Advertisement

NOTICE: We are experiencing technical issues with Academy members trying to log into the JAND site using Academy member login credentials. We are working to resolve the issue as soon as possible. Alternatively, if you are an Academy member, you can access the JAND site by registering for an Elsevier account and claiming access using the links at the top of the JAND site. Email us at [email protected] for assistance. Thanks for your patience!

Harvest for Health, a Randomized Controlled Trial Testing a Home-Based, Vegetable Gardening Intervention Among Older Cancer Survivors Across Alabama: An Analysis of Accrual and Modifications Made in Intervention Delivery and Assessment During COVID-19

Open AccessPublished:May 06, 2022DOI:https://doi.org/10.1016/j.jand.2022.05.005

      Abstract

      Background

      Accelerated functional decline is a concern among older cancer survivors that threatens independence and quality of life. Pilot studies suggest that vegetable gardening interventions ameliorate functional decline through improved diet and physical activity.

      Objective

      The aim of this article was to describe the rationale, recruitment challenges, and enrollment for the Harvest for Health randomized controlled trial (RCT), which will test the impact of a home-based, vegetable gardening intervention on vegetable and fruit consumption, physical activity, and physical functioning among older cancer survivors. Modifications made to the intervention and assessments to assure safety and continuity of the RCT throughout the COVID-19 pandemic also are reported.

      Design

      Harvest for Health is a 2-year, 2-arm, single-blinded, wait-list controlled RCT with cross-over.

      Participants/setting

      Medicare-eligible survivors of cancers with ≥60% 5-year survival were recruited across Alabama from October 1, 2016 to February 8, 2021.

      Intervention

      Participants were randomly assigned to a wait-list control or a 1-year home-based gardening intervention and individually mentored by extension-certified master gardeners to cultivate spring, summer, and fall vegetable gardens.

      Main outcome measures

      Although the RCT’s primary end point was a composite measure of vegetable and fruit consumption, physical activity, and physical functioning, this article focuses on recruitment and modifications made to the intervention and assessments during COVID-19.

      Statistical analyses performed

      χ2 and t tests (α < .05) were used to compare enrolled vs unenrolled populations.

      Results

      Older cancer survivors (n = 9,708) were contacted via mail and telephone; 1,460 indicated interest (15% response rate), 473 were screened eligible and consented, and 381 completed baseline assessments and were randomized. Enrollees did not differ from nonrespondents/refusals by race and ethnicity, or rural-urban status, but comprised significantly higher numbers of comparatively younger survivors, those who were female, and survivors of breast cancer (P < .001). Although COVID-19 delayed trial completion, protocol modifications overcame this barrier and study completion is anticipated by June 2022.

      Conclusions

      This RCT will provide evidence on the effects of a mentored vegetable gardening program among older cancer survivors. If efficacious, Harvest for Health represents a novel, multifaceted approach to improve lifestyle behaviors and health outcomes among cancer survivors—one with capacity for sustainability and widespread dissemination.

      Keywords

      Research Question: Vegetable gardening interventions hold promise for improving lifestyle behaviors and health of older cancer survivors; however, what proportion are interested and eligible for these interventions and what modifications were necessary during the COVID-19 pandemic?
      Key Findings: This 2-arm, wait-list controlled, cross-over trial accrued 381 cancer survivors who are individually mentored by certified master gardeners to maintain spring, summer, and fall vegetable gardens. Roughly 15% of those contacted were interested in participation, with uptake considerably greater among younger survivors and breast cancer survivors. Several adaptations were made in intervention delivery and assessment, which successfully sustained trial viability via remote means during the COVID-19 pandemic.
      Currently, there are approximately 17 million cancer survivors in the United States.
      • Miller K.D.
      • Nogueira L.
      • Mariotto A.B.
      • et al.
      Cancer treatment and survivorship statistics, 2019.
      Given the confluence of population aging, early detection, and improved treatment, this number is expected to exceed 22 million by 2030.
      • Miller K.D.
      • Nogueira L.
      • Mariotto A.B.
      • et al.
      Cancer treatment and survivorship statistics, 2019.
      Most survivors (65%) are 65 years or older, and are at increased risk of new or worsening chronic health conditions associated with both aging and cancer-related sequelae.
      • Miller K.D.
      • Nogueira L.
      • Mariotto A.B.
      • et al.
      Cancer treatment and survivorship statistics, 2019.
      Compared with individuals without a cancer history, cancer survivors are at increased risk for second malignancies, cardiovascular disease, osteoporosis, and functional impairment.
      Institute of Medicine, National Research Council
      From Cancer Patient to Cancer Survivors: Lost in Transition.
      • Travis L.B.
      • Demark Wahnefried W.
      • Allan J.M.
      • et al.
      Aetiology, genetics and prevention of secondary neoplasms in adult cancer survivors.
      • Armenian S.H.
      • Xu L.
      • Ky B.
      • et al.
      Cardiovascular disease among survivors of adult-onset cancer: A community-based retrospective cohort study.
      • Schoormans D.
      • Vissers P.A.
      • van Herk-Sukel M.P.
      • et al.
      Incidence of cardiovascular disease up to 13 year after cancer diagnosis: A matched cohort study among 32 757 cancer survivors.
      • Strongman H.
      • Gadd S.
      • Matthews A.
      • et al.
      Medium and long-term risks of specific cardiovascular diseases in survivors of 20 adult cancers: A population-based cohort study using multiple linked UK electronic health records databases.
      • Donin N.
      • Filson C.
      • Drakaki A.
      • et al.
      Risk of second primary malignancies among cancer survivors in the United States, 1992 through 2008.
      • Sung H.
      • Hyun N.
      • Leach C.R.
      • et al.
      Association of first primary cancer with risk of subsequent primary cancer among survivors of adult-onset cancers in the United States.
      • Barzi A.
      • Hershman D.L.
      • Till C.
      • et al.
      Osteoporosis in colorectal cancer survivors: Analysis of the linkage between SWOG trial enrollees and Medicare claims.
      • Shapiro C.L.
      • Van Poznak C.
      • Lacchetti C.
      • et al.
      Management of osteoporosis in survivors of adult cancers with nonmetastatic disease: ASCO Clinical Practice Guideline.
      • Anderson C.
      • Gapstur S.M.
      • Leach C.R.
      • et al.
      Medical conditions and physical function deficits among multiple primary cancer survivors.
      • Blair C.K.
      • Jacobs Jr., D.R.
      • Demark-Wahnefried W.
      • et al.
      Effects of cancer history on functional age and mortality.
      • Garcia S.F.
      • Wortman K.
      • Cella D.
      • et al.
      Implementing electronic health record-integrated screening of patient-reported symptoms and supportive care needs in a comprehensive cancer center.
      • Leach C.R.
      • Bellizzi K.M.
      • Hurria A.
      • et al.
      Is it my cancer or am I just getting older?: Impact of cancer on age-related health conditions of older cancer survivors.
      Because 44% of cancer survivors are living at least 10 years beyond diagnosis,
      • Miller K.D.
      • Nogueira L.
      • Mariotto A.B.
      • et al.
      Cancer treatment and survivorship statistics, 2019.
      interventions are needed to prevent or delay age- and treatment-related morbidity.
      Despite evidence of the benefits of a healthy lifestyle on physical health and psychosocial well-being, many cancer survivors do not adhere to recommended guidelines, such as those proposed by the National Comprehensive Cancer Network.
      • Denlinger C.S.
      • Sanft T.
      • Moslehi J.J.
      • et al.
      NCCN Guidelines Insights: Survivorship, Version 2.2020.
      Although interventions have shown efficacy in improving vegetable and fruit (V&F) intake, physical activity (PA), and physical functioning (PF) among older cancer survivors,
      • Morey M.C.
      • Snyder D.C.
      • Sloane R.
      • et al.
      Effects of home-based diet and exercise on functional outcomes among older, overweight long-term cancer survivors: RENEW: A randomized controlled trial.
      • Demark-Wahnefried W.
      • Morey M.C.
      • Sloane R.
      • et al.
      Reach Out to Enhance Wellness home-based diet-exercise intervention promotes reproducible and sustainable long-term improvements in health behaviors, body weight, and physical functioning in older, overweight/obese cancer survivors.
      • Demark-Wahnefried W.
      • Clipp E.C.
      • Lipkus I.M.
      • et al.
      Main outcomes of the FRESH START trial: A sequentially tailored, diet and exercise mailed print intervention among breast and prostate cancer survivors.
      most lack capacity for broader adoption, implementation, and sustainability in real-world settings, indicating a considerable research-to-practice gap.
      Gardening, especially vegetable gardening, is associated with healthier diets, increased PA, better health-related quality of life (HRQOL), and lower mortality.
      • Leng C.H.
      • Wang J.D.
      Daily home gardening improved survival for older people with mobility limitations: An 11-year follow-up study in Taiwan.
      • Park S.
      • Shoemaker C.
      • Haub M.
      Can older gardeners meet the physical activity recommendation through gardening?.
      • Soga M.
      • Cox D.T.
      • Yamaura Y.
      • et al.
      Health benefits of urban allotment gardening: Improved physical and psychological well-being and social integration.
      • Van den Berg A.E.
      • Van Winsum-Westra M.
      • De Vries S.
      • et al.
      Allotment gardening and health: A comparative survey among allotment gardeners and their neighbors without an allotment.
      • Barnidge E.K.
      • Hipp P.R.
      • Estlund A.
      • et al.
      Association between community garden participation and fruit and vegetable consumption in rural Missouri.
      • Sommerfeld A.J.
      • McFarland A.L.
      • Waliczek T.M.
      • et al.
      Growing minds: Evaluating the relationship between gardening and fruit and vegetable consumption in older adults.
      • Machida D.
      Relationship between community or home gardening and health of the elderly: A web-based cross-sectional survey in Japan.
      • Kegler M.C.
      • Prakash R.
      • Hermstad A.
      • et al.
      Home gardening and associations with fruit and vegetable intake and BMI.
      • Scott T.L.
      • Masser B.M.
      • Pachana N.A.
      Positive aging benefits of home and community gardening activities: Older adults report enhanced self-esteem, productive endeavours, social engagement and exercise.
      Older gardeners, as compared with nongardeners, demonstrate better balance and gait speed, and report fewer comorbidities, functional limitations, and falls.
      • Chen T.Y.
      • Janke M.C.
      Gardening as a potential activity to reduce falls in older adults.
      ,
      • Chen T.Y.
      • Janke M.C.
      Effectiveness of gardening activities on improving older adults' gait and balance.
      To date, most gardening programs have been based in school or community settings,
      • Soga M.
      • Cox D.T.
      • Yamaura Y.
      • et al.
      Health benefits of urban allotment gardening: Improved physical and psychological well-being and social integration.
      ,
      • Van den Berg A.E.
      • Van Winsum-Westra M.
      • De Vries S.
      • et al.
      Allotment gardening and health: A comparative survey among allotment gardeners and their neighbors without an allotment.
      ,
      • Strout K.
      • Jemison J.
      • O'Brien L.
      • et al.
      GROW: Green Organic Vegetable Gardens to Promote Older Adult Wellness: A feasibility study.
      • Lee R.E.
      • Parker N.H.
      • Soltero E.G.
      • et al.
      Sustainability via Active Garden Education (SAGE): Results from two feasibility pilot studies.
      • Davis J.N.
      • Martinez L.C.
      • Spruijt-Metz D.
      • et al.
      LA Sprouts: A 12-week gardening, nutrition, and cooking randomized control trial improves determinants of dietary behaviors.
      • Gatto N.M.
      • Martinez L.C.
      • Spruijt-Metz D.
      • et al.
      LA Sprouts randomized controlled nutrition, cooking and gardening programme reduces obesity and metabolic risk in Hispanic/Latino youth.
      • Wood C.J.
      • Pretty J.
      • Griffin M.
      A case-control study of the health and well-being benefits of allotment gardening.
      • Spees C.K.
      • Hill E.B.
      • Grainger E.M.
      • et al.
      Feasibility, preliminary efficacy, and lessons learned from a garden-based lifestyle intervention for cancer survivors.
      • Hume A.
      • Wetten A.
      • Feeney C.
      • et al.
      Remote school gardens: Exploring a cost-effective and novel way to engage Australian Indigenous students in nutrition and health.
      with a few smaller-scale programs evaluated in health care facilities.
      • White P.C.
      • Wyatt J.
      • Chalfont G.
      • et al.
      Exposure to nature gardens has time-dependent associations with mood improvements for people with mid- and late-stage dementia: Innovative practice.
      • Brown V.M.
      • Allen A.C.
      • Dwozan M.
      • et al.
      Indoor gardening older adults: Effects on socialization, activities of daily living, and loneliness.
      • Lee Y.
      • Kim S.
      Effects of indoor gardening on sleep, agitation, and cognition in dementia patients—A pilot study.
      Vegetable gardening, regardless of setting, is a relatively low-cost, and multifaceted strategy to promote physical and psychosocial health and well-being and has great potential for sustainability of health behaviors and outcomes, as gardening involves nonrepetitive activities that prevent the burnout that is common with other forms of exercise, and allows enjoyment of the outdoors
      • Kerr J.
      • Marshall S.
      • Godbole S.
      • et al.
      The relationship between outdoor activity and health in older adults using GPS.
      • Kerr J.
      • Sallis J.F.
      • Saelens B.E.
      • et al.
      Outdoor physical activity and self rated health in older adults living in two regions of the U.S.
      • Marsh A.P.
      • Katula J.A.
      • Pacchia C.F.
      • et al.
      Effect of treadmill and overground walking on function and attitudes in older adults.
      ; provides a sense of achievement and zest for life that come from nurturing and observing new life and growth
      • Sommerfield A.J.
      • Zajicek J.
      • Waliczek T.M.
      Growing minds: Evaluating the effect of gardening on quality of life in older adults.
      ; and imparts natural prompts because plants require regular care and attention and serve as continual and dynamic behavioral cues.
      Harvest for Health, a home-based, individually mentored, vegetable gardening intervention that pairs cancer survivors with certified master gardeners (MGs) is designed with a focus on implementation, dissemination, and sustainability. This project builds on the extant infrastructure of the US Department of Agriculture Cooperative Extension System’s Master Gardener Program that includes as its mission “to empower communities to meet the challenges they face to improve nutrition and protect the environment.”
      Cooperative Extension System
      US Department of Agriculture, National Institute of Food and Agriculture.
      Throughout the United States, the MG program operates largely via land-grant universities in each state. Certified MGs complete state-designated hours of training and annual community service to maintain active status. Harvest for Health was developed in the southern United States—a region that is home to 10.8 million older cancer survivors.
      State cancer profiles
      National Program of Cancer Registries.
      Although designed as a 3-growing season program, Harvest for Health can be adapted to regions with fewer growing seasons and people with other chronic diseases for which PF and lifestyle behaviors are key.
      • Blair C.K.
      • Harding E.M.
      • Adsul P.
      • et al.
      Southwest Harvest for Health: Adapting a mentored vegetable gardening intervention for cancer survivors in the southwest.
      Results from a series of 3 pilot studies in samples ranging from 12 to 82 cancer survivors indicated success in establishing feasibility, safety, and satisfaction, and suggest that the gardening interventions increased V&F intake and PA, attenuated increases in waist circumference, and improved PF and performance.
      • Blair C.K.
      • Madan-Swain A.
      • Locher J.L.
      • et al.
      Harvest for health gardening intervention feasibility study in cancer survivors.
      • Bail J.R.
      • Fruge A.D.
      • Cases M.G.
      • et al.
      A home-based mentored vegetable gardening intervention demonstrates feasibility and improvements in physical activity and performance among breast cancer survivors.
      • Cases M.G.
      • Fruge A.D.
      • De Los Santos J.F.
      • et al.
      Detailed methods of two home-based vegetable gardening intervention trials to improve diet, physical activity, and quality of life in two different populations of cancer survivors.
      • Demark-Wahnefried W.
      • Cases M.G.
      • Cantor A.B.
      • et al.
      Pilot randomized controlled trial of a home vegetable gardening intervention among older cancer survivors shows feasibility, satisfaction, and promise in improving vegetable and fruit consumption, reassurance of worth, and the trajectory of central adiposity.
      To evaluate the efficacy and cost–benefit of the intervention on health behaviors and outcomes, the program was expanded to a statewide randomized controlled trial (RCT). To date, Harvest for Health is the largest RCT to evaluate the efficacy of gardening on health outcomes in adults. We will detail the enrollment, as well as challenges encountered and protocol modifications made to assure safety and continuity of the RCT throughout the COVID-19 pandemic.

