Community Investments and Diet-related Outcomes: A longitudinal study of residents of two urban neighborhoods

Abstract

Introduction:

Investments in historically oppressed neighborhoods through food retail, housing, and commercial development are hypothesized to improve residents’ health, nutrition, and perceptions of their neighborhood as a place to live. Although place-based development (e.g., housing, retail, business assistance) is happening in many communities, there is little evidence on the long-term correlates of multiple investments such as health and nutrition among residents.

Methods:

We conducted a quasi-experimental longitudinal study using a cohort of randomly sampled households in two low-income, predominantly African American neighborhoods in Pittsburgh, PA, with surveys assessing residents’ food insecurity, perception of their neighborhood as a place to live, perception of access to healthy foods, and dietary outcomes in 2011 and seven years later (2018), with an interim assessment in 2014. Analyses conducted in 2022 compared changes among residents of one neighborhood which had 2.6 times the investments over a seven-year period with changes among residents of a socio-demographically similar neighborhood that received fewer investments.

Results:

We found that residents in the neighborhood receiving substantial investments demonstrated statistically significant improvements in neighborhood satisfaction (12.6 percent improvement compared with 2.2 percent decrease) and perceived access to healthy food (52 percent improvement compared with 18.2 percent improvement), and marginally significant change in food security (14 percent compared with 4.8 percent improvement) compared with residents in the neighborhood receiving fewer investments.

Conclusions:

Multiple place-based investments in neighborhoods can potentially induce positive change for residents in health and nutrition outcomes.

Introduction

Inequities in access to community-based amenities, especially in low-income urban areas, may be related to discriminatory development policies like residential redlining and racial bias in economic investment decisions. ^1–3 Community investment through improving resources and assets has been hypothesized to improve the health and nutrition of residents. For example, providing access to food retail has been posited to give opportunities for residents to consume healthier foods, and accessible green spaces have been theorized to provide opportunities to engage in physical activity and to achieve a greater sense of well-being. ⁴ Neighborhood assets and resources can come in the form of quality schools, housing, or better air quality. Such social and physical features may impact resident health and wellbeing including satisfaction with one’s neighborhood as a place to live.

Research has examined the effects of place-based investment, yet limited work has longitudinally examined health and nutrition outcomes among residents following multiple neighborhood investments over a substantial amount of time. This analysis builds upon and expands prior work that has examined associations between a Healthy Food Financing Initiative (HFFI)-funded supermarket’s opening and residents’ diet, obesity, neighborhood satisfaction, perceived access to fruits and vegetables and food insecurity, as well as associations between greenspace renovations and housing improvements with physical activity, psychological distress, sleep and cardiometabolic outcomes. ^11–16 Those prior studies spanned two- and three-year periods, focusing on data collected at baseline (prior to improvement) and just after improvements were implemented.

This analysis focuses on resident status seven years after baseline, looking at longer term changes and a larger set of investments of greater total value. The current and prior work employed a quasi-experimental longitudinal design comparing changes in an intervention and a matched comparison neighborhood that experienced more and less investment, respectively. We hypothesized that residents of the neighborhood with greater investments would experience more positive changes over this period.

Methods

Study Sample

Data come from the Pittsburgh Hill/Homewood Research on Eating, Shopping and Health (PHRESH) project, a quasi-experimental study which followed a representative cohort of randomly selected households from two sociodemographically matching Pittsburgh neighborhoods, Hill District and Homewood. At baseline in 2011, residents in both neighborhoods were predominantly African American with median household income of approximately $20,000,¹⁷ and comparable in terms of other neighborhood and resident characteristics, as established using American Community Survey data.¹⁷

As outlined previously,¹¹ the study surveyed a random sample of households in 2011 and conducted in-person interviews with the primary food shopper in 1,372 households. A larger sample was recruited from the intervention neighborhood, Hill District, than the comparison, Homewood, to facilitate analyses specific to understanding supermarket access¹¹ (897 from the Hill District, 475 from Homewood). In 2014, the team reinterviewed 831 households (571 in the Hill District, 260 in Homewood). In 2018, 597 households were reinterviewed (418 in the Hill District, 179 in Homewood). The current analysis included those completing a baseline (2011) interview and one or both follow-up interview(s) (603 from the Hill District; 272 from Homewood). Although the focus of this analysis was change from 2011 to 2018, using all available data improved efficiency (see Statistical Analyses, below). Of the 875 households, 553 completed both follow-up interviews; 44 completed only 2018; 278 completed only 2014 (see Appendix Exhibit 1). Another 15 participants were dropped from analysis due to missing dietary recalls for an analytic sample size of 860. Loss to follow-up in the Hill District was 32.8%, while loss to follow-up in Homewood was 42.7%. Households (n=497) were lost to follow up for the following reasons: deceased (n=76), moved too far away (n=122), physically or mentally unable to participate (n=31), study refusal (n=101), and unable to contact (n=167).

