Race, Ethnicity, and Cultural Identity Modify Postpartum Participation in the Minnesota WIC Program

Abstract

Objective

To examine the racial, ethnic and cultural differences in postpartum participation of women who participated in the Special Supplemental Nutrition Program for Women, Infants and Children (WIC) during pregnancy by completing a retrospective analysis of observational data on 35,903 women who enrolled in Minnesota WIC during pregnancy, from April 2018 to March 2020.

Methods

Descriptive analyses were completed using chi-square tests of association to show differences in postpartum WIC participation by maternal demographics and health risk codes of the WIC participants. Binary logistic regression and multivariate logistic regression were used to obtain odds ratios to compare the likelihood of postpartum WIC participation across different races, ethnicities and cultural groups.

Results

Asian/Pacific Islander, East African, Hispanic, Hmong, Multigenerational Black, and Other Black pregnant participants were more likely than White participants to return to WIC postpartum (adjusted odds ratio (AOR) 2.54, 95% confidence interval (CI) 1.87–3.46; AOR 3.35, 95% CI 2.40–4.66; 1.30, 95% CI 1.10–1.54; AOR 6.76, 95% CI 4.39–10.42; AOR 1.40, 95% CI 1.11–1.77, AOR 1.52, 95% CI 1.26–1.83, respectively). American Indian pregnant participants were less likely than White participants to return to WIC postpartum (AOR 0.70, 95% CI 0.54–0.92).

Conclusions for Practice

These findings can help the Minnesota WIC program, as well as other WIC programs, better understand which cultural groups may need more specific outreach strategies to keep women participating in the program after giving birth. Further research is needed to understand why postpartum women choose to participate, or choose not to participate, in WIC.

Introduction

The Special Supplemental Nutrition Program for Women Infants and Children (WIC) is a federal program aimed at improving the health of low-income pregnant and postpartum women, infants and children (U.S. Department of Agriculture, 2013). WIC families receive supplemental nutritious foods, nutrition education and counseling, and referrals to health and social services. Women are eligible for WIC if they are pregnant or recently postpartum, are at nutritional risk, and meet income guidelines or participate in adjunctively eligible programs such as Medicaid. Most Minnesota participants learn about WIC from a family member or friend, a health care provider, social services or a financial assistance agency, with some differences by race/ethnicity (Lundmark, 2023). Hispanic and Black participants in Minnesota WIC are more likely to learn about the program through a health care provider than other races.

Evidence shows WIC can improve the health of women and children (Anderson et al., 2022; Sonchak, 2016; Venkataramani et al., 2022; Zimmer & Vernarelli, 2020). WIC participation results in fewer neonatal intensive care unit admissions and reduced risk for low birth weight, prematurity and infant mortality (Sonchak, 2016; Venkataramani et al., 2022). The American Hospital Association recommends WIC as a means of reducing maternal mortality, and can act as a gateway to healthcare, as well as provide nutrition support during the critical postpartum period (American Hospital Association, 2021; National WIC Association, 2019). WIC participation has also been shown to improve diet quality and decrease household food insecurity (Anderson et al., 2022; Zimmer & Vernarelli, 2020).

Despite these benefits, only half of eligible families participate in the WIC program (US Department of Agriculture, 2023a). Non-breastfeeding postpartum women have the highest participation rate (82%) compared to 60% of breastfeeding women. Participation rates declined between 2011 and 2021, indicating a need to identify enrollment barriers (US Department of Agriculture, 2023b).

This decline is set against a backdrop of differences in participation rates by race and ethnicity. Hispanic WIC-eligible individuals have the highest coverage rates, followed by Black WIC-eligible individuals, with White WIC-eligible individuals at the lowest coverage rates (US Department of Agriculture, 2023a). In Minnesota, coverage rates are higher than many other states, with 61.2% of WIC-eligible individuals participating.

