The Influence of Mental Health,Confidence,and Acculturation on Breastfeeding in Hispanic People

Abstract

Purpose

To examine breastfeeding behaviors among Hispanic women who participated in a postpartum nurse home visiting program with a focus on improving breastfeeding rates overall.

Design

An analysis comparing breastfeeding rates between Hispanic and non-Hispanic women and the impacts of anxiety, depression, breastfeeding confidence, and acculturation.

Setting

Yolo County, CA.

Subjects

Participants of the First 5 Yolo Welcome Baby: Road to Resilience home visiting program who completed both an initial and follow-up visit (N = 158).

Measures

Mothers self-reported breastfeeding behavior, breastfeeding confidence, anxiety, depression, ethnicity, insurance status, access to WIC, and acculturation (ie, language preference).

Analysis

Multinomial logistic regression with ridge regularization.

Results

Overall, anxiety and breastfeeding confidence were associated with more breastfeeding at 3 months postpartum. Anxiety had a greater negative impact on exclusive breastfeeding for non-Hispanic participants compared to Hispanic participants. High breastfeeding confidence was associated with increased exclusive breastfeeding at 3 months only for Spanish-preferring Hispanic participants.

Conclusion

Understanding health behaviors and resilience factors in Hispanic communities may inform interventions and policies in the US that improve breastfeeding rates overall.

Purpose

Despite structural barriers to good health, Hispanic people tend to live longer and have better health outcomes, a phenomenon known as the Hispanic Health Paradox. ¹ Breastfeeding promotes short- and long-term maternal and infant health,^1-3 and Hispanic women in the US have the highest rates of breastfeeding initiation and duration with Hispanic immigrant parents being more likely to breastfeed compared to those born in the US.^2-5 Acculturation is associated with lower breastfeeding rates among Hispanic people,^5,6 which may increase Hispanic health disparities. Thus, it is important to understand factors that may support breastfeeding and counter the effects of acculturation in this vulnerable population. For example, home visiting (HV) interventions are known to support breastfeeding within and outside of the US, especially in underserved populations.^7-9 The purpose of this study was to examine factors associated with breastfeeding maintenance among Hispanic women who participated in a postpartum nurse home visiting program in California.

Methods

Design and Sample

The UC Davis Internal Review Board approved this study and determined that there was no or minimal risk to subjects. We conducted a secondary analysis of cross-sectional data from First 5 Yolo Welcome Baby: Road to Resilience (Welcome Baby), a free and voluntary postpartum home visiting program for Medi-Cal-insured and uninsured women who delivered babies in Yolo County, CA between March 2022 and March 2023 (N = 252). Bilingual home visitors were available based on participant preference and surveys were offered and administered in participants’ preferred language. To examine breastfeeding maintenance, we included participants who had completed both an initial home visit at 1-2 weeks postpartum and a follow-up visit at home or via phone at 3 months postpartum (N = 188). Data collected via surveys included sociodemographics, mental health screenings, and breastfeeding information. Only participants who opted into screening for each of these categories and attended both the initial and follow-up visit were included in the final data set (N = 158).

Study Variables and Measures

Outcome Variable

The primary outcome was self-reported breastfeeding status at 3 months postpartum recorded as breastmilk only, some breastmilk, and no breastmilk.

Independent Variables

Five predictors were examined. Breastfeeding confidence was measured at 3 months postpartum as: confidence stayed the same, a little more confident, and much more confident. Maternal anxiety was measured at both timepoints using the Generalized Anxiety Disorder 7-item (GAD-7). ¹⁰ Maternal depression was measured also at both timepoints using the Patient Health Questionnaire-9 (PHQ-9). ¹¹ The PHQ-9 and GAD-7 are both reliable measures for depression and anxiety with Cronbach’s αs of .89 and .92, respectively.^10,11 Ethnicity was dichotomized as “Hispanic” and “non-Hispanic” based on self-reported ethnic identification upon enrollment. Language preference served as a proxy for acculturation among Hispanic participants ⁴ with those who preferred Spanish assigned to the lower acculturation group and those who preferred English to the higher acculturation group. Current literature informed covariate selection from available data (eg, access to WIC services, insurance status).

Analysis

Data cleaning and analyses were performed in R. ¹² Descriptive statistics summarized the sociodemographic characteristics. We used multinomial logistic regression to model the associations between breastfeeding status and maternal anxiety, maternal depression, and breastfeeding confidence at 3 months with a moderation analysis for Hispanic ethnicity. In a subgroup analysis of Hispanic participants, we used multinomial logistic regression to test associations between breastfeeding behavior and maternal anxiety, maternal depression, and breastfeeding confidence at 3 months with a moderation analysis for acculturation level. All models used ridge regression (a.k.a., L2 regularization) to account for small sample size and multicollinearity, increasing precision. ¹³ We used bootstrapping to extract 95% confidence intervals and determine statistical significance. ¹⁴ Models were built by progressively adding variables to see whether the effects of interest changed after controlling for other variables.

