Abstract MP19: Academic-community Partnership To Evaluate The Effectiveness Of The Greater Boston Food Bank’s Healthy Pantry Program

Background: Food pantries are a significant nutrition source for food-insecure households. Traffic-light labeling is a new strategy in some pantries to help clients and staff identify healthier food choices. Healthy Pantry Program (HPP) is an online training designed by The Greater Boston Food Bank (GBFB) to teach staff of its affiliated pantries how to promote healthier choices by implementing traffic-light labels using Supporting Wellness at Pantries and behavioral nudges, such as placing healthy foods at eye level. Our academic research team partnered with GBFB to evaluate HPP by assessing if food pantries that participated in HPP increased their GBFB purchases of healthier (green and yellow) foods more than matched control pantries.

Methods: We conducted an observational study of 107 food pantries in the GBFB network in eastern Massachusetts, including 10 pantries that completed HPP training and 97 control pantries matched by organizational characteristics and baseline food purchases. Five HPP pantries participated during Wave 1 (Oct 2018-Jan 2019), and 5 participated during Wave 2 (May 2019-Aug 2019). Outcomes were changes in percentage of pantries’ monthly green and yellow purchases from GBFB at 6-month (Waves 1 and 2) and 12-month (Wave 1 only) follow-ups compared to baseline purchases (6 months pre-HPP). We used a difference-in-differences analysis to examine whether HPP pantries increased green-yellow food purchases more than control pantries.

Results: In Wave 1, HPP pantries had small but nonsignificant increases in green-yellow purchases at 6 and 12 months compared to matched controls (Table). In Wave 2, HPP and control pantries had similar increases in green-yellow purchases at 6 months.

Conclusions: An online training implemented by a food bank to improve healthy choices in food pantries did not significantly improve pantries’ purchases of healthier foods. Academic-community organization partnerships to evaluate health promotion initiatives can help guide effective program design and development.

Life expectancy and agricultural environmental impacts in Addis Ababa can be improved through optimized plant and animal protein consumption

In Ethiopia, children and adults face a double burden of malnutrition, with undernutrition and stunting coexisting with non-communicable diseases. Here we use a framework of comparative risk assessment, local dietary surveys and relative risks from large observational studies to quantify the health and environmental impacts of meeting adult and child recommended daily protein intakes in urban Addis Ababa. We find that plant-based foods, especially legumes, would have the lowest environmental impact and substantially increase life expectancy in adults, while animal-source proteins could be beneficial for children. This context-specific approach-accounting for regional constraints and trade-offs-could aid policymakers in developing culturally appropriate, nutritionally adequate and sustainable dietary recommendations.

Patient Characteristics Associated with Uptake of a Monthly Free Produce Distribution at a Health Center

Objectives

In recent decades it has been well established that social determinants have a profound impact on health. In response, some healthcare providers are implementing social service programs alongside traditional medical care. Research on these interventions, however, points to low patient uptake. We aim to understand patient characteristics associated with use of a monthly, free produce market at a health center.

Methods

The participants of this study are patients at a health system in Massachusetts, which operates a free, monthly produce market for patients and the public. There are no eligibility requirements for market registration. Participants were recruited in primary care waiting rooms and at the market. The analysis uses baseline surveys of study participants and administrative attendance records from the market. Risk ratios for attending at least one market were estimated using multivariable robust Poisson regression.

Results

Of the 712 patients enrolled in the study, 35% attended at least one market. After attending their first market, they attended 43% (IQR: 0%, 70%) of the following markets. Factors associated with attending at least one market include having at least one child in the household (Risk ratio: 1.32; 95% CI: 0.98, 1.78), having a monthly income below $2500 (1.39; 95% CI: 1.07, 1.80), being married (1.47; 95% CI: 1.06, 2.03), and being a homemaker (1.44; 95% CI: 1.07, 1.94) or unemployed (1.41; 95% CI: 0.99, 2.00). The magnitudes of the associations differ by the participant's language (English, Spanish, Portuguese).

Conclusions

Patients more likely to use a free produce market at a health center include those with lower incomes, dependent family members, and fewer work-related responsibilities. The use of the market by low-income patients supports a hypothesis that healthcare-based food assistance programs open to all will be used by those most in need. By eliminating eligibility requirements, health center produce markets reduce both the administrative burden on the health system and the stigma associated with use of food assistance. Qualitative research is ongoing to determine how to increase accessibility of the produce market to those who may have limited available time.

Funding Sources

This work is supported by SIREN UCSF, the Office of the Massachusetts Attorney General, and general operating funds from The Greater Boston Food Bank.

Issues in defining and measuring facial cleanliness for national trachoma control programs

The WHO trachoma control strategy includes promotion of facial cleanliness to interrupt transmission. However, scant data exist on measuring a clean face. Our goals were to determine reliable indicators of a clean face in a child, and disparities between the assessments of faces at a clinic versus at home. Five hallmarks of a clean face were assessed on 50 children in Tanzania. In five districts, 973 children (age 1-5 years) were evaluated for facial cleanliness at home, and again 2 days later at a central location for a trachoma examination. Data on environmental risk factors were collected. Only three signs, ocular and nasal discharge, and flies on the face, were reliable indicators of facial cleanliness (kappa>0.6); dust and food were less reliable. Unclean faces were more prevalent when measured at home (62%) than at the clinic (51%), although both were related to trachoma. The environmental markers of absence of latrines and tin roofs were consistent markers across all districts of households at risk of poor facial cleanliness. We conclude that for accurate estimates of clean faces, some assessment strategy outside the clinical environment is necessary. Additionally, behavior change programs in these districts in Tanzania should especially target families without tin roofs or latrines.