Pregnant women are eligible for continued enrollment postpartum unless their household income changes, yet some women do not return to WIC after pregnancy. Few studies have examined factors that predict continuation of WIC participation into the postpartum period, and none have disaggregated broad racial categories into specific cultural identities. Cultural identity allows participants to choose how they identify culturally within a race, as health practices can vary greatly within race by culture. For example, Black participants might identify as Somali American, or multigenerational Black American. Data disaggregation is critical to identify and address gaps in public health efforts (Kauh et al., 2021). Given the established evidence that WIC participation varies across races and ethnicities, we hypothesized that postpartum participation in WIC among women who participated in pregnancy would also vary by race and ethnicity. By investigating the differences in WIC participation across cultural identities, programs can identify specific populations that may benefit from additional outreach. Using data from the Minnesota WIC program, our study aimed to quantify differences in postpartum WIC participation among Minnesotans who participated in WIC prenatally by cultural identity.

Methods

Data Collection

The data for this secondary data analysis were obtained from the Minnesota Department of Health WIC Program and combined with Minnesota birth records. These are administrative data and therefore the study did not require Institutional Review Board approval. The data included pregnant women who entered the WIC program between April 1, 2018 and March 31, 2020. These years were chosen as they were the most recent data available prior to the coronavirus pandemic. Birth records and postpartum WIC records were matched with pregnancy records. Records for multiple infants were removed to retain only one postpartum record, leaving a dataset of 38,176 records. The data were then examined for implausible values, which removed 386 records. Next, for women with more than one pregnant certification within the two-year period, records were randomly removed so that only one record remained per individual (2,048 records). Additionally, records with unknown race and ethnicity were removed (19 records). The final dataset included 35,903 records.

Variables

The dependent variable for this study was postpartum WIC participation, determined by attendance at a postpartum WIC certification appointment. Minnesota birth records were matched to participants with a postpartum WIC record.

Cultural identity category was the primary independent variable. Minnesota WIC clients identify their race as Black/African American, Asian, Native Hawaiian or Pacific Islander, American Indian or Alaskan Native, and/or White. They are also asked to choose their ethnicity: Hispanic or Latino/a or not Hispanic or Latino/a. Cultural identity is asked of WIC participants who identify as non-Hispanic Black/African American or as non-Hispanic Asian. Choices for cultural identity include: Black Multigenerational American, Somali, Liberian, Sudanese, Ethiopian, Kenyan, Oromo, Other- Black, Hmong, KaRen, Bhutanese-Nepali, Cambodian, Karenni, Laotian, Vietnamese, and Other- Asian. For this analysis, Somali, Sudanese, Ethiopian, Kenyan, and Oromo were grouped under “East African”. Black individuals who did not have a cultural group listed, or individuals identifying culturally as Liberian or Other- Black were classified as “Other Black”. Hmong was the only distinct Asian cultural group used for analysis due to the low frequency of other Asian-identifying cultural groups, who were grouped in “Asian/Pacific Islander”. Native Hawaiian/Pacific Islander were also included under “Asian/Pacific Islander”. Individuals who identified as more than one race were counted in a single category, prioritized in the following order for this analysis: East African, Hmong, Multigenerational Black, Hispanic, Native American/American Indian, non-Hmong Asian with Native Hawaiian/Pacific Islander, Other – Black, and White. If cultural identity was not obtained at WIC, data from the birth records were used (917 records).

Geographical area of residence was included for analysis. These categories were determined using the National Center for Health Statistics Urban-Rural Classification Scheme for Counties (Centers for Disease Control and Prevention, 2017). The 87 Minnesota counties were categorized as urban (Large Central Metro, Large Fringe Metro, Medium Metro or Small Metro), or rural (Micropolitan and Non-Core).

Age, calculated as age at expected delivery date, was used as a continuous variable for the multivariate analysis but as a categorical variable for the descriptive analysis. Age was divided into 5-year increments from 20 years through 40 years or older.