      Methods

      Inter-Institutional Partnership Between a Comprehensive Cancer Center and a Land Grant University

      The Harvest for Health program is a partnership between the O’Neal Comprehensive Cancer Center at UAB (OCCC-UAB) and the Alabama Cooperative Extension System (ACES) at Auburn University, the major land grant university in the state. The OCCC-UAB identifies, enrolls, and evaluates the cancer survivor participants in the gardening intervention, and ACES oversees the MGs and delivers the intervention. Both organizations monitor intervention fidelity. Similar to all states, ACES provides intensive horticultural training and certification to individuals, who then volunteer as MGs in their communities. In Alabama, there are roughly 2,000 MGs, with each volunteering more than 50 hours/year to maintain certification. As an initial step, and to gauge potential capacity for Harvest for Health, ACES conducted an online statewide survey among MGs; results indicated that 26% of respondents were “interested” and 71% were “exceptionally interested” in committing volunteer hours to this project,
      • Blair C.K.
      • Madan-Swain A.
      • Locher J.L.
      • et al.
      Harvest for health gardening intervention feasibility study in cancer survivors.
      thus assuring an adequate workforce.

      Study Design

      Harvest for Health is a 2-arm, single-blinded, RCT that uses a wait-list controlled, cross-over design. Participants are randomly assigned with equal allocation to either the immediate intervention or the delayed intervention (wait-list control) arms, arms are crossed-over at 1 year and followed for an additional year. This design, as used and described in the RENEW (Reach Out to Enhance Wellness ) diet and exercise RCT of 641 older cancer survivors,
      • Demark-Wahnefried W.
      • Morey M.C.
      • Sloane R.
      • et al.
      Reach Out to Enhance Wellness home-based diet-exercise intervention promotes reproducible and sustainable long-term improvements in health behaviors, body weight, and physical functioning in older, overweight/obese cancer survivors.
      allows testing for efficacy (between-arm differences from baseline to 1-year follow-up), exploration of durability (comparison of 1- to 2-year outcomes in the immediate intervention arm), and exploration of repeatability (contrasting the effects observed with the delayed intervention compared with those of the immediate intervention). Outcome measures are assessed on all study participants during observational assessments (baseline and 1- and 2-year follow-up), and with mailed surveys (6- and 18-month follow-up). Figure 1 illustrates the study flow from identification of potential cases through screening and randomization, and also illustrates projected follow-up. The Harvest for Health study protocol was approved by the UAB Institutional Review Board and all participants provide written informed consent. The trial was registered with the National Institutes of Health (ClinicalTrials.gov ID: NCT02985411).
      Figure thumbnail gr1
      Figure 1Study flow diagram of the Harvest for Health randomized controlled trial that tests the impact of a vegetable gardening intervention on vegetable & fruit (V&F) consumption, physical activity, and physical functioning among older cancer survivors residing in Alabama.

      Study Aims

      The primary aim tests the impact of the gardening intervention on V&F intake, PA, and PF. Secondary aims were to assess the intervention’s effects on secondary end points, for example, plasma α-carotene (biomarker of V&F consumption), inflammatory biomarkers (serum interleukin 6 and tumor necrosis factor–α), toenail cortisol, the fecal microbiome, adiposity (waist circumference and body mass index [calculated as kg/m2]), sleep, reassurance of worth (ie, feeling of adding value), and HRQOL; evaluate the durability and repeatability of the intervention; explore participant factors related to program efficacy (eg, sex, comorbidity, and age); and assess the value of health improvements relative to intervention costs.

      Recruitment, Eligibility, and Consent

      Recruitment for Harvest for Health spanned from October 1, 2016 to February 8, 2021. Similar to recruitment procedures documented as successful by Patterson and colleagues,
      • Patterson R.E.
      • Marinac C.R.
      • Natarajan L.
      • et al.
      Recruitment strategies, design, and participant characteristics in a trial of weight-loss and metformin in breast cancer survivors.
      cases were ascertained from cancer registries (UAB and Alabama state) employing the following procedure:
      • 1.
        Identify Medicare-eligible participants (noting that for most, a minimum age of 65 years is used) who have been diagnosed and completed curative therapy for cancers associated with 5-year survival rates of ≥60% (ie, in situ or localized cancers of the bladder, cervix; gastric cardia, larynx or early stage multiple myeloma; in situ or locoregionally staged melanoma and cancers of the colorectum, endometrium, kidney/renal pelvis, oral cavity/pharynx, ovary, prostate, soft-tissue sarcoma, thyroid, and female breast [only female—as defined by sex at birth and obtained from cancer registries or medical records—participants with breast cancer were considered, given poorer documented survival among men],
        • Wang F.
        • Shu X.
        • Meszoely I.
        • et al.
        Overall mortality after diagnosis of breast cancer in men vs women.
        as well as melanoma, and all stages of testes cancer, leukemia, and Hodgkin or non-Hodgkin lymphoma).
      • 2.
        Contact oncologists of participants to gain permission for contact.
      • 3.
        Mail letters of invitation and then follow-up with telephone contact (placing up to 6 call attempts at various days and times) to assess study interest and eligibility.
      Other eligibility criteria include:
      • 1.
        Reporting ≤5 daily servings of V&F (assessed using the totals from 2 brief screening questions: “On average, how many cups of raw green leafy vegetables do you eat per day and “On average, how many cups of other fruits and vegetables do you eat per day (not including lettuce, potatoes, fruit juices, dried beans)?;
      • 2.
        Reporting <150 minutes of weekly moderate-to-vigorous PA
        Physical activity intensity. Centers for Disease Control and Prevention.
        (assessed using the screening question: “On average how many minutes do you exercise a week?);
      • 3.
        Reporting ≥1 PF limitation on the PF subscale of the 36-Item Short Form Survey
        • Ware J.E.
        • Kosinski M.
        SF-36® Physical & Mental Health Summary Scales: A Manual for Users of Version 2.
        ;
      • 4.
        Absence of pre-existing medical conditions that preclude gardening (eg, severe orthopedic conditions, pending hip/knee replacement, paralysis, dementia, blindness, untreated stage 3 hypertension, or conditions such as myocardial infarction that require oxygen or hospitalization within the past 6 months);
      • 5.
        Community dwelling;
      • 6.
        No existing or recent vegetable garden (eg, neither themselves nor a household member tended a vegetable garden in the past 2 years);
      • 7.
        Residence can accommodate 1 raised bed (4 × 8 feet) or 4 grow boxes (24 × 20 inches) with ample sunshine (6 or more hours per day) and have ready access to water;
      • 8.
        English-speaking and writing; and
      • 9.
        Willingness to be randomized to either study arm and to participate for 2 years.
      Additional recruitment strategies included study flier distribution, radio public service announcements, and community presentations. Eligible respondents were mailed 2 copies of the study consent form and a telephone appointment was scheduled with study staff to review the trial requirements and obtain informed signed consent (1 copy of the signed form was returned in a pre-addressed, postage-paid envelope).