At each timepoint, participants completed a 60-minute interviewer-administered survey and a 24-hour dietary recall; a second dietary recall was obtained seven to 11 days later. Participants received compensation for participation. Interviewers living in or familiar with study neighborhoods (also, usually the same race/ethnicity as the participant) received training to conduct door-to-door recruitment and to enroll the primary food shopper in the household (if 18 years or older). Once enrolled, study participants received postcards, phone calls and invitations to town hall meetings to ensure continued engagement. All study protocols were approved by RAND’s Institutional Review Board (IRB).

Measures

To understand neighborhood-level investments between 2011 and 2018, data was gathered on the type, timing, and funding of all neighborhood-based developments that were at least partially publicly funded, through interviews and data requests to the Housing Authority of the City of Pittsburgh, the Pittsburgh Urban Redevelopment Authority, the Pennsylvania Housing Finance Agency, and the City of Pittsburgh. Investments in both neighborhoods consisted of a combination of housing (e.g., public housing improvements, construction of new mix-income and affordable housing), and non-residential projects (food retail, green space renovations, economic and commercial development). The intervention neighborhood received $255 million in investments compared with $99 million in the comparison. A table of all development costs can be found in Appendix Table 1.

Neighborhood satisfaction was assessed with the question, “All things considered, would you say you are very satisfied, satisfied, dissatisfied, very dissatisfied, or neutral - neither satisfied or dissatisfied with your neighborhood as a place to live?”¹⁸

Household food insecurity was assessed using the USDA validated 10-item Adult Food Security Survey Module, which assesses conditions and behaviors that are characteristic of households lacking financial resources to meet basic food needs (e.g., how often the participant’s household ran out of food and didn’t have money to buy more). Higher scores indicate higher food insecurity. We employed USDA scoring, classifying households as food insecure (low or very low food security (1)) or food secure (high or marginal food security (0)).¹⁹

Body mass index (BMI) (or weight in kg/height in m²) was calculated from interviewer-measured height and weight (respondents were measured without shoes). Interviewers measured height to the nearest eighth inch using a carpenter’s square (triangle) and an 8-foot folding wooden ruler marked in inches. Weight was measured to the nearest tenth of a pound using the SECA Robusta 813 digital scale.

Diet was assessed with two interviewer-administered 24-hour dietary recalls spaced seven to eleven days apart in 2011, 2014, and 2018, employing the ASA24 (Automated Self-Administered 24-Hour) Dietary Assessment Tool, which collects data on all food and beverages consumed in the 24 hours prior to data collection.²⁰ From the dietary recalls, we computed Healthy Eating Index-2010 (HEI-2010) scores to measure overall dietary quality based upon compliance with the United States Dietary Guidelines for Americans. We calculated a single HEI-2010 score at each survey wave based on the two recalls.²¹ HEI can range from 0 to 100, with higher scores indicating better quality of diet. We also calculated added sugar intake (teaspoons per day).

Perceived access to healthy foods was assessed through a series of 4 questions on a 5-point (strongly agree-strongly disagree) scale about the ease of buying, selection, quality, and price of fruits and vegetables in participants’ neighborhood.^22–24

Sociodemographic measures included race/ethnicity, age, gender, per capita household income, Supplemental Nutrition Assistance Program (SNAP) participation, marital status, educational attainment, any children in the household, access to a car (own or borrow), and years lived in the neighborhood.

Statistical Analysis

Between-group differences (i.e., Hill District versus Homewood) in baseline demographics, education, per capita household income, marital status, any children in the household, years lived in neighborhood, access to a vehicle, SNAP participation, and BMI were analyzed using t-tests for continuous variables and chi-squared tests for categorical variables. Next, linear mixed models were specified for all study outcomes (food insecure, dietary quality, added sugars, BMI, perceived access to healthy foods, and reported satisfaction with one’s neighborhood as a place to live), including a participant-level random intercept and covariates (gender, age, education, per capita household income, marital status, children in the household, and access to a vehicle). Interaction terms between neighborhood (Hill District=1 and Homewood=0) and time (year=2011, 2014, or 2018) were included in the model to test for significant differences in change over time between neighborhoods. ²⁵ Time (year) was coded so that the coefficient of the interaction term gives the estimated change between 2011 and 2018 in the Hill District relative to Homewood; specifically, we assigned these values: 2011=0, 2014=0.5, and 2018=1.

Analyses were performed using Proc Mixed in the statistical software SAS, version 9.4 (SAS Institute, Cary, NC). All analyses were completed in 2022. Mixed modeling was used to account for possible non-independence of the data because of repeated measurement within participants, and to deliver unbiased standard errors of the parameter estimates. ^26–28 In a comparison of the baseline and follow-up samples, we found that males, those with less than high school educational attainment, and those with access to a car were more likely to attrit. To address potential bias, all analyses included nonresponse weights. The nonresponse weights are calculated as the inverse of the probability of response estimated using logistic regression with socio-demographics and additional baseline characteristics as predictors. ²⁹

Results

Table 1 shows baseline characteristics of the analytic sample, which was predominantly African American, low-income, and on average middle-aged. There were significantly more female participants, less car ownership, greater neighborhood tenure, and higher neighborhood satisfaction in the Hill District relative to Homewood. We also show (unadjusted) baseline and follow-up values of study outcomes. Only satisfaction with one’s neighborhood as a place to live differed between the two groups at baseline–in the Hill District 66.2% of participants were satisfied or very satisfied, compared to 54.2% in Homewood. At follow-up this difference was even greater. Perceived access to healthy foods was comparable at baseline; at follow-up, the proportion of Hill District participants in agreement had increased by 25% relative to Homewood.

Table 1.

Attrition-Weighted Characteristics at Baseline and Outcomes at Baseline and Follow-up, By Neighborhood

Characteristic	Total (n=860)	Hill District neighborhood (Intervention) (n=591)	Homewood neighborhood (Comparison) (n=269)
Participant characteristics at baseline
Race/ethnicity
African American/Black (%)	94.9	94.0	96.7
Age in years, mean (SD)	53.3 (17.4)	52.7 (17.3)	54.6 (17.4)
Female (%)	75.3	78.0 *	69.9 *
Annual per-capita household income in $1000s, mean (SD)	13.6 (13.9)	13.3 (13.6)	14.1 (14.4)
SNAP participant in past year (%)	51.5	53.3	48.0
Marital status (%)
Married or living with a partner	18.1	16.1	21.8
Never married	43.1	45.2	39.1
Widowed, divorced, or separated	38.8	38.7	39.0
Educational attainment (%)
Less than high school	13.1	14.1	11.0
High school diploma	37.8	40.2	33.0
Some college or technical school	34.5	31.8	39.8
College degree	14.6	13.9	16.1
Any children in household (%)	28.7	28.7	28.9
Own or borrow a car (%)	56.6	53.8 *	62.0 *
Years lived in the neighborhood, mean (SD)	26.2 (23.1)	30.7 (23.8) ***	17.3 (18.3) ***
Outcomes at Baseline and Follow-up
Food insecure (low or very low food security, %)
Baseline (2011)	31.7	32.9	29.4
Follow up (2018)	21.0	20.0	23.1
Healthy eating index, mean (SD)
Baseline (2011)	48.5 (12.8)	48.4 (12.7)	48.5 (13.7)
Follow up (2018)	50.7 (12.4)	50.8 (12.1)	50.7 (13.2)
Daily teaspoons of added sugar, mean (SD)
Baseline (2011)	15.3 (11.3)	15.2 (10.8)	15.4 (12.4)
Follow up (2018)	13.3 (9.5)	13.3 (9.3)	13.4 (9.9)
Body mass index (kg/m2), mean (SD)
Baseline (2011)	30.4 (7.2)	30.3 (7.0)	30.6 (7.8)
Follow up (2018)	30.5 (8.2)	30.3 (7.3)	31.0 (9.9)
Easy to buy fruits and vegetables in my neighborhood (agree or strongly agree, %)
Baseline (2011)	17.5	16.5	19.5
Follow up (2018)	53.4	61.8 ***	36.1 ***
Satisfied or very satisfied with neighborhood as a place to live (%)
Baseline (2011)	62.1	66.2 ***	54.2 ***
Follow up (2018)	68.4	77.3 ***	50.1 ***

SOURCE Authors’ analysis of study data collected in the Pittsburgh Hill/Homewood Research on Eating, Shopping, and Health cohort.

NOTES Analysis adjusted for attrition weights. Note that 2018 values represent n=583 households. Significance tests used are Rao-Scott chi-squared test for categorical and t-test for continuous measures. Boldface indicates statistical significance (p<.05). Asterisks designate p-value limits (*p<0.05, **p<0.01, ***p<0.005).

Results from mixed modeling are shown in Table 2. A significant coefficient of the interaction term between “Follow-up period 2018” and “Hill District” is of primary interest implying different amounts of change in the Hill District relative to Homewood between 2018 and 2011. Two of these interactions were statistically significant: perceived access to healthy foods (beta = .34, 95% C.I. = (.23, .44), p < 0.001) and neighborhood satisfaction (beta=0.15, 95% CI = (0.05, 0.25), p = 0.004). There was a marginally significant coefficient for food insecurity. We show unadjusted results in Appendix Table 1.

Table 2.

Predictors of Change in Selected Outcomes for Study Participants from Baseline (2011) to Follow-Up (2018)

Parameter	Food Insecure a	HEI	Added Sugars	BMI	Easy to buy fruits and vegetables in my neighborhood	Neighborhood Satisfaction
Intercept	0.72 (0.61, 0.83) ***	44.05 (40.98, 47.13) ***	18.12 (15.55, 20.69) ***	31.92 (30.14, 33.69) ***	0.28 (0.17, 0.39) ***	0.31 (0.20, 0.43) ***
Neighborhood (ref=Homewood)	−0.01 (−0.07, 0.04)	0.93 (−0.59, 2.44)	−1.31 (−2.62, 0.01)	−0.69 (−1.49, 0.11)	0.07 (0.01, 0.12) *	0.14 (0.09, 0.20) ***
Year	−0.05 (−0.13, 0.03)	1.38 (−0.92, 3.69)	−1.62 (−3.51, 0.28)	0.57 (−0.89, 2.03)	0.18 (0.10, 0.27) ***	−0.02 (−0.11, 0.06)
Neighborhood*Year interaction	−0.09 (−0.19, 0.00)	0.72 (−2.09, 3.54)	−0.63 (−2.94, 1.68)	−0.66 (−2.44, 1.11)	0.34 (0.23, 0.44) ***	0.15 (0.05, 0.25) ***
Gender (ref=female)	0.01 (−0.04, 0.05)	−1.61 (−2.88, −0.34) *	3.48 (2.42, 4.53) ***	−2.67 (−3.41, −1.93) ***	0.05 (0.01, 0.10) *	0.01 (−0.04, 0.05)
Age in years	−0.00 (−0.01, 0.00) ***	0.14 (0.10, 0.18) ***	−0.09 (−0.12, −0.06) ***	−0.03 (−0.06, −0.01) ***	−0.00 (−0.00, −0.00) ***	0.00 (0.00, 0.01) ***
Education (ref=college degree)
Less than high school	0.01 (−0.06, 0.08)	−8.21 (−10.34, −6.07) ***	2.35 (0.57, 4.13) **	1.36 (0.11, 2.61) *	0.01 (−0.07, 0.09)	−0.03 (−0.11, 0.05)
High School	−0.04 (−0.10, 0.02)	−6.01 (−7.72, −4.31) ***	1.54 (0.12, 2.96) *	0.75 (−0.25, 1.75)	0.02 (−0.04, 0.08)	−0.00 (−0.06, 0.06)
Some college or technical school	−0.04 (−0.09, 0.02)	−4.04 (−5.71, −2.37) ***	0.99 (−0.40, 2.38)	1.37 (0.39, 2.35) **	−0.02 (−0.08, 0.05)	−0.05 (−0.11, 0.01)
Per capita household income in $1000s	−0.01 (−0.01, 0.00) ***	0.06 (0.01, 0.10) **	0.00 (−0.03, 0.04)	0.02 (−0.01, 0.04)	−0.00 (−0.00, 0.00)	−0.00 (0.00, 0.00)
Married	−0.08 (−0.13, −0.04) ***	1.53 (0.16, 2.89) *	−0.90 (−2.03, 0.23)	−0.77 (−1.56, 0.03)	−0.01 (−0.06, 0.05)	0.06 (0.01, 0.11) *
Children living in household	−0.08 (−0.13, −0.03) ***	−1.18 (−2.60, 0.25)	1.17 (−0.02, 2.35)	1.44 (0.61, 2.28) ***	0.02 (−0.03, 0.07)	−0.06 (−0.11, −0.01) *
Access to a vehicle	−0.05 (−0.09, −0.01) *	0.15 (−1.00, 1.29)	0.41 (−0.54, 1.36)	0.24 (−0.42, 0.91)	0.00 (−0.04, 0.04)	0.04 (0.00, 0.08)

SOURCE Authors’ analysis of study data collected in the Pittsburgh Hill/Homewood Research on Eating, Shopping, and Health cohort.

NOTES ^a Food insecure is binary; and reflects low/very food security (using standard USDA definition). A negative coefficient implies lower food insecurity (which is the same as greater food security). Ref = reference. Analyses were adjusted for attrition weights. Boldface indicates statistical significance (p<.05). Asterisks designate p-value limits (*p<0.05, **p<0.01, ***p<0.005). Main effect of “neighborhood” captures differences between Hill District and Homewood at baseline (2011). Main effect of “year” shows change in the reference neighborhood (Homewood) between 2011 and 2018. Interaction between neighborhood and year shows change in the intervention neighborhood (Hill District) compared with Homewood between 2011 and 2018.

Figure 1 provides side-by-side comparisons of the estimated change in Hill District participants (in blue) and Homewood participants (in red) using the covariate-adjusted regression model in Table 2. For the binary outcomes (food insecurity, neighborhood satisfaction, perceived access), the y-axis shows change in (absolute) percentage points. For the continuous outcomes (HEI, added sugars, BMI), the y-axis shows the amount of change as a percentage of baseline value. Participants in both neighborhoods experienced improvements in all outcomes except for neighborhood satisfaction, which decreased slightly among Homewood participants, and BMI, where we observed a small increase in Homewood. We observed significant relative improvements among Hill District participants in neighborhood satisfaction, and perceived access to healthy foods, and marginally significant relative improvement in food insecurity. Changes in dietary quality, added sugars, and BMI were similar between the two neighborhoods.

Figure 1: Overall Percent Change between 2011 and 2018 for Study outcomes.

NOTES Estimates of change are produced from mixed regression models that include covariate adjustment and attrition weights. Binary variables are percent change and linear variables are expressed as amount of change as a percent of baseline value (at 2011). The outcome food insecure indicates low/very food security (using standard USDA definition). A negative coefficient implies lower food insecurity (or more food security).

Food insecurity decreased by 14% among Hill District participants and by about 5% in Homewood. The relative difference between neighborhoods was 9%. Neighborhood satisfaction increased by 12.6% in Hill District and decreased by 2.4% among Homewood participants, a relative difference of 15%. In Hill District, ease of buying fruits and vegetables increased by 52 percentage points; the increase in Homewood was 18 percentage points.

Discussion

Across the U.S., there have been place-based efforts investing in historically oppressed communities to improve residents’ quality of life and health. From community development corporations who work on economic development, to housing developers who focus on promoting affordable housing, there is a hypothesized relationship between improving neighborhood assets and conditions and resident health. While recent research ³⁰ has begun to uncover the pathways between changes on the neighborhood-level and resident responses, adaptations, and ability to thrive ³¹, none to the team’s knowledge has followed a cohort of residents over a seven-year period to observe how they change as their neighborhoods change. Prior work with this study’s population examined short-term changes associated with community investments (i.e. one to two years). ^13,16,32 Additional years of follow-up data in this study allowed an examination of longer-term outcomes among residents following community investments that included a full-service supermarket, housing renovation and development, other commercial and economic development, and greenspace improvement in one neighborhood.

Between 2011 and 2018, residents in the Hill District neighborhood, receiving substantially more investments than Homewood, demonstrated greater improvements in neighborhood satisfaction and perceived access to healthy foods beyond that of Homewood neighborhood residents and a marginally significant trend toward greater reductions in food insecurity. While we observed decreases in food insecurity in both neighborhoods, there was a trend toward steeper decline among residents in Hill District. Results suggest that revitalization of the physical and economic environment may yield healthier communities.

Although residents demonstrated improvement in multiple outcomes, there may be unintentional harms associated with investing in neighborhoods, specifically caused by gentrification (i.e., the process by which disinvested neighborhoods experience a renewal, driven by increases in college-educated residents and upwardly trending housing prices). Other research by the research team in these neighborhoods demonstrated that improvements in social cohesion and neighborhood satisfaction were smaller when residents were living in a census tract (i.e., sub-neighborhood geographic area) undergoing gentrification ³³. To the extent that investments trigger gentrification, the improvements observed in the present study may be less likely to occur.

Original analysis of the supermarket’s one-year effect found that while there were initial diet and related health improvements after it opened, use of the new supermarket was only associated with neighborhood residents’ perception of access to healthy foods. ³² This raised the possibility that residents’ diet-related outcomes and wellbeing may have improved because the supermarket acted as an asset and resource to the community – signaling economic change,³⁴ fostering hope, and/or a sense of inclusion in or recognition by the broader society given the investment made.

In the present study, the relatively greater increase in perceived access to healthy foods in the Hill District neighborhood suggests that the investments played a role, either as community resources more broadly, or because the availability of healthy foods did indeed increase with the addition of a neighborhood supermarket. Smaller increases in the Homewood neighborhood may indicate secular trends in store offerings more generally. We observed dietary improvements during the seven-year period among both Hill District and Homewood residents.

Over the course of seven years, neighborhood satisfaction among residents in the Hill District significantly improved compared to residents in Homewood. Neighborhood satisfaction could be an important part of understanding the potential broader impact of neighborhood investments, including local job creation and economic growth, as well as community empowerment.³⁵

Limitations

Findings from this study of residents from two low-income, predominantly African American urban neighborhoods in Pittsburgh, PA may not generalize to neighborhoods with different socio-demographic profiles. This participant sample is older than the average age in neighborhood census data; this study enrolled the primary food shopper, so that female and older residents were more likely to be enrolled.

This study is further limited by its focus on two neighborhoods -- that may have differed at baseline and/or had different experiences over the seven years studied, beyond differential investment. This was controlled for as best as possible by choosing communities with similar observed characteristics at baseline (as measured by ACS data) with similar geographic, policy and social environments by virtue of their geographic closeness to one another, and by statistically controlling for baseline demographic characteristics. Nonetheless, there may be unmeasured and unrecognized differences influencing results. While true experimental research is probably impossible to bring to bear on the question at hand, it is one worth answering. Additional studies in other neighborhoods with investments are needed to provide the accumulation of evidence that might increase (or decrease) confidence in these findings.

In addition, prior research has shown that residents rely on resources outside of their neighborhoods. ^38–40 Not accounting for investments outside of the study neighborhoods is a limitation to this study. While not part of these analyses, data was collected on where residents did their main food shopping and reliance on and distance to supermarkets outside of the neighborhoods were comparable across both neighborhoods over time. That is, changes in and use of development that occurred outside of the neighborhoods, at least in terms of food access, was very similar.

Studying changes within neighborhoods is challenging, especially over a longer timeframe. While differential loss does not necessarily mean differential bias,⁴¹ there was 42.7% attrition in the comparison neighborhood (Homewood), compared with 32.8% attrition in the Hill District. It is possible that those residents who stayed in the neighborhood (and the study) produced a “resident survivor effect” (i.e., their responses may be more positive than residents’ who moved and thus attrit). While nonresponse weighting and covariate adjustment was included to address this limitation, elimination of all bias cannot be certain.

A variety of investments occurred in both neighborhoods, and while the magnitude of the investments through careful data collection can be quantified, disentangling the effects of each of the investments in the current study nor the exact mechanisms through which different neighborhood investments/development may impact outcomes is possible. The investments in both neighborhoods may have bolstered resident wellbeing and improved their heath behavior by directly influencing local socioeconomics.³⁴ Although the investments were greater in the intervention neighborhood, the specific kinds of improvements in both neighborhoods could have influenced the outcomes.

Conclusions

In this examination of a neighborhood with an HFFI-supported supermarket opening, there was not a long-term significant difference in dietary outcomes. However, improvements in neighborhood satisfaction and residents’ perceptions of access to healthy foods was observed. This suggests that policies such as HFFI are not a magic-bullet solution to poor diet or diet-related disease; however, the observed improvements in other outcomes warrant further consideration, and aligning such outcomes with equity and economic opportunity is a critical direction for public health policy research.⁴² Thus, understanding the process of neighborhood infrastructure investment and the health and wellbeing of residents is a critical public health need and challenge.