The variable primigravida was defined as women who reported the pregnancy as their first time being pregnant. Pre-pregnancy weights were self-reported; height, weight and hemoglobin were collected at least once during pregnancy. We created a two-level categorical variable defined as gestational weight gain within or not within (greater or less than) the Institute of Medicine gestational weight gain guidelines (Institute of Medicine (US) and National Research Council (US) Committee to Reexamine IOM Pregnancy Weight Guidelines, 2009). Low hemoglobin was included in the analysis. Gestational diabetes mellitus (GDM), diabetes mellitus (DM), and hypertension or prehypertension were self-reported at WIC. The variable for smoking was indicated if the participant reported currently smoking cigarettes at their certification appointment. Alcohol or substance use was assigned as a WIC risk code if a woman reported drinking any alcohol or using any substances/drugs not prescribed by her doctor currently in her pregnancy. Multifetal gestation was assigned if a woman self-reported being pregnant with multiples. Pregnancy at a young age was assigned if a woman became pregnant at age 17 or younger. Lastly, the variable “High Risk- Other” was included; this indicates that WIC staff coded the participant as “High Risk” but did not assign a specific “High Risk” code.

Data Analysis

Descriptive analyses were completed using chi-square tests of association to test differences in maternal characteristics by postpartum WIC participation. Binary logistic regression was used to obtain an unadjusted odds ratio assessing the relationship between cultural identity and likelihood of postpartum WIC participation. White was used as the reference group because this was the largest group in this dataset. A multi-variable binary logistic regression model was conducted to find adjusted odds ratios and 95% confidence intervals. Variables were first tested for correlation. If two variables had a correlation of greater than 0.50 the one considered most relevant based on a review of the literature was chosen for inclusion in the model. After the initial full regression model was run, variables were removed one by one if the p-value was > 0.05, watching for significant changes (> 0.1) in odds ratios upon removal. Interactions (cultural identity by smoking status, age, weight gain, urban, GDM, DM, and primigravida) were added to the regression model and kept if p-value was < 0.05 and inclusion resulted in significant change (> 0.1) in parameter estimates. All analyses were completed in SAS Studio version 9.04.01.

Results

Maternal Characteristics

Almost 90% of individuals who participated in Minnesota WIC prenatally returned postpartum (Table 1) The largest race/ethnicity group in the study population was White (40.6%) followed by Black/African American (28.6%), Hispanic (15.1%), Asian/Pacific Islander (11.6%), and American Indian/Alaskan Native (4%). After dividing races into cultural groups, 6.1% of individuals identified as Hmong, 8% identified as East African, and 7.5% identified as Multigenerational Black. 13.1% of participants were in the Other-Black category.

Table 1.

Characteristics of women enrolled in the Minnesota WIC program who participated prenatally from April 2018 to March 2020