      Study Flow Overview

      Harvest for Health is delivered on an individual basis, although for logistical reasons (ie, focused recruitment and supervision of MGs, delivery of intervention supplies, and scheduling of assessment travel), a rolling recruitment plan is followed that corresponds to county-dependent planting dates appropriate for spring, summer, and fall gardens. Therefore, baseline assessments with randomization occur on a rotating trimester basis. For participants who randomize to the immediate intervention, the year-long mentored gardening intervention starts with a meet-and-greet kick-off event, and for those assigned to the delayed intervention, the intervention begins 1 year later. Participants are assessed every 6 months, alternating between in-person and survey-based assessments.

      Assessments

      Assessments are conducted at baseline and throughout the 2-year study period on the following schedule: every 6 months for primary outcomes and every 12 months for secondary outcomes (with the exception of health care utilization, which is assessed semi-annually). The primary outcome is multifaceted in nature and relies on achieving the following 3 goals: increase of ≥1 V&F servings/d; increase in moderate-to-vigorous PA by ≥30 min/wk; and improvement in PF. Primary outcome data for each of these domains are gathered using self-administered surveys composed of validated instruments for which study participants receive either a mailed questionnaire or an e-mailed link to an online REDCap survey. Secondary outcomes include a broad array of measures that assess different domains that also serve various functions, including the following: objective measures that corroborate self-reported primary outcomes data; anthropometric measures of adiposity; health status and utilization; HRQOL, stress, and self-worth; sleep; and impact on the gut microbiome, cortisol, and biomarkers of inflammation. Given that many secondary outcomes rely on physical assessments and the collection of biospecimens, Harvest for Health planned for in-person assessments at participants’ homes or community facilities (eg, medical centers, clinics, senior centers, and town offices). Thus, 2 to 3 weeks before each annual assessment, participants receive a mailed packet that includes the following: an instruction sheet to prepare for phlebotomy (fast for ≥4 hours and be well-hydrated); instructions to clip toenails and baggies in which to place them for cortisol assays; activity and sleep logs and a programmed accelerometer (Actigraph) for a 7-day collection; and a fecal wipe and bag (to assess the microbiome). At the time of the in-person visit, study staff perform phlebotomy; conduct other in-person assessments; and collect accelerometers, nails, and fecal wipes. Blood samples are configured into sera, plasma, and buffy coat (half immersed in RNAlater [Invitrogen] and the other half stored as is). All biospecimens are stored at –80o C until analysis. Figure 2 details each of the study measures.
      • Chen T.Y.
      • Janke M.C.
      Gardening as a potential activity to reduce falls in older adults.
      ,
      • Sommerfield A.J.
      • Zajicek J.
      • Waliczek T.M.
      Growing minds: Evaluating the effect of gardening on quality of life in older adults.
      ,
      • Ware J.E.
      • Kosinski M.
      SF-36® Physical & Mental Health Summary Scales: A Manual for Users of Version 2.
      • Thompson F.E.
      • Subar A.F.
      • Smith A.F.
      • et al.
      Fruit and vegetable assessment: Performance of 2 new short instruments and a food frequency questionnaire.
      • Stewart A.L.
      • Mills K.M.
      • King A.C.
      • et al.
      CHAMPS physical activity questionnaire for older adults: Outcomes for interventions.
      • Bieri J.G.
      • Brown E.D.
      • Smith J.C.
      Determination of individual carotenoids in human plasma by high performance liquid chromatography.
      • Rikli R.E.
      • Jones C.J.
      Development and validation of a functional fitness test for community-residing older adults.
      • Taekema D.G.
      • Gussekloo J.
      • Maier A.B.
      • et al.
      Handgrip strength as a predictor of functional, psychological and social health. A prospective population-based study among the oldest old.
      • Brazier J.E.
      • Roberts J.
      The estimation of a preference-based measure of health from the SF-12.
      • Fillenbaum G.G.
      Multidimensional Functional Assessment of Older Adults: The Duke Older Americans Resources and Services Procedures.
      • Cutrona C.E.
      • Russell D.
      The provisions of social relationships and adaptation stress.
      • Buysse D.J.
      • Reynolds 3rd, C.F.
      • Monk T.H.
      • et al.
      The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research.
      • Fruge A.D.
      • Cases M.G.
      • Howell C.R.
      • et al.
      Fingernail and toenail clippings as a non-invasive measure of chronic cortisol levels in adult cancer survivors.
      • Warnock F.
      • McElwee K.
      • Seo R.J.
      • et al.
      Measuring cortisol and DHEA in fingernails: A pilot study.
      • Fruge A.D.
      • Van der Pol W.
      • Rogers L.Q.
      • et al.
      Fecal Akkermansia muciniphila is associated with body composition and microbiota diversity in overweight and obese women with breast cancer participating in a presurgical weight loss trial.
      • Caporaso J.G.
      • Kuczynski J.
      • Stombaugh J.
      • et al.
      QIIME allows analysis of high-throughput community sequencing data.
      • Caporaso J.G.
      • Lauber C.L.
      • Walters W.A.
      • et al.
      Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample.
      Human Microbiome Project Consortium
      Structure, function and diversity of the healthy human microbiome.
      • Faith D.P.
      • Baker A.M.
      Phylogenetic diversity (PD) and biodiversity conservation: Some bioinformatics challenges.
      Figure 2Study outcomes of the Harvest for Health vegetable gardening intervention in older cancer survivors in Alabama.
      OutcomeMeasure
      Primary outcome: Composite measure derived from the following subjective measures
      V&F
      V&F = vegetable and fruit.
      intake
      The 10-item, Eating at America’s Table Screener assesses frequencies and portion sizes of V&Fs. This NCI
      NCI = National Cancer Institute.
      -developed tool has proven reliability and validity, and strong concordance with 24-hour recalls.
      • Thompson F.E.
      • Subar A.F.
      • Smith A.F.
      • et al.
      Fruit and vegetable assessment: Performance of 2 new short instruments and a food frequency questionnaire.
      PA
      PA = physical activity.
      The 41-item Community Healthy Activities Models Program for Seniors questionnaire has proven validity for capturing PA in older adults (ICC
      ICC = intraclass correlation.
       = 0.81-0.88 for moderate intensity activity).
      • Stewart A.L.
      • Mills K.M.
      • King A.C.
      • et al.
      CHAMPS physical activity questionnaire for older adults: Outcomes for interventions.
      PF
      PF = physical function.
      The 10-item PF subscale of the 36-Item Short Form Survey assesses general PF and is valid and reliable for use in healthy and chronically ill adults. Internal consistency is excellent: α = .89-.92; it has published norms and is sensitive to change.
      • Ware J.E.
      • Kosinski M.
      SF-36® Physical & Mental Health Summary Scales: A Manual for Users of Version 2.
      Secondary outcomes: Objective measures below serve to validate subjective measures above
      Plasma α-caroteneCarotenoids are analyzed via high-performance liquid chromatography as per Bieri and colleagues
      • Bieri J.G.
      • Brown E.D.
      • Smith J.C.
      Determination of individual carotenoids in human plasma by high performance liquid chromatography.
      on a Hitachi 911 clinical analyzer with standard chemistries by Roche. Lipid-adjusted values of α-carotene are explicitly selected for analyses because these are least likely to be affected by supplement use.
      PA7-day accelerometer-assisted collection using ActiGraph
      Physical performanceThe Senior Fitness Battery objectively assesses physical performance in several domains, is sensitive to change, devoid of ceiling effects, and has normative scores.
      • Rikli R.E.
      • Jones C.J.
      Development and validation of a functional fitness test for community-residing older adults.
      Assessments included: 30-second chair stand (lower body strength); 8-foot get up and go test (agility, dynamic balance); 8-foot walk (gait speed); sit and reach (flexibility); back scratch (flexibility); 2-minute step test (endurance); 30-second arm curl (upper arm strength); and grip strength, (disability and functional limitation predictor)
      • Taekema D.G.
      • Gussekloo J.
      • Maier A.B.
      • et al.
      Handgrip strength as a predictor of functional, psychological and social health. A prospective population-based study among the oldest old.
      as assessed using an Omron dynamometer per manufacturer’s instructions.

