplace lactation support. Large companies that can prepare lactation support facilities usually have a breastfeeding-friendly environment, policies, and other facilities, such as lactation rooms and more comprehensive equipment. 40 Support from superiors must encompass both emotional and structural contexts, with the emotional component comprising verbal communication between superiors and their employees, and the structural component comprising company support, such as facilities and infrastructure that facilitate breastfeeding. 41
In addition, support from superiors can be an indicator that the job accepted by female workers in the textile industry has a high workload. In the textile industry, production runs 24 hours a day, workers work in shifts, and face high targets, making it difficult for them to leave their workstations even when they have received support from their superiors. Other studies also support this: even though workers understand the benefits of breastfeeding and there is normative support from supervisors, if the lactation room is not suitable or available, the right to breastfeed cannot be exercised due to work time constraints and poor communication. 42 Thus, support from supervisors does not have a significant impact on increasing exclusive breastfeeding if there are no organizational changes.
Industrial workers have limited knowledge, low skills, and no authority over their work. 43 Thus, the role of workplace supervisors, especially female supervisors, is a key factor in fostering positive experiences related to breastfeeding. When supervisors offer flexibility to negotiate breastfeeding breaks at work, this provides positive emotional support for mothers.16,44 Workers generally perceive managerial support as important in their decision to continue and successfully practice exclusive breastfeeding. 45 In terms of the support companies can provide, the role of companies in offering both physical and non-physical support is very important. In this context, companies can facilitate workplace breastfeeding support by focusing on programs that foster a breastfeeding-friendly culture, including interventions to strengthen peer support. 46
Household Income
Another finding from this study is that household income is related to exclusive breastfeeding. High income actually reduces exclusive breastfeeding among working mothers. A previous study found that economic burdens, such as financial pressure, a lack of dual-income households, and limited access to government cash benefits, influence mothers to return to work immediately rather than take their maternity leave to earn additional income for their families. 47
On the other hand, workers with high family incomes can access formula milk. Other studies have also shown that high income can have a positive or negative impact on exclusive breastfeeding, depending on whether it enables the purchase of formula milk, especially when there are obstacles or discomfort in continuing to breastfeed. 48 In addition, high income is also linked to employment status, high work pressure, and low flexibility. As a result, working mothers tend to return to work sooner after their maternity leave. 12
The impact of high income on exclusive breastfeeding is also influenced by other factors, such as social culture, breastfeeding facilities at work, and workloads that challenge working mothers to continue or stop breastfeeding. In West Java, Indonesia, a cultural practice holds that giving formula milk to babies under 6 months of age is considered normal. 49 Research conducted in West Bandung Regency found that there are several socio-cultural factors that hinder exclusive breastfeeding, including myths about breast milk, such as the belief that babies will go hungry if they are fed only breast milk, so parents try to give them food other than breast milk, especially if they have a higher income and can access formula milk. 50
Limitations
This study has several limitations. The first limitation is the limited generalizability resulting from purposive sampling. The sample was dominated by respondents from Company A, the only large-scale company, so the proportion of respondents was female workers at Company A, while the other 3 companies were medium-scale industries with a limited number of female workers who could be respondents. In addition, the choice of a cross-sectional study design limits the ability to draw causal conclusions, so future research should use a design that can better establish causal relationships. The range of sociodemographic variables, such as parity and husband’s occupation, that were not all measured in this study may limit interpretation. Furthermore, cultural differences across other regions of Indonesia need to be considered, as this study was conducted in only 1 region, whose cultural values may differ from those of others. The sample selection criteria, which aimed to capture the experiences of working mothers who had breastfed in the textile industry within the last 3 years, could introduce bias into the study. In addition, the measurement of supervisor support variables is limited to emotional support aspects, such as providing flexibility and information about policies, and does not comprehensively cover more tangible indicators of support.
Conclusion
The breastfeeding success model proposed in this study suggests that coworker support, supervisor support, and household income should be considered when developing lactation support programs, especially in the textile industry. Increased exclusive breastfeeding among female textile industry workers can be achieved by empowering coworkers and supervisors to create a breastfeeding-friendly work environment. However, companies also have a role to play in providing structural and written breastfeeding facilities that can accelerate the formation of a breastfeeding-friendly culture in the workplace. The government and health workers also have a role in providing educational support to the community on exclusive breastfeeding.
Footnotes
Acknowledgements
The authors thank all participating textile companies and workers for their valuable contributions to this research.
Ethical Considerations
This study has obtained approval from the Ethics Committee of the Faculty of Public Health, University of Indonesia, with Approval Letter Number: Ket-101/UN2.F10.D11/PPM.00.02/2023. This step was conducted prior to data collection. The ethical review was conducted to ensure that the data collection process did not violate research ethical principles.
Consent to Participate
Before respondents agreed to participate in the study, the researchers requested their consent and explained that there was no coercion and that all information provided would be kept confidential and would not adversely affect respondents’ careers or personal lives.
Consent for Publication
All authors gave their consent to publish this manuscript.
Author Contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a grant from the Directorate of Research and Development, University of Indonesia with research contract number: NKB-716/UN2.RST/HKP.05.00/2023.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.