		All women n (%)	Participated Postpartum n (%)	Did Not Participate Postpartum n (%)	Chi Squarea	P-value < .05
Sample		35904	32225(89.8)	3679(10.3)
Characteristic
Race/Ethnicity					< .001	*
	Hispanic	5414 (15.1)	4854 (15.1)	560 (15.1)
	American Indian	1445 (4)	1209 (3.8)	236 (6.4)
	Asian/PI	1975 (5.5)	1796 (5.6)	179 (4.8)
	East African	2875 (8)	2773 (8.6)	102 (2.8)
	Hmong	2194 (6.1)	2141 (6.6)	53 (1.4)
	Multigenerational Black	2698 (7.5)	2400 (7.5)	298 (8.1)
	Other Black	4711 (13.1)	4186 (13)	525 (14.2)
	White	14591 (40.6)	12843 (39.9)	1748 (47.2)
Geographic location					< .001	*
Urban	Large Central Metro	12755 (36.3)	11579 (36.2)	1176 (37.4)
	Large Fringe Metro	7538 (21.4)	6846 (21.4)	692 (22)
	Medium Metro	1313 (3.7)	1154 (3.6)	159 (5.1)
	Small Metro	4183 (11.9)	3816 (11.9)	367 (11.7)
Rural	Micropolitan	5127 (14.6)	4720 (14.7)	407 (12.9)
	Noncore	4233 (12)	3888 (12.1)	345 (11)
Age group					< .001	*
	Under 20	2451 (6.9)	2165 (6.7)	286 (9)
	20–24	8816 (24.9)	7863 (24.4)	953 (30.1)
	25–29	10556 (29.8)	9675 (30)	881 (27.8)
	30–34	8222 (23.2)	7622 (23.7)	600 (19)
	35–39	4233 (12)	3910 (12.1)	323 (10.2)
	40+	1089 (3.1)	967 (3)	122 (3.9)
Primigravida		8252 (23)	7266 (22.6)	986 (26.6)	< .001	*
Smoking		3525 (9.8)	3062 (9.5)	463 (12.5)	< .001	*
Underweight		943 (2.6)	836 (2.6)	107 (2.9)	0.288
Gestational Weight Gainb		21923 (61.1)	19752 (61.3)	1530 (58.7)	0.002	*
Low Hemoglobin		4577 (12.7)	4167 (12.9)	410 (11.1)	0.001	*
Hyperemesis Gravidarum		313 (0.9)	294 (0.9)	19 (0.5)	0.013	*
Gestational diabetes		1021 (2.8)	965 (3)	56 (1.5)	< .001	*
Pregnancy at a Young Age		1013 (2.8)	912 (2.8)	101 (2.7)	0.720
Multifetal Gestation		577 (1.6)	520 (1.6)	57 (1.5)	0.732
Pregnant & currently breastfeeding		466 (1.3)	416 (1.3)	50 (1.4)	0.763
Gastrointestinal disorders		190 (0.5)	159 (0.5)	31 (0.8)	0.006	*
Diabetes Mellitus		508 (1.4)	425 (1.3)	83 (2.2)	< .001	*
Hypertension or Prehypertension		650 (1.8)	568 (1.8)	82 (2.2)	0.051
Other Medical Conditions		460 (1.3)	413 (1.3)	47 (1.3)	0.949
Alcohol & Substance Use		259 (0.7)	220 (0.7)	39 (1.1)	0.012	*
Median (interquartile range)
Age at estimated due date		28(9)	28(9)	26(9)	< 0.001 c	*

^a:Chi-Square test of association conducted using characteristic by postpartum WIC participation

^b:Variable present if participant gained weight outside of Institute of Medicine gestational weight gain guidelines

^c:P-value of t-test comparing the mean age of two groups: participated postpartum and did not participate postpartum

The majority of participants lived in urban (73.3%) versus in rural (26.5%) areas, with 36.3% of participants living in a Large Central Metro county. The largest age group was 25–29 year-olds (29.8%). Participant characteristics and differences in postpartum participation are reported in Table 1.

Participation Characteristics

Before adjustment for confounders or assessment for interactions, there were clear differences by cultural identity in postpartum WIC participation. See Table 2 for unadjusted odds ratios (OR) for each cultural identity. (Table 2) Compared to the largest group (White participants), Asian/Pacific Islander (OR 1.37, 95% confidence interval (CI) 1.16–1.61), East African (OR 3.70, 95% CI 3.01–4.54), Hispanic (OR 1.18, 95% CI 1.07–1.31), and Hmong (OR 5.50, 95% CI 3.72–6.76) participants were more likely to participate postpartum. American Indian participants were less likely to participate postpartum compared to White participants (OR 0.70, 95% CI 0.60–0.81). Multigenerational Black and Other Black participants were not significantly different than White participants in postpartum WIC participation.

Table 2.