      Balance testing was done using iPAD minis (Apple) uploaded with the Sway app and using the Physical Performance Mobility Examination protocol (ie, eyes open and then shut for side by side, semi-tandem, and tandem stances).
      Other secondary outcomes
      HRQOL
      HRQOL = Health-related quality of life.
      (QALYs
      QALY = quality-adjusted life-year.
      )
      The 36-Item Short Form Survey HRQOL Index (version 2) provides a global measure of mental and physical HRQOL 8 subscales.
      • Ware J.E.
      • Kosinski M.
      SF-36® Physical & Mental Health Summary Scales: A Manual for Users of Version 2.
      The internal consistency and reliability for all 8 subscales is high, ICC ranging from 0.78 to 0.93. A subset of output data will also be used to create QALYs.
      • Brazier J.E.
      • Roberts J.
      The estimation of a preference-based measure of health from the SF-12.
      Health care utilization“Any changes in health” are assessed every 6 months; health events, hospitalizations, and medication changes (type and dose) are reported.
      ComorbidityThe Older Americans Resources & Services Comorbidity Index (43 items) used in multiple studies in older adults will be used to assess the number of chronic medical conditions and symptoms and their functional impact (severity).
      • Fillenbaum G.G.
      Multidimensional Functional Assessment of Older Adults: The Duke Older Americans Resources and Services Procedures.
      Because falls are a particular issue in this population, an item validated by Chen and Janke
      • Chen T.Y.
      • Janke M.C.
      Gardening as a potential activity to reduce falls in older adults.
      that assesses falls in the past year will also be included.
      Reassurance of worth1 of 6 subscales of the Revised Social Provision Scale, this 4-item subscale measure has reliability estimates of 0.60-0.70 and will be used to assess the psychosocial benefits of gardening.
      • Sommerfield A.J.
      • Zajicek J.
      • Waliczek T.M.
      Growing minds: Evaluating the effect of gardening on quality of life in older adults.
      ,
      • Cutrona C.E.
      • Russell D.
      The provisions of social relationships and adaptation stress.
      Sleep qualityThe Pittsburgh Sleep Quality Index is a 19-item scale composed of 7 components, with proven reliability and consistency (Cronbach α = .83).
      • Buysse D.J.
      • Reynolds 3rd, C.F.
      • Monk T.H.
      • et al.
      The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research.
      Sleep quality also will be measured by accelerometry (actigraphy).
      AnthropometricsAll anthropometric assessments are measured in light clothing and no shoes and include: height assessed using a portable stadiometer; weight assessed using a calibrated digital scale; waist circumference is measured in duplicate at the level of the umbilicus and upon exhale using a tension-controlled nonstretch tape measure (Gulick II; Fisher Scientific); and percent body fat via bioimpedance (Omron HBF-306C). This measure is only performed on participants without pacemakers.
      StressToenail cortisol levels: In previous studies, excellent correlations (r = 0.58; P < .0001) were found between salivary cortisol and toenail cortisol.
      • Fruge A.D.
      • Cases M.G.
      • Howell C.R.
      • et al.
      Fingernail and toenail clippings as a non-invasive measure of chronic cortisol levels in adult cancer survivors.
      Therefore, toenails are collected, stored at room temperature, and assayed using the methods of Warnock and colleagues.
      • Warnock F.
      • McElwee K.
      • Seo R.J.
      • et al.
      Measuring cortisol and DHEA in fingernails: A pilot study.
      InflammationSera from venipuncture is analyzed for interleukin-6 and tumor necrosis factor–α via electrochemiluminescence (Meso-scale Discovery).
      MicrobiomeAs in previous studies,
      • Fruge A.D.
      • Van der Pol W.
      • Rogers L.Q.
      • et al.
      Fecal Akkermansia muciniphila is associated with body composition and microbiota diversity in overweight and obese women with breast cancer participating in a presurgical weight loss trial.
      participants use sterile wipes to collect fecal samples, which are stored in their home freezer until staff collection. Microbe DNA extraction is performed using a ZR Fecal DNA Miniprep, then amplified, sequenced, and purified. Analyses are performed with the Quantitative Insight into Microbial Ecology suite, version 1
      • Caporaso J.G.
      • Kuczynski J.
      • Stombaugh J.
      • et al.
      QIIME allows analysis of high-throughput community sequencing data.
      and the QWRAP wrapper pipeline. α-Diversity measures are calculated using Chao1, observed species, Shannon Index, and whole tree Phylogenetic Diversity.
      • Caporaso J.G.
      • Lauber C.L.
      • Walters W.A.
      • et al.
      Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample.
      Human Microbiome Project Consortium
      Structure, function and diversity of the healthy human microbiome.
      • Faith D.P.
      • Baker A.M.
      Phylogenetic diversity (PD) and biodiversity conservation: Some bioinformatics challenges.
      a V&F = vegetable and fruit.
      b NCI = National Cancer Institute.
      c PA = physical activity.
      d ICC = intraclass correlation.
      e PF = physical function.
      f HRQOL = Health-related quality of life.
      g QALY = quality-adjusted life-year.
      In addition to serving as a secondary outcome, comorbidity will be explored as a potential moderator. Furthermore, as per the logic model (Figure 3), social support and self-efficacy will be explored as potential mediators. The widely used social support scales of Sallis and colleagues
      • Sallis J.F.
      • Pinski R.B.
      • Grossman R.M.
      • et al.
      The development of self-efficacy scales for healthrelated diet and exercise behaviors.
      to eat more V&Fs (10 items) and get more exercise (13 items) are used, given their strong psychometric properties (α = .70). Self-efficacy items adapted from Bandura and found to successfully mediate increases in exercise and eating more V&F among cancer survivors are also used, with additional adaptation to assess gardening self-efficacy (ie, how sure are you that you could maintain a thriving vegetable garden?).
      • Bandura A.
      Social Learning Theory.
      ,
      • Mosher C.E.
      • Lipkus I.
      • Sloane R.
      • et al.
      Long-term outcomes of the FRESH START trial: Exploring the role of self-efficacy in cancer survivors' maintenance of dietary practices and physical activity.
      Figure thumbnail gr2
      Figure 3Logic model of Harvest for Health illustrating the potential mediators of the gardening intervention and their potential influence on study outcomes (primary outcomes are featured in italicized font). IL6 = interleukin-6; TNF-α = tumor necrosis factor–α; V&F = vegetables and fruit.
      As indicated in the Recruitment, Eligibility, and Consent section and Figure 1, cancer registries largely served as the information source for cancer type and stage, sex, age, race and ethnicity, and county of residence; among cases who self-referred, this information was captured via verification from treating oncologists. Data on the highest grade of educational attainment, income, employment status, tobacco use, and types of cancer treatment were captured via self-report on the baseline survey.

      Randomization

      On completion of the baseline assessment, participants are considered enrolled and block randomized within each county on a trimester basis. Participants are allocated evenly to either the immediate intervention or delayed intervention arms.

      Vegetable Gardening Intervention

      Theoretical Framework

      The Harvest for Health intervention is guided by Social Cognitive Theory
      • Bandura A.
      Social Learning Theory.
      and the Social Ecological Model.
      • Bronfenbrenner U.
      The Ecology of Human Development, Experiments by Nature and Design.
      Aligned with Social Cognitive Theory, MGs serve as role models and promote gardening self-efficacy by reinforcing key concepts and providing encouragement and guidance in overcoming barriers. Self-monitoring, a key Social Cognitive Theory concept, is addressed through the assignment of a gardening journal. Social Ecological Model concepts of social and environmental support are addressed via interactions with the MG and with nature. The logic model for Harvest for Health (Figure 3) depicts potential mediators of the gardening intervention and their potential influence on study outcomes.

      Intervention Delivery and Support

      Because Harvest for Health promotes home-based vegetable gardening and requires in-person visits at survivors’ homes, ACES performs criminal background checks on all MGs, although to date no one has failed the background check (note that home-based visits also raised safety concerns during COVID-19; adaptations to intervention delivery and assessment are covered in an upcoming subsection). MGs also complete 4 hours of intensive training on content co-developed by OCCC-UAB and ACES that covers cancer, cancer survivorship, and delivery of the intervention. Contact between the MG and survivor occurs on a biweekly basis alternating between monthly home visits and interactions that transpire via telephone, text, or e-mail (the total time commitment averages 1.5 hours per month: 1 hour for the home visit and 30 minutes for other communications). Efforts are made to pair MGs with survivors who live within 15 miles of one another to enhance interactions. Survivor–MG dyads initially become acquainted at meet-and-greet group functions scheduled at county extension offices, churches, or public libraries, and also feature 20- to 30-minute presentations by OCCC-UAB and ACES staff to review the project and field questions. Distribution of gardening supplies also occurs at these events.
      MGs help survivors plan and set-up their gardens during the initial home visit (this session is longer and requires 2 hours). Thereafter, biweekly contacts are used to review progress (eg, status of plants and harvesting), promote optimal garden management (ie, soil/pest management and watering), and answer questions or troubleshoot problems (eg, weeds and mildew). During monthly home visits and as old plants expire, the MG assists the participant in substituting new items.
      Strategies to monitor and enhance fidelity to intervention delivery by MGs and to gardening by participants are based on the National Institutes of Health Behavior Change Consortium recommendations.
      • Bellg A.J.
      • Borrelli B.
      • Resnick B.
      • et al.
      Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH Behavior Change Consortium.
      First, MGs log all encounters with their assigned participant on the ACES website, which is tracked by Auburn staff to ensure biweekly contact, completion of gardening activities (eg, plant rotation), and review of core content (eg, soil care). Similarly, survivors are asked to log MG contact and provide time-stamped photographs of the garden at each MG visit, which are submitted to the OCCC-UAB study office. These photographs are entered into monthly contests that are judged by study staff for creativity and demonstration of best practices; winning MG–survivor dyads receive garden gloves and seed packets. Tracking of both the MG logs and the survivor logs with accompanying photographs serves as a measure of adherence.
      Education is a key component of the intervention and to be sustainable the participant must learn as much as possible about plants (eg, purchasing and growing from seeds) and gardening (eg, planting, tending, and harvesting). The MG mentor imparts knowledge by teaching participants about sustainable gardening principles and “soft chemistries” for fertilizing and pest control; visiting local garden centers with participants to impart skills in selecting healthy plants and additional supplies (eg, extra soil or fertilizer); and conducting gardening education systematically and in brief segments to promote self-mastery among participants. Thus, in accordance with enforcing incremental goals,
      • Bandura A.
      Social Learning Theory.
      participants’ first gardens rely on basic knowledge and skill levels and, as plants are rotated and gardens evolve, higher levels of skill are gained. In addition, ACES maintains a dedicated, private Facebook site to reinforce this knowledge and promote social support by encouraging interactive sharing of gardening experiences, suggestions, recipes, and photos among study participants and MGs.

      Supplies

      According to preference and gardening conditions, each Harvest for Health participant receives either 1 raised bed kit or 4 grow boxes that offer comparable square footage and are ideal for balconies, patios, or high-shade environments (Figure 4). Standard issue supplies also include the following: gardening journal; soil; fertilizer; mulch; frost cloth; trellis; gardening hose; watering can; trowel; cultivator; hoe; soft chemistry insect control products; seeds; transplants; and tomato cages. Uncommon supplies (eg, deer fences) are managed on a case-by-case basis. Participants also receive a hat and sunscreen to promote sun safety and a binder that includes contact information and a checklist to track MG contact, brochures on garden design and management, safety, and health (eg, sun safety, protecting knees/back, and lymphedema prevention), nutritional benefits of V&F, V&F recipes, and cooking methods to preserve nutrients. MGs receive a similar binder and sun safety supplies. To further enhance engagement, participants receive monthly postcards highlighting seasonal vegetables, nutrition facts, and healthy recipes. Delayed intervention participants and MGs receive identical supplies on completion of 12-month follow-up.
      Figure thumbnail gr3
      Figure 4Participants of the Harvest for Health randomized controlled trial received 1 raised bed kit (4 × 8 feet × 10.5 inches) or 4 grow box kits (24.5 × 20.5 inches) depending on preferences and residential accommodations.

      Statistical Analyses

      Descriptive statistics are calculated for all study variables and normality checks are performed with subsequent transformation, if indicated. All statistical tests are 2-sided using a significance level of 5%. Statistical analyses are conducted using SAS, version 9.4
      or higher. Our initial statistical analyses focus on determining potential demographic and cancer type differences between cancer survivors who express interest in the gardening intervention and those who refuse or are unresponsive, and those who enroll in the RCT vs those not enrolled. Here, t tests are conducted for continuous variables, such as age, and χ2 tests are performed for categorical variables, such as sex, race and ethnicity, residence in a rural- or urban-classified county, and cancer type. These analyses are now complete and results are presented in this article—findings that are integral in assessing program interest and to appropriately generalize the main outcomes of this trial on its completion.

      Primary Outcome

      The primary study end point will be a composite dichotomous score (yes or no) based on the increase from baseline to 1-year follow-up on achieving the following 3 goals (based on findings of the first Harvest for Health pilot study):
      • Blair C.K.
      • Madan-Swain A.
      • Locher J.L.
      • et al.
      Harvest for health gardening intervention feasibility study in cancer survivors.
      1) increase of ≥1 V&F serving/d; 2) increase in moderate-to-vigorous PA by ≥30 min/wk; and 3) any improvement in self-reported PF. These analyses will be conducted in 2 ways: 1) on self-reported data (in which achievement of all 3 aforementioned goals is coded as a yes or no), and 2) only on self-reported data that are corroborated by objective measures (ie, as per Brevik and colleagues,
      • Brevik A.
      • Andersen L.F.
      • Karlsen A.
      • Trygg K.U.
      • Blomhoff R.
      • Drevon C.A.
      Six carotenoids in plasma used to assess recommended intake of fruits and vegetables in a controlled feeding study.
      plasma α-carotene increases of ≥10% are used to confirm ≥1 serving/d increases in V&F consumption, increases in accelerometer-derived PA confirm self-reports, and enhanced scores on at least two-thirds of tests within the physical performance battery lend support to self-reported increases in PF). Findings of both analyses will be reported with the χ2 test used to test for arm differences in goal achievement. Intent-to-treat analysis will be conducted; a score of 0 (not achieving 3 of 3 goals) will be used for missing data. Proportions of the immediate intervention arm who achieve the composite goal, along with 95% CIs, are computed for year 1 and compared with proportions and 95% CIs generated for year 2 to assess durability. For repeatability, the proportion and CIs achieved for the immediate intervention arm at year 1 are compared with the proportion and CIs achieved in the delayed intervention arm at year 2. Using a 2-sided .05 significance level, a χ2 test with 160 participants/arm yields >90% power to detect between-arm differences (assumptions: 15% attrition from the planned enrollment of 185 participants/arm and a moderate response difference (≥20%) according to Cohen).
      • Cohen J.
      Statistical Power Analysis for the Behavioral Sciences.

      Secondary Outcomes

      Secondary outcomes include HRQOL; self-efficacy and social support to eat an additional daily serving of V&F and get at least an additional 30 minutes of PA per week (including gardening); sleep; reassurance of worth; adiposity (BMI and waist circumference); toenail cortisol, serum interleukin-6 and tumor necrosis factor–α microbiome α-diversity, comorbidity, and health care utilization (as captured from health events, hospitalizations, and medication use). All of these outcomes are hypothesized to improve in the immediate vs the delayed intervention arm at 1-year follow-up, with all secondary analyses considered exploratory. Mixed-models repeated-measures analysis will be performed using an appropriate structure for the covariance matrix to account for potential correlation among repeated measurements from the same participant. When a model term is statistically significant, the Tukey-Kramer multiple comparisons test will be used to determine which specific pairs’ means are significantly different. This modeling method allows for simultaneous estimation of within-group and between-group change. Models will include terms for group, time, and group × time.

      Mediators and Moderators

      For this trial, it is hypothesized that both increased self-efficacy and social support related to gardening, V&F intake, and PA will be significantly associated with the impact of the intervention. Logistic regression will be used to evaluate the association of potential moderators (eg, age, number of comorbidities, and sex) with improvements in outcomes. Interaction terms related to participant characteristics by treatment arm will be added to models to discern survivors for whom this intervention is more effective.

      Economic Analysis

      Intervention implementation costs and participants’ health care costs (based on self-reported health care utilization) will be calculated. The latter will be compared across arms to determine whether Harvest for Health leads to health care cost savings. A within-trial cost-effectiveness analysis
      • Glick H.A.
      • Doshi J.A.
      • Sonnad S.S.
      • et al.
      Economic Evaluation in Clinical Trials.
      • Neumann P.J.
      • Kim D.D.
      • Trikalinos T.A.
      • et al.
      Future directions for cost-effectiveness analyses in health and medicine.
      • Neumann P.J.
      • Sanders G.D.
      Cost-effectiveness analysis 2.0.
      • Weinstein M.C.
      • Russell L.B.
      • Gold M.R.
      • et al.
      Cost-Effectiveness in Health and Medicine.
      • Ramsey S.
      • Willke R.
      • Briggs A.
      • et al.
      Good research practices for cost-effectiveness analysis alongside clinical trials: The ISPOR RCT-CEA Task Force report.
      will be conducted to compare intervention implementation costs net of health care cost savings with the effectiveness of the intervention measured by the gain in quality-adjusted life-years derived from the 36-Item Short Form Survey.
      • Brazier J.E.
      • Roberts J.
      The estimation of a preference-based measure of health from the SF-12.
      ,
      • Kharroubi S.A.
      • Brazier J.E.
      • Roberts J.
      • et al.
      Modelling SF-6D health state preference data using a nonparametric Bayesian method.
      ,
      • McCabe C.
      • Brazier J.
      • Gilks P.
      • et al.
      Using rank data to estimate health state utility models.
      Incremental cost-effectiveness ratios will be calculated and compared with commonly used thresholds (eg, $50,000 per quality-adjusted life-year)
      • Neumann P.J.
      • Cohen J.T.
      • Weinstein M.C.
      Updating cost-effectiveness—The curious resilience of the $50,000-per-QALY threshold.
      to determine whether Harvest for Health is worth its cost.

      Modifications Due to COVID-19

      The COVID-19 pandemic necessitated several protocol modifications to protect study participants, MGs, and study staff, as well as to provide a means to continue study accrual, deliver the intervention with fidelity, and perform rigorous measures that would assure project continuity. For example, informed consent is no longer obtained using mailed forms, but rather facilitated by Adobe e-sign. In-person assessments are largely replaced by virtual visits wherein participants are asked to recruit a partner to record them as study staff lead them through anthropometric measures, performance testing, and blood collection on Zoom Meetings (Zoom) using password protected log-ins; participants and their partners are e-mailed 2 video links to view before these sessions (https://youtu.be/lbxctNuOgLk and https://youtu.be/lBPLS4PoHv4). Adaptations from in-person to virtual assessments are detailed in Figure 5,
      The 4-Stage Balance Test. Centers for Disease Control and Prevention.
      and require the following mailed supplies:
      • 2 ribbons and a felt-tip marker (to measure and mark waist circumference in duplicate);
      • an 8-foot length of cord and 2 stickers (to mark the distance for the 8-foot walk and up and go performance tests);
      • 2 plastic orange soccer cones (to increase visualization of the 8-foot walk course on virtual assessments);
      • a 36-inch vinyl tape measure and 1 sticker (to measure step height for the 2-minute step test);
      • a dried blood spot kit, which includes a collection card, lancet, foil storage pouch, and desiccant (to self-collect a blood sample); and
      • a prepaid, pre-addressed mailer to return the accelerometer, questionnaires (if not completed online), both ribbons, toenail clippings, fecal wipe, and dried blood spot card.
      Figure 5Study outcome modifications of the Harvest for Health vegetable gardening intervention in older cancer survivors in Alabama to overcome the challenges posed by COVID-19. aFor measures that have been omitted, analyses will still occur, but will only be performed on the sample that completed ≥12 months on study (ie, have both baseline and follow-up measures).
      OutcomeMeasure
      Primary outcome: Composite measure derived from the following subjective measures
      Vegetable and fruit intakeNo change
      Physical activityNo change
      Physical functionNo change
      Secondary outcomes: Objective measures below serve to validate subjective measures above
      Plasma α-caroteneIn adapting blood collection protocols from venipuncture to dried blood spots (DBS), there are 3 issues that precluded continued use of this measure: 1) inability to separate plasma; 2) rapid degradation of light sensitive carotenoids; and 3) nonexistence of carotenoid assays using DBS eluates.a
      Physical activityAccelerometry procedures are unchanged.
      Physical performanceThe following assessments from the Senior Fitness Battery are preserved, but implemented through Zoom after assuring safety and accuracy: 30-second chair stand; 8-foot get up and go test; 8-foot walk; sit and reach; back scratch; and 2-minute step test. Because the internet distorts the relative transmission of light and sound, Zoom calls were recorded for timed measures, then immediately replayed to extract the appropriate outcomes and recorded on REDCap, and then destroyed. Given the need for specialized equipment and budget restrictions, the 30-second arm curl and grip strength are omitted.a

      Balance testing: To circumvent the need for special equipment, the Centers for Disease Control and Prevention protocol for side by side, semi-tandem, and tandem stances was employed.
      The 4-Stage Balance Test. Centers for Disease Control and Prevention.
      To reduce ceiling effects, the latter test was extended for up to 2 minutes (or until the time at which the stance was broken).
      Other secondary outcomes
      Health-related quality of lifeNo change
      Health care utilizationNo change
      ComorbidityNo change
      Reassurance of worthNo change
      Sleep qualityNo change
      AnthropometricsMeasures are adapted to Zoom and include: acceptance of self-reported height, because it is unlikely to vary over the 2-year study; weight assessed on the participant’s scale or a mailed scale that is zeroed on camera; and waist circumference is taken in duplicate using the ribbons provided (see text). Participants bare their midriff on camera and staff instruct the partner to place the taped end of the ribbon on the umbilicus and then wrap it snugly around the waist; the participant is asked to rotate in front of the camera to assure the ribbon is flat and parallel to the ground; once assured, the partner marks the place of overlap with the felt-tipped marker provided. Both ribbons are sent back to the study office with other supplies. percent body fat via bioimpedance is not performed.a
      StressNo change
      InflammationNo change
      MicrobiomeNo change
      Currently, separate validation studies are underway (including interleukin-6 and tumor necrosis factor–α determinations from circulating sera and dried blood spot cards) and will guide whether data from pre– vs post–COVID-19 cohorts can be combined or require separate analyses.
      To promote safety during the gardening intervention, meet-and-greets were also adapted to Zoom Meetings and supplies distributed via “parking lot pick-ups,” in which masked cancer survivors remain in their vehicles while masked ACES staff load garden boxes, soil, and other supplies into their trunks. Also, both MGs and survivors were advised to restrict in-person meetings solely to outdoor interactions using ample personal protective equipment and adhering to social distancing; in-person meetings are advocated only if both the MG and survivor indicated mutual agreement and neither are currently COVID-19–positive or –exposed.

      Results

      As indicated, recruitment for Harvest for Health is now complete. Figure 1 details the study trajectory from self-referral or registry ascertainment to study completion. Notable challenges include outdated status and invalid contact information from registries, as evidenced by statistics showing that approximately 2% of cancer registry patients were deceased, approximately 4% had inactive mailing addresses, and approximately 24% were unreachable by telephone. Of the 13,738 cancer survivors ascertained, verified contact could only be assured among 9,708 (self-referrals [326 cases, 2 that could not be verified] + cancer registry referrals [13,412 cases – 511 undeliverable/returned letters – 316 decedents – 1,954 no answer/unverified telephone numbers – 1,247 disconnected telephones]). Of these cancer survivors, most refused participation, with leading reasons for disinterest reported as “physically unable” (18.5%) and “lack of time” (12.2%) or “lack of space” (7.0%),” although general disinterest ranked uppermost. Nonetheless, a substantial proportion (15% response rate) of older cancer survivors expressed interest in participation and underwent screening, with roughly two-thirds screening out. Leading reasons for ineligibility were already vegetable gardening (41.7%) or exercising (16.2%), or medical exclusions (13.7%). Throughout the recruitment period, the OCCC-UAB study team obtained signed consent from 473 eligible older cancer survivors. However, completion of baseline assessments was thwarted by natural disasters (2 serious tornado systems and 1 hurricane that ravaged several counties); the 2017-2018 influenza outbreak; and, finally, COVID-19. Altogether these events caused substantial delays; however, the study is now in the field and completion is anticipated by June 2022. Table 1 presents characteristics of those who were interested in participating in the gardening intervention (vs those who did not respond or refused), as well as those who screened eligible and enrolled (vs those who were not enrolled).
      Table 1Characteristics of the 9,708 cancer survivors who were contacted for participation in the Harvest for Health randomized controlled trial: Those who expressed interest vs not and those who were deemed eligible and enrolled vs not
      CharacteristicContacted pool of cancer survivors (n = 9,708)Cancer Survivors Expressing Interest in the Intervention vs Unresponsive/RefusedCancer Survivors Enrolled vs Unenrolled in the Randomized Controlled Trial
      Interested (n = 1,460)Unresponsive/refused (n = 8,248)P value
      Test for significance (t test for age or χ2 for the other characteristics) for between-group comparisons for interested vs unresponsive/refused and enrolled vs unenrolled (note that for the comparisons on race and ethnicity that only non-Hispanic White and non-Hispanic Black cancer survivors were compared because the proportion of others/refuse/missing was low; missing values were excluded from the comparisons of county of residence and of age).
      Enrolled (n = 381)Unenrolled (n = 9,327)P value
      Test for significance (t test for age or χ2 for the other characteristics) for between-group comparisons for interested vs unresponsive/refused and enrolled vs unenrolled (note that for the comparisons on race and ethnicity that only non-Hispanic White and non-Hispanic Black cancer survivors were compared because the proportion of others/refuse/missing was low; missing values were excluded from the comparisons of county of residence and of age).
      ←mean (SD) →←mean (SD)→
      Age, y71.5 (7.4)69.7 (6.7)71.9 (7.4)<.00169.8 (6.4)71.6 (7.4)<.001
      ←% (n)→←% (n)→
      Female sex55.7 (5,407)63.1 (921)54.4 (4,486)<.00169.0 (263)55.2 (5,144)<.001
      Race and ethnicity.009.79
      Non-Hispanic White78.9 (7,661)67.9 (991)80.9 (6,670)80.6 (307)78.8 (7,354)
      Non-Hispanic Black17.4 (1,688)17.7 (259)17.3 (1,429)18.4 (70)17.3 (1,618)
      Other
      Other is composed of cancer survivors of Hispanic ethnicity, as well as who identify with racial groups described as Asian, American Indian, or multi-race.
      /refused/missing
      3.7 (359)14.4 (210)1.8 (149)1.0 (4)3.8 (335)
      Rural county of residence29.0 (2,816)27.9 (408)29.2 (2,408).6226.5 (101)29.1 (2,715).28
      Cancer type<.001<.001
      Breast29.7 (2,879)31.9 (466)29.3 (2,413)39.1 (149)29.3 (2,730)
      Prostate20.0 (1,940)15.2 (222)20.8 (1,718)13.4 (51)20.3 (1,889)
      Colorectal8.9 (865)7.0 (102)9.3 (763)6.6 (25)9.0 (840)
      Other/missing41.4 (4,024)45.9 (670)40.7 (3,354)40.9 (156)41.5 (3,868)
      a Test for significance (t test for age or χ2 for the other characteristics) for between-group comparisons for interested vs unresponsive/refused and enrolled vs unenrolled (note that for the comparisons on race and ethnicity that only non-Hispanic White and non-Hispanic Black cancer survivors were compared because the proportion of others/refuse/missing was low; missing values were excluded from the comparisons of county of residence and of age).
      b Other is composed of cancer survivors of Hispanic ethnicity, as well as who identify with racial groups described as Asian, American Indian, or multi-race.
      The recruitment of MGs is also key to this study and Harvest for Health has been buoyed by strong MG support. More than 300 MGs across the state volunteered to serve as gardening mentors, with most serving more than 1 year. In some cases, the number of eligible cancer survivors and the number of available MGs within a county were misaligned, as evidenced by very strong MG organizations in some rural counties that had fairly low numbers of cancer survivors (especially those who were not already tending vegetable gardens); however, these discrepancies were fairly rare and some MGs were willing to travel to adjacent counties that had more cancer survivors and greater need. A surplus of additional MGs was helpful in instances when the originally assigned MG was unable to fulfill the year-long commitment. MGs in New Mexico and Vermont are already participating in other home-based or hospital-based vegetable gardening projects,
      • Blair C.K.
      • Harding E.M.
      • Adsul P.
      • et al.
      Southwest Harvest for Health: Adapting a mentored vegetable gardening intervention for cancer survivors in the southwest.
      ,
      Gardening for Health Pilot Program
      The University of Vermont Medical Center.
      and MGs in several other states have expressed interest (eg, Tennessee and Washington); however, the capacity to implement vegetable gardening projects rests largely on MGs’ willingness to volunteer and competing local priorities. A compilation of the number of MGs for most states in the United States and their estimated number of volunteer hours was tabulated from a recent 2020 report (Table 2).
      Extension Master Gardener 2020 National Report Overview. US Department of Agriculture.
      Table 22020 Extension master gardener national report overview
      Extension Master Gardener 2020 National Report Overview. US Department of Agriculture.
      StateEstimated no. of volunteer hoursEstimated no. of volunteers (new/existing/returning)
      Alabama127,5001,891
      Arizona29,6001,003
      Arkansas100,4003,763
      California328,3006,200
      Colorado44,3001,457
      Delaware7,700237
      Florida156,8005,257
      Georgia111,1002,253
      Illinois92,7003,439
      Indiana98,6003,120
      Iowa65,8001,594
      Kansas36,500640
      Kentucky11,3001,111
      Louisiana39,8001,272
      Maryland37,4002,319
      Michigan51,2002,677
      Minnesota69,8003,034
      Montana6,600434
      Nebraska15,500782
      Nevada9,000388
      New Hampshire9,200621
      New Mexico30,5001,136
      New York96,9001,860
      North Carolina109,6002,978
      North Dakota11,900262
      Oklahoma36,2001,521
      Oregon77,6002,703
      Pennsylvania114,4003,533
      Rhode Island29,800702
      South Carolina45,500898
      South Dakota3,100385
      Tennessee125,8003,133
      Texas450,0007,300
      Vermont6,200366
      Virginia250,0005,620
      Washington203,7004,751
      West Virginia16,2001,416
      Wisconsin39,3002,854
      WyomingNo data265

      Discussion

      The Harvest for Health home-based vegetable gardening intervention represents a novel and multifaceted approach aimed at improving a triad of important outcomes (ie, PF, V&F consumption, and PA) among a growing patient population that is at risk for comorbidity, second cancers, and functional decline.
      Institute of Medicine, National Research Council
      From Cancer Patient to Cancer Survivors: Lost in Transition.
      • Travis L.B.
      • Demark Wahnefried W.
      • Allan J.M.
      • et al.
      Aetiology, genetics and prevention of secondary neoplasms in adult cancer survivors.
      • Armenian S.H.
      • Xu L.
      • Ky B.
      • et al.
      Cardiovascular disease among survivors of adult-onset cancer: A community-based retrospective cohort study.
      • Schoormans D.
      • Vissers P.A.
      • van Herk-Sukel M.P.
      • et al.
      Incidence of cardiovascular disease up to 13 year after cancer diagnosis: A matched cohort study among 32 757 cancer survivors.
      • Strongman H.
      • Gadd S.
      • Matthews A.
      • et al.
      Medium and long-term risks of specific cardiovascular diseases in survivors of 20 adult cancers: A population-based cohort study using multiple linked UK electronic health records databases.
      • Donin N.
      • Filson C.
      • Drakaki A.
      • et al.
      Risk of second primary malignancies among cancer survivors in the United States, 1992 through 2008.
      • Sung H.
      • Hyun N.
      • Leach C.R.
      • et al.
      Association of first primary cancer with risk of subsequent primary cancer among survivors of adult-onset cancers in the United States.
      • Barzi A.
      • Hershman D.L.
      • Till C.
      • et al.
      Osteoporosis in colorectal cancer survivors: Analysis of the linkage between SWOG trial enrollees and Medicare claims.
      • Shapiro C.L.
      • Van Poznak C.
      • Lacchetti C.
      • et al.
      Management of osteoporosis in survivors of adult cancers with nonmetastatic disease: ASCO Clinical Practice Guideline.
      • Anderson C.
      • Gapstur S.M.
      • Leach C.R.
      • et al.
      Medical conditions and physical function deficits among multiple primary cancer survivors.
      • Blair C.K.
      • Jacobs Jr., D.R.
      • Demark-Wahnefried W.
      • et al.
      Effects of cancer history on functional age and mortality.
      • Garcia S.F.
      • Wortman K.
      • Cella D.
      • et al.
      Implementing electronic health record-integrated screening of patient-reported symptoms and supportive care needs in a comprehensive cancer center.
      • Leach C.R.
      • Bellizzi K.M.
      • Hurria A.
      • et al.
      Is it my cancer or am I just getting older?: Impact of cancer on age-related health conditions of older cancer survivors.
      Although there have been feasibility studies of vegetable gardening interventions conducted in community
      • Spees C.K.
      • Hill E.B.
      • Grainger E.M.
      • et al.
      Feasibility, preliminary efficacy, and lessons learned from a garden-based lifestyle intervention for cancer survivors.
      and home-based settings,
      • Blair C.K.
      • Madan-Swain A.
      • Locher J.L.
      • et al.
      Harvest for health gardening intervention feasibility study in cancer survivors.
      • Bail J.R.
      • Fruge A.D.
      • Cases M.G.
      • et al.
      A home-based mentored vegetable gardening intervention demonstrates feasibility and improvements in physical activity and performance among breast cancer survivors.
      • Cases M.G.
      • Fruge A.D.
      • De Los Santos J.F.
      • et al.
      Detailed methods of two home-based vegetable gardening intervention trials to improve diet, physical activity, and quality of life in two different populations of cancer survivors.
      all of which show promise in affecting lifestyle behaviors and other health outcomes among cancer survivors, all of these studies have relied on small samples and are underpowered. The Harvest for Health RCT described in this article will be the first fully powered trial to test the impact of a vegetable gardening intervention. Similar to other trials,
      • McDonald A.M.
      • Knight R.C.
      • Campbell M.K.
      • et al.
      What influences recruitment to randomised controlled trials? A review of trials funded by two UK funding agencies.
      recruitment for Harvest for Health has been costly and labor-intensive. As indicated, the data on accrual are important to appropriately generalize the trial outcomes, and data on interest may be useful for future dissemination beyond the RCT. A significantly higher proportion of breast cancer survivors, as well as survivors who were relatively younger and female, expressed interest and were ultimately enrolled. Although lower positive response rates were observed among non-Hispanic White survivors, these differences diminished on enrollment; moreover, no urban-rural differences were observed with regard to either levels of interest or enrollment.

      Conclusions

      Cancer survivors experience long-term adverse sequelae that reduce PF and that may negatively impact independent living and HRQOL. Interventions that are effective, durable, and disseminable are needed urgently. Results from this study will provide evidence on the effects of vegetable gardening among older cancer survivors. If effective, this intervention could be disseminated broadly. Already Harvest for Health has been adapted for use in the Southwestern United States,
      • Blair C.K.
      • Harding E.M.
      • Adsul P.
      • et al.
      Southwest Harvest for Health: Adapting a mentored vegetable gardening intervention for cancer survivors in the southwest.
      and could lay the groundwork for a pragmatic trial that is national in scope. The home-based intervention and the remote-based assessment methods that were adapted in response to COVID-19 lay the groundwork for dissemination of this particular intervention and can also be adapted to expand the reach of other diet-related research projects.

