Use patterns of a food referral program for pregnant individuals: findings from the Mid-Ohio Farmacy

OBJECTIVE: Food and nutrition insecurity affects 1 in 4 pregnant individuals and increases the risk for adverse pregnancy outcomes.^1,2 Despite increasing calls to action to address food and nutrition insecurity in pregnancy,² there remains limited evidence that describe the use of prenatal clinic–based referrals to community-based food pantries to improve food security for pregnant individuals.³

The objective of this study was to describe the use of Mid-Ohio Farmacy (MOF) food pantries among pregnant and postpartum individuals with food insecurity.

STUDY DESIGN: We conducted a retrospective analysis of pregnant and postpartum individuals referred from high-risk prenatal clinics to the MOF from November 2019 to June 2022. The MOF is a partnership between a regional foodbank and local healthcare providers.⁴ Pregnant and postpartum individuals who reported unmet social needs as part of prenatal care were screened for food insecurity.⁴ Those who screened positive for food insecurity received an electronic referral to the food pantry.

The goal of the MOF is to connect individuals to fresh produce and healthy food as frequently as every week through a local food pantry. MOF participants can choose either a produce-specific food selection or a general box that primarily includes nonperishable items.

We compared the characteristics of individuals who accessed a food pantry with those who did not using chi-square or t tests. A multivariable logistic regression was used to examine the likelihood of accessing a food pantry based on patient characteristics; covariates were selected for inclusion based on a previously published conceptual model.⁵ We reported the results of this model as average marginal effects, which can be interpreted as the percentage change in the likelihood of accessing a food pantry.

RESULTS: Among 80 pregnant and postpartum individuals (70 pregnant; median gestational age, 18 weeks) with food insecurity who were referred to the food pantry as part of prenatal care, the mean age was 27 years, 49% self-identified as non-Hispanic Black, and 79% were Medicaid insured. Nearly two-fifths (39%) had accessed a food pantry before their pregnancy referral.

In total, 41% of individuals who were referred to the MOF accessed a food pantry at least once for a total of 448 food pantry visits. The average number of food pantry visits was 6 (standard deviation, 6.7).

In a multivariable analysis, existing food pantry users had a 31% higher likelihood of accessing a food pantry following a pregnancy referral than those without a history of accessing a food pantry (P=.03). Hispanic individuals had a 47% higher likelihood of accessing a food pantry than non-Hispanic White individuals (P=.01). Other characteristics were not associated with food pantry access (Table).

Table.

Characteristics of pregnant and postpartum individuals who accessed the Mid-Ohio Farmacy