Odds ratios: participation in Minnesota WIC postpartum by cultural identity

	Unadjusted Odds Ratio	95% Confidence Interval	Adjusted Odds Ratio	95% Confidence Interval
American Indian vs White	0.70*	0.60, 0.81	0.70*	0.54, 0.92
Asian/PI vs White	1.37*	1.16, 1.61	2.54*	1.87, 3.46
East African vs White	3.70*	3.01, 4.54	3.35*	2.40, 4.66
Hispanic vs White	1.18*	1.07, 1.31	1.30*	1.10, 1.54
Hmong vs White	5.50*	3.72, 6.76	6.76*	4.39, 10.42
Multigenerational Black vs White	1.10	0.87, 1.17	1.40*	1.11, 1.77
Other Black vs White	1.09	0.98, 1.25	1.52*	1.26, 1.83
Asian/PI vs Hmong	0.25*	0.18, 0.34	0.38*	0.22, 0.63
East African vs Multigenerational Black	3.38*	2.68, 4.25	2.40*	1.63, 3.53

^*significant at 95% confidence level

Odds ratios at mean age 27.91, weight gain within Institute of Medicine guidelines

Model adjusting for primigravida versus multigravida, geographical area of residence, gestational diabetes, diabetes mellitus, and gestational weight gain

The adjusted odds ratios (AOR) are reported in Table 2. The odds of participation among Other Black participants (AOR 1.52, 95% CI 1.26, 1.83) and Multigenerational Black participants (AOR 1.40, 95% CI 1.11, 1.77) compared to White increased and reached significance after adjusting for multigravidity, rurality, GDM, DM, and gestational weight gain.

There were also differences, in both the unadjusted and adjusted models, within the Black and Asian racial groups. East African women were 2.4 times more likely than Multigenerational Black women to participate postpartum (AOR 2.40, 95% CI 1.63–3.53). Asian/Pacific Islander participants were 68% less likely than Hmong participants to remain in WIC postpartum (AOR 0.38, 95% CI 0.22, 0.63).

Smoking decreased likelihood of returning to WIC for all cultural identities. However, since less than 10% of participants overall were smokers, and many cultural identities had even lower numbers of women who smoked, the interaction term was not included in the final model. Removing the interaction term did not drastically change the odds ratios for the adjusted model.

By adding the interaction term for gestational weight gain by cultural identity, gestational weight gain became mostly not significant; only Asian/Pacific Islander participants had a significant difference for postpartum return rate, between weight gain within IOM guidelines and outside IOM guidelines.

Table 3 shows the adjusted odds ratios for other variables included in the final model (Table 3) Primigravid women, those living in large urban counties, smokers, women without GDM, and those self-reporting DM had lower postpartum participation rates. There were no significant interactions, other than smoking and weight gain, between cultural identity and other maternal characteristics. Risk codes found to be not significant (p-value > 0.5) included: underweight, multifetal gestation, gastrointestinal disorders, central nervous system disorders, low hemoglobin, hypertension or prehypertension, alcohol and substance use, other medical conditions, and “High risk- other”.

Table 3.

Odds ratios: variables associated with participation in Minnesota WIC postpartum after adjustment for cultural identity

	Adjusted Odds Ratio	95% Confidence Interval
Primigravida vs Multigravida	0.82*	0.75, 0.90
Urban vs Rural	0.66*	0.60, 0.72
Gestational Diabetes	1.64*	1.23, 2.18
Diabetes Mellitus	0.67*	0.51, 0.89
Smoking	0.79*	0.70, 0.89
Gestational Weight Gain
American Indian	1.08	0.90, 1.20
Asian/PI	0.83*	0.68, 0.96
East African	1.24	0.99, 1.42
Hispanic	1.11	0.97, 1.18
Hmong	0.99	0.74, 1.21
Multigenerational Black	0.95	0.80, 1.03
Other Black	0.99	0.86, 1.07
White	1.04	0.98, 1.10