Results

Sample Characteristics and Descriptive Statistics

Over half of participants identified as Hispanic (N = 90, 57%). Remaining participants identified as Asian (N = 28, 18%), White (N = 14, 9%), Black/African American (N = 5, 3%), multi-racial (N = 4, 2%), and unknown or declined to state (N = 17, 11%). In the Hispanic subgroup, 34% preferred English and 66% preferred Spanish. The mean depression and anxiety scores postpartum were 2.17 (SD = 3.60) and 2.48 (SD = 4.27), respectively. At the first home visit, most participants were exclusively breastfeeding (N = 103, 65%), followed by some breastmilk (N = 43, 27%), and no breastmilk (N = 12, 7%). At the 3-month follow-up visit, fewer than half of participants were exclusively breastfeeding (N = 78, 49%), followed by some breastmilk (N = 51, 32%), and no breastmilk (N = 29, 18%). Most participants reported feeling “much more confident” (N = 93, 59%) in breastfeeding.

Primary Outcomes

Maternal Mental Health, Breastfeeding Confidence, and Breastfeeding Behavior

Anxiety was positively associated with exclusive breastfeeding at 3 months postpartum for all participants (OR = 1.27, 95% CI: [1.05, 1.47]). Breastfeeding confidence was also positively associated with partial (OR = 1.54, 95% CI: [1.04, 2.22]) and exclusive breastfeeding (OR = 2.01, 95% CI: [1.45, 2.72]) in the full model and preliminary modeling for all participants. The full model included all independent variables, moderated by ethnicity, and the selected covariates (ie, insurance status, WIC services). Compared to Hispanic parents, non-Hispanic parents with anxiety were more likely to use partial breastfeeding (OR = 1.30, 95% CI: [1.06, 1.69]). There was no statistically significant difference in breastfeeding confidence or breastfeeding behavior across the 2 groups.

Language Preference, Breastfeeding, and Breastfeeding Confidence for Hispanic Participants

Breastfeeding confidence had a significant positive association with exclusive breastfeeding in the preliminary and full model for both groups (OR = 1.81, 95% CI: [1.31, 2.41]). The full model included all independent variables, moderated by language preference, and the selected covariates (ie, insurance status, WIC services). Interestingly, there was a significant interaction between acculturation (ie, language preference) and high breastfeeding confidence where breastfeeding confidence had a stronger impact on breastfeeding for the Spanish-preferring group (OR = 1.51, 95% CI: [1.15, 1.92]). There was no statistically significant difference in breastfeeding behavior, depression, or anxiety across groups.

Discussion

Summary

Postpartum anxiety may have differential impacts on breastfeeding for Hispanic and non-Hispanic parents, and acculturation seems to moderate the relationship between breastfeeding confidence and breastfeeding for Hispanic mothers. Prior research demonstrates that anxiety negatively impacts breastfeeding initiation and duration. ¹⁵ Our findings suggest that interventions and policies to promote breastfeeding should incorporate mental health assessments and care referrals as needed and consider cultural context. Future studies should apply both quantitative and qualitative approaches to further elucidate how mental health, confidence, and race/ethnicity affect breastfeeding. Here, anxiety had a positive impact on breastfeeding for both Hispanic and non-Hispanic parents indicating that anxiety may have nuanced impacts on breastfeeding; future research is needed to better understand this relationship.

Strengths and Limitations

These results should be interpreted in the context of 3 limitations. First, our small sample size limits the generalizability and interpretability of our findings. Second, our sample size examining acculturation was halved, making it difficult to detect within-group differences. Third, our secondary analysis of data limited our ability to explore unmeasured factors (eg, age, education, income) that may also contribute to breastfeeding behaviors. This study also had many strengths. To our knowledge, this is the first study to examine maternal mental health, breastfeeding confidence, and acculturation on breastfeeding moderated by race/ethnicity. There is limited empirical research on what factors may influence health behaviors in Hispanic communities, and we identified mental health, breastfeeding confidence, and acculturation as important contributors. Additionally, by using ridge regularization in our modeling, we increased the stability of our models and reduced overfitting, increasing the reliability of our findings.

Significance

Of the policies and programs in the US that support breastfeeding, many are not applicable for Hispanic immigrants without papers. Learning more about unique resilience factors, buffers, or differences in how psychosocial factors influence health behaviors in different communities could benefit future intervention implementation. Given that Hispanic people are the largest growing ethnic group in the US, it is crucial to learn from, support, and include them in discussions around healthcare policy and legislation.

So What?

What is Already Known About This Topic?

Hispanic individuals in the United States have the highest breastfeeding rates and are the largest growing ethnic group in the US.

What Does This Article Add?

This article augments understanding of how mental health status, breastfeeding confidence, and acculturation may affect breastfeeding behaviors.

What Are the Implications for Health Promotion Practice or Research?

Understanding factors that contribute to breastfeeding behavior may inform policies and interventions aimed at maintaining breastfeeding practices, which may promote health among Hispanic individuals who experience disparities.