Study design, data collection and study population

This was a cross-sectional study conducted in seven WIC program locations in six states (i.e., Georgia, Massachusetts, Nevada, Pennsylvania, Wisconsin, Wyoming) and in one Indian Tribal Organization (ITO; Chickasaw Nation, located in Oklahoma) between July 2020 and August 2020. In the overarching study, data were collected among WIC mothers of young children and expectant mothers in their third trimester aged 18 years old or above as part of a baseline survey of a one-year breastfeeding campaign in the six states and the ITO [19]. The campaign builds on the proven strengths of a social marketing approach for breastfeeding promotion among WIC mothers. The campaign interventions include social marketing activities, the buddy program (a program that matches breastfeeding WIC mothers or expectant mothers whose babies were born or will be born around the same date to share experiences, encourage each other to breastfeed, and celebrate their children’s milestones) [20], and WIC clinic staff education. However, in this study, we restricted our study population to WIC mothers with infants aged 5 months old or less, as this is the age group of infants who are recommended to be exclusively breastfed according to the American Academy of Pediatrics [3].

Sampling strategy

Participants for the study were recruited using a convenient sampling approach. The WIC staff in the six states and the ITO administered the surveys either electronically through an online survey application or on paper to WIC-participating mothers. Responses to both online and paper surveys were de-identified. WIC staff members in each program location kept the list of WIC participants numbers onsite and assigned a project ID to participants for use on the surveys. While responses to the online survey application went directly into a survey application, responses to the paper surveys were entered into the survey application by WIC staff. Data were collected on self-reported intention to breastfeed (but reported after birth), breastfeeding-related practices, obstetric characteristics, and socio-demographics. Data also were collected on breastfeeding services and resources available through the WIC program.

Primary outcome

The primary outcome of interest is EBF, which was defined as the mother feeding her infant only breastmilk up through 5 months of age.


The predictors of interest in our study include laws supporting breastfeeding and WIC breastfeeding consultation. Employment status was assessed as an effect modifier.

Laws related to the promotion of breastfeeding in each state or ITO were assessed by extracting data from the US National Conference of State Legislatures (NCSL) on breastfeeding laws, which provides a summary of breastfeeding laws from 50 states, the District of Columbia, Puerto Rico and the Virgin Islands, as of September 2020 [21]. Program coordinators confirmed that laws in Chickasaw Nation were similar to those in Oklahoma, the state that surrounds Chickasaw Nation on three sides. We categorized laws supporting breastfeeding into five themes: (1) employers are encouraged or required to provide break time and private space for breastfeeding employees; (2) employers are prohibited from discriminating against breastfeeding employees; (3) breastfeeding is permitted in any public or private location; (4) breastfeeding is exempt from public indecency laws; and (5) breastfeeding women are exempt from jury duty (Additional file 1) [11]. The number of laws supporting breastfeeding in each state or ITO were summed to create a continuous score (i.e., a score of 1 for each theme and the total score range from 1 to 5). Breastfeeding laws in themes (1) and (2) were grouped together as employment-related laws. A binary indicator was created for whether the states or ITO had any employment-related breastfeeding laws or not. WIC breastfeeding consultation was defined as mothers who self-reported that they received breastfeeding information from a WIC program staff member, lactation consultant or peer counselor (yes, no). Employment status was also self-reported, and we dichotomized employment into any employment versus no employment. Mothers were considered employed if they reported being employed full-time or part-time.


There were both individual- and program-level covariates. Individual-level factors include mother’s age (years), education (< high school graduate, ≥ high school graduate), marital status (married/living with partner, single), has an adult support for baby at home (yes, no), current smoker (yes, no), and infant age (< 8 weeks, 8–16 weeks, > 16 weeks). Other individual-level factors include the mother’s previous history of breastfeeding (yes, no), parity (primiparity, multiparity), infant was delivered by cesarean section (yes, no), and infant was born premature (yes, no). Program-level factors include whether the mother received any breastfeeding promotional messages (yes, no), and whether she received breastfeeding information from any breastfeeding support group (yes, no).

Data analysis

Descriptive statistics of the study population were stratified by program locations. The overall prevalence of categorical variables, and the means and standard deviations of continuous variables are presented separately for each of the program locations.

We used four separate multivariable mixed models to conduct our analyses. Each of these models used modified Poisson regression with a robust variance estimator to estimate the prevalence ratio, and accounted for clustering by program location using a random intercept.

The first model assessed the relationship between the number of laws supporting breastfeeding in each program location and EBF, while adjusting for potential confounders, which include WIC breastfeeding consultation, individual- and program-level factors. Biologically plausible potential confounders were specified a priori and were controlled for in the model (Additional file 2).

The second model was used to estimate the relationship between program location’s employment-related breastfeeding laws and EBF practice. We introduced an interaction term (program location’s employment-related breastfeeding laws*employment status) in the model and assessed effect modification by employment status on the additive and multiplicative scales, while adjusting for the number of laws supporting breastfeeding, WIC breastfeeding consultation, individual-and program-level factors.

The third model assessed the association between WIC breastfeeding consultation and EBF practice, while adjusting for the number of laws supporting breastfeeding, employment status and individual- and program-level factors. The fourth model was similar to the third but included an interaction term between WIC breastfeeding consultation and employment status (WIC breastfeeding consultation*employment status). Effect modification was assessed on the additive and multiplicative scale, while adjusting for the same variables as in model 3.

Effect modification on the additive scale was assessed using the relative excess risk due to interaction (RERI). The additive scale can reliably identify the correct group of individuals to intervene in order to achieve a significant public health impact [22, 23]. Therefore, it is the most appropriate public health measure for assessing effect modification on the additive scale. RERI is an extra risk due to interaction, which is estimated as the difference between the (expected) effect based on the summation of the separate effects of the two predictors under study and the (observed) effect in the joint exposure category [24]. When RERI is equal to zero, it means no interaction, while a RERI greater than zero means positive interaction, and a RERI less than zero means negative interaction [25]. The RERI and corresponding 95% confidence intervals were estimated using publicly available SAS code [26]. On the multiplicative scale, an estimate of one means no effect modification, and an estimate of less than one signifies negative effect modification and an estimate greater than one signifies positive effect modification [27]. Our effect modification results were presented in a tabular format recommended by Knol and VanderWeele [27].

While we did not assess the role of each of the specific breastfeeding laws on EBF, we did conduct sensitivity analyses to compare the prevalence of EBF among mothers in program locations with laws encouraging or requiring employers to provide break time and private space for breastfeeding employees to mothers in program locations without such laws. We also compared the prevalence of EBF among mothers in program locations with laws that prohibit employers from discriminating against breastfeeding employees to mothers in program locations that do not prohibit such discrimination. We conducted sensitivity analyses to see whether the relationship between the number of laws supporting breastfeeding and EBF remained the same across different infant age (0–1, 2–3 & 4–5 months) groups. We also conducted sensitivity analysis to assess the relationship between a 10-unit increase in number of law-years in each program location and EBF. Number of law-years was defined as the number of total years of laws enacted in each state (Additional file 1). Data analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (