Healthy Families: a Cluster Randomized Study to Change Obesity in Public Housing

Purpose and design The purpose of this study was to test the effects of a environmental-level package of interventions on public housing residents' obesity and obesity-related behaviors. The design of the study and its intervention have been previously published . Simply, this was a cluster randomized trial, with public housing developments (PHDs) serving as the unit of randomization and analysis. This document describes the patterns of change in the main behavioral and weight outcomes for adult women before and after the intervention.

Participants Public housing in Boston is administered by the Boston Housing Authority (BHA), a public agency that provides subsidized housing to low- and moderate-income individuals and families, disabled individuals, and elderly individuals. There are 64 public housing developments, 37 are designated as elderly/disabled developments and 27 are designated as family developments. Approximately 27,000 people are housed under the public housing program.

Family (vs. elderly) designated PHDs with more than 200 residents which were not undergoing renovations requiring residents to move out of the development for a period of time in the City of Boston were eligible to participate in this study (n=24). The investigators recruited 10 developments to participate in the Healthy Families study; 5 serving as intervention PHDs and 5 serving as control PHDs. PHDs were randomly assigned to either condition, in matched pairs for size of development and existence of health activities in the development. PHDs randomized to the intervention group received all intervention components (see Intervention section below) and developments randomized to the control group did not receive any intervention components.

Data collection In both intervention and control group PHDs, a randomly sampled group of female residents and their daughters ages 8-16 were recruited into an evaluation cohort to examine study outcomes. The investigators selected mothers because in the family developments over 80% of heads of household (per aggregate data compiled from tenant agreements) were women. The investigators selected daughters aged 8-15 as development of obesity often occurs during this time period for females. To be eligible for this study, participants were required to be female, age 18-72, live in one of the recruited PHD, and planning to remain so for two years, had responsibility for a girl age 8-15 (also living in the public housing residence), were English or Spanish speaking, and were able to make changes to their diet and physical activity habits if desired. Exclusion criteria included: the adult was not able to complete the survey tools or was not interested in participating. All study materials were available and used in both English and Spanish.

Survey assistants approached randomly selected apartment units within each of the 10 housing developments. Using a standardized protocol, survey assistants assessed individual's interest in participating and eligibility. If interested and eligible, the participant provided their written informed consent to participate, girls aged 12-15 provided written assent, and girls aged 8-11 gave verbal assent. The survey assistant then administered the baseline survey and recorded the (adult) participant's responses. After completing the survey, the women and girl's height and weight were measured and recorded. Project staff returned to re-assess the original evaluation cohort at one- year follow-up.