      Acknowledgement

      The authors thank all of the cancer survivors, master gardener volunteers, and Alabama Cooperative Extension agents for their participation, especially Leonora Roberson, a master gardener who earns special mention. Also deserving of special mention are Tony Glover and Renee Thompson, both are Harvest for Health heroes! The authors also acknowledge the tremendous contributions and dedication of pre- and postdoctoral fellows, student interns, work study students, and staff members (especially Walker Cole, MPH, Teri Hoenemeyer, PhD, and Daniel Edwards, MLS, ASCPCM). Thanks also go to the study advisors, Jennifer De Los Santos, MD Mark Conaway, PhD, and Patricia Ganz, MD, as well as Madeline Harris, RN, MSN, OCN, and Amy Clayton. Donations made by the following organizations were also appreciated: Cawaco Resource, Conservation & Development Council, Birmingham, AL (seeds); Edgewell Personal Care, LLC, Shelton, CT (sunscreen); Hannah’s Garden Shop, Birmingham, AL (seeds); Johnny’s Selected Seeds, Fairfield, ME (seeds); Safer Brand, Lititz, PA; Scott’s Miracle-Gro, Marysville, OH; Sway Medical, Aledo, TX (balance testing app); and Walmart Charitable Community Giving, Bentonville, AR (various supplies). Finally, appreciation also goes to the private donors whose financial donations were instrumental in helping the entire project. In closing, we also remember Dr Meredith Kilgore, a colleague and friend. Note, that permissions have been obtained for all individuals who are specifically named above (or from their family members, if they are now deceased).

      Author Contributions

      Conceptualization: W. Demark-Wahnefried, C. K. Blair, R. A. Oster, J. L. Locher, and K. P. Smith. Funding acquisition: W. Demark-Wahnefried, R. A. Oster, J. L. Locher, and K. P. Smith. Data collection and recruitment: W. Demark-Wahnefried, J. R. Bail, K. P. Smith, H. Kaur, A. D. Frugé, and G. Rocque. Formal analysis: R. A. Oster. Writing (original draft): J. R. Bail, C. K. Blair, and W. Demark-Wahnefried. Writing (review and editing): J. R. Bail, C. K. Blair, K. P. Smith, R. A. Oster, H. Kaur, J. L. Locher, A. D. Frugé, G. Rocque, M. Pisu, H. J. Cohen, and W. Demark-Wahnefried. All authors have read and agreed to the published version of the manuscript. J. R. Bail and C. K. Blair share co-first authorship.