Characteristic	Frequency, overall and by food pantry use n (%)			Likelihood of accessing a food pantry (n=80)
	Overall N=80	No n=47	Yes n=33	Marginal effecta (%)	95% Confidence interval	P value
Sociodemographics
Age (y), mean (SD)	27.1 (6.7)	26.5 (7.0)	28.0 (6.2)	0.5	−0.01 to 0.02	.5
Race and ethnicity
Non-Hispanic White	21 (26.2)	14 (29.8)	7 (21.2)	Ref	Ref	Ref
Non-Hispanic Black	39 (48.8)	25 (53.2)	14 (42.4)	5.5	−0.18 to 0.29	.7
Hispanic	11 (13.8)	4 (8.5)	7 (21.2)	46.9	0.09–0.85	.01
Non-Hispanic otherb	9 (11.2)	4 (8.5)	5 (15.2)	34.9	0.02–0.68	.04
Household size (no. of individuals), mean (SD)	2.4 (2.0)	1.8 (1.8)	3.2 (2.1)	3.5	−0.02 to 0.09	.2
Existing food pantry user
No	49 (61.3)	36 (76.6)	13 (39.4)	Ref	Ref	.03
Yes	31 (38.8)	11 (23.4)	20 (60.6)	30.9	0.03–0.58
Insurance type
Medicaid	62 (77.5)	37 (78.7)	25 (75.8)	Ref	Ref	.2
Private	18 (22.5)	10 (21.3)	8 (24.2)	17.8	−0.12 to 0.48
Trimester of pregnancy at food pantry referral
First	27 (33.8)	16 (34.0)	11 (33.3)	Ref	Ref	Ref
Second	27 (33.8)	15 (31.9)	12 (36.4)	8.1	−0.17 to 0.33	.5
Third	18 (22.5)	11 (23.4)	7 (21.2)	4.7	−0.24 to 0.33	.7
Fourth	8 (10.0)	5 (10.6)	3 (9.1)	8.9	−0.29 to 0.46	.6
Chronic comorbid conditions (not mutually exclusive)
Diabetes mellitus
No	72 (90.0)	44 (93.6)	28 (84.8)	Ref	Ref	.8
Yes	8 (10.0)	3 (6.4)	5 (15.2)	5.3	−0.33 to 0.43
Chronic hypertension
No	77 (96.2)	44 (93.6)	29 (87.9)	Ref	Ref	.3
Yes	3 (3.8)	3 (6.4)	4 (12.1)	21.4	−0.16 to 0.59
Obesity
No	39 (48.8)	25 (53.2)	14 (42.4)	Ref	Ref	.8
Yes	41 (51.2)	22 (46.8)	19 (57.6)	2.7	−0.18 to 0.24
Substance use disorder
No	69 (86.2)	40 (85.1)	29 (87.9)	Ref	Ref	.9
Yes	11 (13.8)	7 (14.9)	4 (12.1)	1.8	−0.29 to 0.32

Notes: *p < 0.05; **p < 0.01.

SD, standard deviation.

^aAverage marginal effect determined using a multivariable regression model adjusted for sociodemographic and chronic comorbid factors

^bOther included individuals who identified as Asian or Pacific Islander, American Indian, or Multiracial.

Walker. Uptake of food referral program among pregnant individuals. Am J Obstet Gynecol Glob Rep 2024.

A comparison of food pantry use between new and existing (ie, previous food pantry use before referral) users showed that existing users accessed a food pantry more frequently (mean visits, 7 vs 4; P<.01), completed more produce-specific visits (mean, 0.7 vs 0.1; P=.02), and visited food pantries over a longer period of time than new users (mean visits >90 days after referral, 6 vs 4; P=.01) (Supplemental Table 1). A comparison of food pantry use between prepregnancy vs postpregnancy referral showed no differences (Supplemental Table 2).

CONCLUSION: Only two-fifths of pregnant and postpartum individuals who were referred to the MOF through prenatal care accessed a food pantry. And those who did were more likely to be existing users with previous experience in accessing a food pantry before pregnancy and to be Hispanic individuals. The relatively low uptake in the current study is consistent with data from nonpregnant adults.^4,5

Limitations include a limited sample size and time period, a lack of pregnancy outcomes and patient-reported barriers to and facilitators of food pantry use, and a lack of data on other programs that addressed food and nutrition insecurity in pregnancy. More data are needed on how other related social needs, including transportation and housing instability, may impact the use of a food pantry and, more broadly, food insecurity during pregnancy, and such data may inform the development of future interventions.

A prenatal clinic–based referral program to a network of community-based food pantries may have the potential to address food and nutrition insecurity in pregnancy.

CRediT authorship contribution statement

Daniel M. Walker: Conceptualization, Data curation, Formal analysis. Jennifer A. Garner: Methodology, Writing – original draft, Writing – review & editing. Joshua J. Joseph: Conceptualization, Data curation, Writing – review & editing. Jiqiang Wu: Data curation, Formal analysis, Software, Validation. Amy Headings: Conceptualization, Funding acquisition, Methodology, Project administration. Aaron Clark: Conceptualization, Funding acquisition, Supervision, Writing – review & editing. Kartik K. Venkatesh: Formal analysis, Methodology, Project administration, Supervision, Writing – original draft.