^*significant at 95% confidence level

Discussion

The purpose of this analysis was to examine factors that may be correlated with postpartum WIC participation, particularly cultural identity. To our knowledge, this is the first study that has analyzed postpartum WIC participation rates by race and ethnicity, although several studies have analyzed prenatal WIC participation rates. Postpartum participation rates are high nationwide, relative to child and pregnant women participation rates, likely due to the high infant participation rates (US Department of Agriculture, 2023a). This analysis also found a high postpartum return rate overall, with statistically significant differences in postpartum participation among cultural identities; these differences remained after adjusting for other variables. The unadjusted and adjusted odds ratios did not differ greatly, indicating that disparities remained among cultural identities that are not explained by variables such as age, health conditions, or geographic location. This study has similar findings to prenatal WIC research in that Hispanic women have higher rates of WIC participation than their non-Hispanic White counterparts (Angley et al., 2018; Bitler et al., 2003; Edmunds et al., 2014; Jacknowitz & Tiehen, 2010; Liu & Liu, 2016; Gray et al., 2019; Swann, 2007; Tiehen & Jacknowitz, 2008). Similarly, children of Hispanic mothers are more likely to enter WIC than children of non-Hispanic mothers (Castner, Mabli & Sykes, 2009).

Other studies have found that Black/African American women and White women participate in WIC at different rates (Angley et al., 2018; Bitler et al., 2003; Brien & Swann, 2001; Edmunds et al., 2014; Gueorguieva et al., 2009; Joyce et al., 2008; Liu & Liu, 2016; Gray et al., 2019; Sonchak, 2016; Tiehen & Jacknowitz, 2008). After adjustment for variables, we did find that Multigenerational Black and Other Black were 1.4 times and 1.5 times more likely, respectively, than White participants to return postpartum. There were notable differences among East African participants and White participants; East African women were over three times as likely as White women to return postpartum.

Another finding was that American Indian participants have the lowest rates of returning to WIC postpartum. Shefer and Smith (2004) found that American Indian/Alaskan Native children have higher rates of dropping out of WIC, compared to children who are Hispanic or Asian. White and Black children also had higher WIC drop-off rates, compared to Hispanic or Asian children, which follows the same trends observed in our study (Shefer & Smith, 2004).

The results presented here can benefit the Minnesota WIC Program by better understanding differences in population sub-group WIC participation during the postpartum period. Further research into factors explaining higher postpartum participation among Hmong and East African communities and examining barriers to postpartum participation among White and American Indian communities is warranted.

There are many factors that influence why people may or may not return after their pregnancy ends. Lack of transportation, paid parental leave, and childcare are structural barriers that reduce postpartum participation, whereas unintended pregnancy and lack of social support may increase WIC participation (Liu & Liu, 2016). Difficulties with understanding eligibility criteria and benefits, gathering proper documentation, lack of understanding of redemption procedures and frustrating interactions with government programs are common administrative barriers (Davis et al., 2022). Stigma is a social barrier to participation. When surveying women about why they did not enroll in WIC during their pregnancy, the most common reasons reported were thinking they did not qualify for WIC and believing they did not need WIC (Westmark, 2016).

According to a study by Westmark (2016), for women who did not participate in WIC prenatally because they thought they did not need WIC, many intended to exclusively breastfeed. WIC provides formula in the infant food package, if needed. Some women who exclusively breastfeed may not see the value of participating in WIC postpartum. Generally, when a woman returns to enroll her infant in WIC, she is also recertified as a postpartum woman. In Minnesota WIC, Hmong infants are less likely to be breastfed than other infants (Minnesota Department of Health, 2018). One hypothesis is that Hmong women have the highest postpartum WIC participation rates because they have high formula use rates. East African infants have high breastfeeding initiation rates, but also have high rates of formula supplementation, potentially contributing to their high postpartum WIC rates. Other studies have found that women enrolled in WIC were less likely to exclusively breastfeed than eligible non-participants (Francescon et al., 2016; Ziol-Guest & Hernandez, 2010). Increased promotion prenatally around the breastfeeding support and breastfeeding food packages that WIC provides may encourage exclusively breastfeeding women to return to WIC postpartum.

