Feasibility of a Pilot Randomized Controlled Trial to Assess Health Impacts of Community Gardening

Abstract

Studies have demonstrated that community gardening is associated with improved physical health through improving food security, increasing fruit and vegetable intake, increasing physical activity, and healthy weight status. However, no randomized controlled trials have been conducted to assess if these effects are causal. This study was conducted to assess the feasibility of a randomized controlled trial of community gardening to assess the health impacts of community gardening in adults. This study was approved by the Institutional Review Board of University of Colorado Boulder. Participants were recruited to the wait list of a community garden in Denver, Colorado and then randomized to the intervention or control group. The control group was asked to not garden for the season, while the intervention group received a plot at the community garden, plant starts, and offered basic gardening education and assistance. At baseline and four month follow up blood samples were analyzed for lipids, inflammatory markers, and hemoglobin A1C; anthropometric measures were taken; three 24 hour diet recalls were collected; ActivPAL physical activity monitors measured activity for six days; and participants completed a survey including questions on physical and mental health, social support, civic engagement, and neighborhood aesthetics and attachment. At follow up participants completed an evaluation survey of the data collection and of garden participation for the intervention group. At baseline and approximately every three weeks for the duration of the study participants collected samples of their tongue, skin, and stool for microbiome analysis. The study aimed to recruit 30 participants, 15 randomized to each group. Despite significant efforts to recruit participants, only 8 were recruited for each group. Perhaps the largest contributor to the low‐level of recruitment was that recruitment did not begin until after the local gardening season had already started (near the end of May). Six out of the 8 participants recruited in each group completed the study. One member of the control group withdrew from the study and joined the garden, and one was lost to follow up. One member of the gardening group withdrew for unknown reasons and another withdrew due to moving out of state. The participants that completed the study did not find the amount of data collected burdensome, but some had difficulty with the dietary recalls and physical activity monitoring. This study concludes that a randomized controlled trial of community gardening is feasible, but timely recruitment is extremely important due to the need for the intervention to coincide with the gardening season. The results of this pilot study will be used to inform the protocol of a large multi‐year randomized controlled trial of community gardening.Support or Funding InformationCUPC Grant‐ Population Center at University of Colorado Boulder, In‐kind contribution – Center for Microbiome Innovation at University of California San Diego, CTRC Microgrant – University of Colorado Denver