This analysis found that American Indian women had the lowest rates of returning to WIC postpartum. American Indians in Minnesota experience disparities among social determinants of health (Minnesota Compass, 2021). American Indians are the most likely of any race or ethnicity in Minnesota to be below the poverty line, to be without health insurance, and to have a disability. In 2021, 24% of American Indians were below the poverty line in Minnesota, compared to 9.2% of the statewide population. Additionally, despite making up less than 1% of the population, 17% of Minnesotans experiencing homelessness are American Indian (Institute for Community Alliance Minnesota, 2023). Although this would seem to make their need for WIC greater, extreme poverty and/or homelessness can create additional barriers to attending WIC appointments, perhaps contributing to their higher rates of not returning postpartum. In 2021, American Indian WIC participants in Minnesota had the highest rates of not redeeming any WIC benefits in a month (Minnesota Department of Health, 2021). This could indicate that American Indian participants see less value in the WIC foods they receive, or that they have difficulty shopping, which discourages them from participating. Targeted efforts, such as interviews or focus groups within this population, is vital to help inform how to encourage their access to this important safety net.

This study has many strengths. It is perhaps the first study to examine differences in postpartum return rates among WIC participants using this method of categorizing participants by cultural identity. This study highlights the importance of looking beyond traditional racial categories such as “Black/African American” and “Asian” as there are varied cultural practices within these groups, providing an opportunity to create targeted outreach strategies. For example, disaggregating the “Asian” category into Hmong and non-Hmong Asian made it clear that efforts to increase postpartum participation should focus on non-Hmong Asian communities. In addition, the American Indian group included multiracial participants, an important improvement over single-race categorization given that, in 2020 census data, almost two-thirds (63%) of children classified in any way as American Indian or Alaskan Native were reported as biracial or multiracial (National Indian Child Welfare Association, 2023).

There are several limitations to this paper. We do not know the outcome of the pregnancy for the women that did not return to WIC. Women who had a miscarriage, stillbirth, or neonatal death may have been less likely to return. Despite being eligible after a pregnancy loss, many individuals may not know this, or they may not see the benefit in participating without an infant on the program. It also could be emotionally difficult for women to think of returning to WIC without a baby, or to discuss their loss with WIC staff. When examining the different cultural identity groups, the Other- Black category is a limitation as we do not know the cultural communities these participants identify with. This makes it difficult to target outreach to this group. Lastly, it is important to note that some women do not return postpartum due to ineligibility, such as an increase in income or moving out of state.

Conclusion

This study found significant differences by cultural identity in the odds of postpartum WIC participation among Minnesota women who enrolled in WIC during pregnancy between 2018 and 2020. Hmong and East African women had the highest rate of postpartum WIC participation, while American Indian and White women had the lowest rates of postpartum WIC participation. Among Asian/Pacific Islander women, Hmong women had the highest participation rates, and among Black/African American women, East African women had the highest participation rates. To our knowledge, this is the first report of differential postpartum WIC participation by cultural identity groups. Efforts to improve program retention should consider cultural identity as a factor that could be used to better target outreach efforts. Educating clients on the offerings of a fully breastfeeding WIC food package, and of WIC culturally-specific breastfeeding peer counseling, is one potential area for improvement. Additionally, efforts to limit barriers to participation for American Indian families are needed. Further research could shed light on women’s reasons for foregoing the many benefits of continued participation in WIC after they deliver their baby. Given the challenge of WIC participation declining, understanding these trends will be especially pertinent to ensuring an equitable and accessible program for all.

Significance.

It is already known that culture, race and ethnicity influence the likelihood of WIC program participation. Research also has shown that WIC participation can have positive impacts on the health of women, infants and children.

This research adds depth to previous findings on prenatal WIC participation by describing the differences in postpartum WIC participation by race, ethnicity and cultural group. This research can be used for outreach planning within the WIC program.

Data Availability

Minnesota WIC individual-level data cannot be shared without a data use agreement.