      References

        • Miller K.D.
        • Nogueira L.
        • Mariotto A.B.
        • et al.
        Cancer treatment and survivorship statistics, 2019.
        CA Cancer J Clin. 2019; 69: 363-385
        • Institute of Medicine, National Research Council
        From Cancer Patient to Cancer Survivors: Lost in Transition.
        National Academies Press, 2005
        • Travis L.B.
        • Demark Wahnefried W.
        • Allan J.M.
        • et al.
        Aetiology, genetics and prevention of secondary neoplasms in adult cancer survivors.
        Nat Rev Clin Oncol. 2013; 10: 289-301
        • Armenian S.H.
        • Xu L.
        • Ky B.
        • et al.
        Cardiovascular disease among survivors of adult-onset cancer: A community-based retrospective cohort study.
        J Clin Oncol. 2016; 34: 1122-1130
        • Schoormans D.
        • Vissers P.A.
        • van Herk-Sukel M.P.
        • et al.
        Incidence of cardiovascular disease up to 13 year after cancer diagnosis: A matched cohort study among 32 757 cancer survivors.
        Cancer Med. 2018; 7: 4952-4963
        • Strongman H.
        • Gadd S.
        • Matthews A.
        • et al.
        Medium and long-term risks of specific cardiovascular diseases in survivors of 20 adult cancers: A population-based cohort study using multiple linked UK electronic health records databases.
        Lancet. 2019; 394: 1041-1054
        • Donin N.
        • Filson C.
        • Drakaki A.
        • et al.
        Risk of second primary malignancies among cancer survivors in the United States, 1992 through 2008.
        Cancer. 2016; 122: 3075-3086
        • Sung H.
        • Hyun N.
        • Leach C.R.
        • et al.
        Association of first primary cancer with risk of subsequent primary cancer among survivors of adult-onset cancers in the United States.
        JAMA. 2020; 324: 2521-2535
        • Barzi A.
        • Hershman D.L.
        • Till C.
        • et al.
        Osteoporosis in colorectal cancer survivors: Analysis of the linkage between SWOG trial enrollees and Medicare claims.
        Arch Osteoporos. 2019; 14: 83
        • Shapiro C.L.
        • Van Poznak C.
        • Lacchetti C.
        • et al.
        Management of osteoporosis in survivors of adult cancers with nonmetastatic disease: ASCO Clinical Practice Guideline.
        J Clin Oncol. 2019; 37: 2916-2946
        • Anderson C.
        • Gapstur S.M.
        • Leach C.R.
        • et al.
        Medical conditions and physical function deficits among multiple primary cancer survivors.
        J Cancer Surviv. 2020; 14: 518-526
        • Blair C.K.
        • Jacobs Jr., D.R.
        • Demark-Wahnefried W.
        • et al.
        Effects of cancer history on functional age and mortality.
        Cancer. 2019; 125: 4303-4309
        • Garcia S.F.
        • Wortman K.
        • Cella D.
        • et al.
        Implementing electronic health record-integrated screening of patient-reported symptoms and supportive care needs in a comprehensive cancer center.
        Cancer. 2019; 125: 4059-4068
        • Leach C.R.
        • Bellizzi K.M.
        • Hurria A.
        • et al.
        Is it my cancer or am I just getting older?: Impact of cancer on age-related health conditions of older cancer survivors.
        Cancer. 2016; 122: 1946-1953
        • Denlinger C.S.
        • Sanft T.
        • Moslehi J.J.
        • et al.
        NCCN Guidelines Insights: Survivorship, Version 2.2020.
        J Natl Compr Canc Netw. 2020; 18: 1016-1023
        • Morey M.C.
        • Snyder D.C.
        • Sloane R.
        • et al.
        Effects of home-based diet and exercise on functional outcomes among older, overweight long-term cancer survivors: RENEW: A randomized controlled trial.
        JAMA. 2009; 301: 1883-1891
        • Demark-Wahnefried W.
        • Morey M.C.
        • Sloane R.
        • et al.
        Reach Out to Enhance Wellness home-based diet-exercise intervention promotes reproducible and sustainable long-term improvements in health behaviors, body weight, and physical functioning in older, overweight/obese cancer survivors.
        J Clin Oncol. 2012; 30: 2354-2361
        • Demark-Wahnefried W.
        • Clipp E.C.
        • Lipkus I.M.
        • et al.
        Main outcomes of the FRESH START trial: A sequentially tailored, diet and exercise mailed print intervention among breast and prostate cancer survivors.
        J Clin Oncol. 2007; 25: 2709-2718
        • Leng C.H.
        • Wang J.D.
        Daily home gardening improved survival for older people with mobility limitations: An 11-year follow-up study in Taiwan.
        Clin Interv Aging. 2016; 11: 947-959
        • Park S.
        • Shoemaker C.
        • Haub M.
        Can older gardeners meet the physical activity recommendation through gardening?.
        Horttechnology. 2008; 18: 639-643
        • Soga M.
        • Cox D.T.
        • Yamaura Y.
        • et al.
        Health benefits of urban allotment gardening: Improved physical and psychological well-being and social integration.
        Int J Environ Res Public Health. 2017; 14: 71
        • Van den Berg A.E.
        • Van Winsum-Westra M.
        • De Vries S.
        • et al.
        Allotment gardening and health: A comparative survey among allotment gardeners and their neighbors without an allotment.
        Environ Health. 2010; 9: 1-12
        • Barnidge E.K.
        • Hipp P.R.
        • Estlund A.
        • et al.
        Association between community garden participation and fruit and vegetable consumption in rural Missouri.
        Int J Behav Nutr Phys Act. 2013; 10: 128
        • Sommerfeld A.J.
        • McFarland A.L.
        • Waliczek T.M.
        • et al.
        Growing minds: Evaluating the relationship between gardening and fruit and vegetable consumption in older adults.
        Horttechnology. 2010; 20: 711-717
        • Machida D.
        Relationship between community or home gardening and health of the elderly: A web-based cross-sectional survey in Japan.
        Int J Environ Res Public Health. 2019; 16: 1389
        • Kegler M.C.
        • Prakash R.
        • Hermstad A.
        • et al.
        Home gardening and associations with fruit and vegetable intake and BMI.
        Public Health Nutr. 2020; 23: 3417-3422
        • Scott T.L.
        • Masser B.M.
        • Pachana N.A.
        Positive aging benefits of home and community gardening activities: Older adults report enhanced self-esteem, productive endeavours, social engagement and exercise.
        SAGE Open Med. 2020; 82050312120901732
        • Chen T.Y.
        • Janke M.C.
        Gardening as a potential activity to reduce falls in older adults.
        J Aging Phys Act. 2012; 20: 15-31
        • Chen T.Y.
        • Janke M.C.
        Effectiveness of gardening activities on improving older adults' gait and balance.
        Gerontologist. 2010; 50: 116
        • Strout K.
        • Jemison J.
        • O'Brien L.
        • et al.
        GROW: Green Organic Vegetable Gardens to Promote Older Adult Wellness: A feasibility study.
        J Community Health Nurs. 2017; 34: 115-125
        • Lee R.E.
        • Parker N.H.
        • Soltero E.G.
        • et al.
        Sustainability via Active Garden Education (SAGE): Results from two feasibility pilot studies.
        BMC Public Health. 2017; 17: 242
        • Davis J.N.
        • Martinez L.C.
        • Spruijt-Metz D.
        • et al.
        LA Sprouts: A 12-week gardening, nutrition, and cooking randomized control trial improves determinants of dietary behaviors.
        J Nutr Educ Behav. 2016; 48: 2-11
        • Gatto N.M.
        • Martinez L.C.
        • Spruijt-Metz D.
        • et al.
        LA Sprouts randomized controlled nutrition, cooking and gardening programme reduces obesity and metabolic risk in Hispanic/Latino youth.
        Pediatr Obes. 2017; 12: 28-37
        • Wood C.J.
        • Pretty J.
        • Griffin M.
        A case-control study of the health and well-being benefits of allotment gardening.
        J Public Health (Oxf). 2016; 38: e336-e344
        • Spees C.K.
        • Hill E.B.
        • Grainger E.M.
        • et al.
        Feasibility, preliminary efficacy, and lessons learned from a garden-based lifestyle intervention for cancer survivors.
        Cancer Control. 2016; 23: 302-310
        • Hume A.
        • Wetten A.
        • Feeney C.
        • et al.
        Remote school gardens: Exploring a cost-effective and novel way to engage Australian Indigenous students in nutrition and health.
        Aust N Z J Public Health. 2014; 38: 235-240
        • White P.C.
        • Wyatt J.
        • Chalfont G.
        • et al.
        Exposure to nature gardens has time-dependent associations with mood improvements for people with mid- and late-stage dementia: Innovative practice.
        Dementia (London). 2018; 17: 627-634
        • Brown V.M.
        • Allen A.C.
        • Dwozan M.
        • et al.
        Indoor gardening older adults: Effects on socialization, activities of daily living, and loneliness.
        J Gerontol Nurs. 2004; 30: 34-42
        • Lee Y.
        • Kim S.
        Effects of indoor gardening on sleep, agitation, and cognition in dementia patients—A pilot study.
        Int J Geriatr Psychiatry. 2008; 23: 485-489
        • Kerr J.
        • Marshall S.
        • Godbole S.
        • et al.
        The relationship between outdoor activity and health in older adults using GPS.
        Int J Environ Res Public Health. 2012; 9: 4615-4625
        • Kerr J.
        • Sallis J.F.
        • Saelens B.E.
        • et al.
        Outdoor physical activity and self rated health in older adults living in two regions of the U.S.
        Int J Behav Nutr Phys Act. 2012; 9: 89
        • Marsh A.P.
        • Katula J.A.
        • Pacchia C.F.
        • et al.
        Effect of treadmill and overground walking on function and attitudes in older adults.
        Med Sci Sports Exerc. 2006; 38: 1157-1164
        • Sommerfield A.J.
        • Zajicek J.
        • Waliczek T.M.
        Growing minds: Evaluating the effect of gardening on quality of life in older adults.
        Hortscience. 2009; 44 (1046-1046)
        • Cooperative Extension System
        US Department of Agriculture, National Institute of Food and Agriculture.
        • State cancer profiles
        National Program of Cancer Registries.
        http://statecancerprofiles.cancer.gov
        Date accessed: March 1, 2022
        • Blair C.K.
        • Harding E.M.
        • Adsul P.
        • et al.
        Southwest Harvest for Health: Adapting a mentored vegetable gardening intervention for cancer survivors in the southwest.
        Contemp Clin Trials Commun. 2021; 21: 100741
        • Blair C.K.
        • Madan-Swain A.
        • Locher J.L.
        • et al.
        Harvest for health gardening intervention feasibility study in cancer survivors.
        Acta Oncol. 2013; 52: 1110-1118
        • Bail J.R.
        • Fruge A.D.
        • Cases M.G.
        • et al.
        A home-based mentored vegetable gardening intervention demonstrates feasibility and improvements in physical activity and performance among breast cancer survivors.
        Cancer. 2018; 124: 3427-3435
        • Cases M.G.
        • Fruge A.D.
        • De Los Santos J.F.
        • et al.
        Detailed methods of two home-based vegetable gardening intervention trials to improve diet, physical activity, and quality of life in two different populations of cancer survivors.
        Contemp Clin Trials. 2016; 50: 201-212
        • Demark-Wahnefried W.
        • Cases M.G.
        • Cantor A.B.
        • et al.
        Pilot randomized controlled trial of a home vegetable gardening intervention among older cancer survivors shows feasibility, satisfaction, and promise in improving vegetable and fruit consumption, reassurance of worth, and the trajectory of central adiposity.
        J Acad Nutr Diet. 2018; 118: 689-704
        • Patterson R.E.
        • Marinac C.R.
        • Natarajan L.
        • et al.
        Recruitment strategies, design, and participant characteristics in a trial of weight-loss and metformin in breast cancer survivors.
        Contemp Clin Trials. 2016; 47: 64-71
        • Wang F.
        • Shu X.
        • Meszoely I.
        • et al.
        Overall mortality after diagnosis of breast cancer in men vs women.
        JAMA Oncol. 2019; 5: 1589-1596
      1. Physical activity intensity. Centers for Disease Control and Prevention.
        • Ware J.E.
        • Kosinski M.
        SF-36® Physical & Mental Health Summary Scales: A Manual for Users of Version 2.
        2nd ed. QualityMetric, 2007
        • Thompson F.E.
        • Subar A.F.
        • Smith A.F.
        • et al.
        Fruit and vegetable assessment: Performance of 2 new short instruments and a food frequency questionnaire.
        J Am Diet Assoc. 2002; 102: 1764-1772
        • Stewart A.L.
        • Mills K.M.
        • King A.C.
        • et al.
        CHAMPS physical activity questionnaire for older adults: Outcomes for interventions.
        Med Sci Sports Exerc. 2001; 33: 1126-1141
        • Bieri J.G.
        • Brown E.D.
        • Smith J.C.
        Determination of individual carotenoids in human plasma by high performance liquid chromatography.
        J Liq Chromatogr. 1985; 8: 473-484
        • Rikli R.E.
        • Jones C.J.
        Development and validation of a functional fitness test for community-residing older adults.
        J Aging Phys Act. 1999; 7: 129-161
        • Taekema D.G.
        • Gussekloo J.
        • Maier A.B.
        • et al.
        Handgrip strength as a predictor of functional, psychological and social health. A prospective population-based study among the oldest old.
        Age Ageing. 2010; 39: 331-337
        • Brazier J.E.
        • Roberts J.
        The estimation of a preference-based measure of health from the SF-12.
        Med Care. 2004; 42: 851-859
        • Fillenbaum G.G.
        Multidimensional Functional Assessment of Older Adults: The Duke Older Americans Resources and Services Procedures.
        Lawrence Erlbaum Associates, 1988: 7-12
        • Cutrona C.E.
        • Russell D.
        The provisions of social relationships and adaptation stress.
        in: Jones W.H. Perlman D. Advances in Personal Relationships. JAI Press, 1987: 37-67
        • Buysse D.J.
        • Reynolds 3rd, C.F.
        • Monk T.H.
        • et al.
        The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research.
        Psychiatry Res. 1989; 28: 193-213
        • Fruge A.D.
        • Cases M.G.
        • Howell C.R.
        • et al.
        Fingernail and toenail clippings as a non-invasive measure of chronic cortisol levels in adult cancer survivors.
        Cancer Causes Control. 2018; 29: 185-191
        • Warnock F.
        • McElwee K.
        • Seo R.J.
        • et al.
        Measuring cortisol and DHEA in fingernails: A pilot study.
        Neuropsychiatr Dis Treat. 2010; 6: 1-7
        • Fruge A.D.
        • Van der Pol W.
        • Rogers L.Q.
        • et al.
        Fecal Akkermansia muciniphila is associated with body composition and microbiota diversity in overweight and obese women with breast cancer participating in a presurgical weight loss trial.
        J Acad Nutr Diet. 2020; 120: 650-659
        • Caporaso J.G.
        • Kuczynski J.
        • Stombaugh J.
        • et al.
        QIIME allows analysis of high-throughput community sequencing data.
        Nat Methods. 2010; 7: 335-336
        • Caporaso J.G.
        • Lauber C.L.
        • Walters W.A.
        • et al.
        Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample.
        Proc Natl Acad Sci U S A. 2011; 108: 4516-4522
        • Human Microbiome Project Consortium
        Structure, function and diversity of the healthy human microbiome.
        Nature. 2012; 486: 207-214
        • Faith D.P.
        • Baker A.M.
        Phylogenetic diversity (PD) and biodiversity conservation: Some bioinformatics challenges.
        Evol Bioinform Online. 2007; 2: 121-128
        • Sallis J.F.
        • Pinski R.B.
        • Grossman R.M.
        • et al.
        The development of self-efficacy scales for healthrelated diet and exercise behaviors.
        Health Educ Res. 1988; 3: 283-292
        • Bandura A.
        Social Learning Theory.
        Prentice Hall, 1977
        • Mosher C.E.
        • Lipkus I.
        • Sloane R.
        • et al.
        Long-term outcomes of the FRESH START trial: Exploring the role of self-efficacy in cancer survivors' maintenance of dietary practices and physical activity.
        Psychooncology. 2013; 22: 876-885
        • Bronfenbrenner U.
        The Ecology of Human Development, Experiments by Nature and Design.
        Harvard University Press, 1979
        • Bellg A.J.
        • Borrelli B.
        • Resnick B.
        • et al.
        Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH Behavior Change Consortium.
        Health Psychol. 2004; 23: 443-451
      2. SAS [computer program]. Version 9.4. SAS Institute, 2013
        • Brevik A.
        • Andersen L.F.
        • Karlsen A.
        • Trygg K.U.
        • Blomhoff R.
        • Drevon C.A.
        Six carotenoids in plasma used to assess recommended intake of fruits and vegetables in a controlled feeding study.
        Eur J Clin Nutr. 2004; 58: 1166-1173
        • Cohen J.
        Statistical Power Analysis for the Behavioral Sciences.
        2nd ed. Hove and London, 1988
        • Glick H.A.
        • Doshi J.A.
        • Sonnad S.S.
        • et al.
        Economic Evaluation in Clinical Trials.
        Oxford University Press, 2007
        • Neumann P.J.
        • Kim D.D.
        • Trikalinos T.A.
        • et al.
        Future directions for cost-effectiveness analyses in health and medicine.
        Med Decis Making. 2018; 38: 767-777
        • Neumann P.J.
        • Sanders G.D.
        Cost-effectiveness analysis 2.0.
        N Engl J Med. 2017; 376: 203-205
        • Weinstein M.C.
        • Russell L.B.
        • Gold M.R.
        • et al.
        Cost-Effectiveness in Health and Medicine.
        Oxford University Press, 1996
        • Ramsey S.
        • Willke R.
        • Briggs A.
        • et al.
        Good research practices for cost-effectiveness analysis alongside clinical trials: The ISPOR RCT-CEA Task Force report.
        Value Health. 2005; 8: 521-533
        • Kharroubi S.A.
        • Brazier J.E.
        • Roberts J.
        • et al.
        Modelling SF-6D health state preference data using a nonparametric Bayesian method.
        J Health Econ. 2007; 26: 597-612
        • McCabe C.
        • Brazier J.
        • Gilks P.
        • et al.
        Using rank data to estimate health state utility models.
        J Health Econ. 2006; 25: 418-431
        • Neumann P.J.
        • Cohen J.T.
        • Weinstein M.C.
        Updating cost-effectiveness—The curious resilience of the $50,000-per-QALY threshold.
        N Engl J Med. 2014; 371: 796-797
      3. The 4-Stage Balance Test. Centers for Disease Control and Prevention.
        • Gardening for Health Pilot Program
        The University of Vermont Medical Center.
      4. Extension Master Gardener 2020 National Report Overview. US Department of Agriculture.
        • McDonald A.M.
        • Knight R.C.
        • Campbell M.K.
        • et al.
        What influences recruitment to randomised controlled trials? A review of trials funded by two UK funding agencies.
        Trials. 2006; 7: 9

      Biography

      J. R. Bail is an assistant professor, College of Nursing, University of Alabama, Huntsville. At the time of the study she was a postdoctoral fellow in the Demark-Wahnefried Laboratory, Department of Nutrition Sciences, University of Alabama, Birmingham.
      C. K. Blair is an associate professor, Department of Internal Medicine, University of New Mexico, Albuquerque. At the time of the study she was a postdoctoral fellow in the Demark-Wahnefried Laboratory, Department of Nutrition Sciences, University of Alabama, Birmingham.
      K. P. Smith is the state master gardener program coordinator, Alabama Cooperative Extension System, Auburn University, Auburn, AL.
      R. A. Oster is a professor, Division of Preventive Medicine, Department of Medicine, University of Alabama, Birmingham.
      H. Kaur is a doctoral student, Department of Nutrition Sciences, University of Alabama, Birmingham.
      J. L. Locher is a professor emeritus, Division of Geriatrics, Department of Medicine, University of Alabama, Birmingham.
      A. D. Frugé is an assistant professor, Department of Nutrition, Dietetics and Hospitality Management, Auburn University, Auburn, AL. At the time of the study he was a postdoctoral fellow in the Demark-Wahnefried Laboratory, Department of Nutrition Sciences, University of Alabama, Birmingham.
      G. Rocque is an assistant professor, Department of Medicine, University of Alabama, Birmingham.
      M. Pisu is a professor, Division of Preventive Medicine, Department of Medicine, University of Alabama, Birmingham.
      H. J. Cohen is a professor, Department of Medicine, Duke University Medical Center, Durham, NC.
      W. Demark-Wahnefried is a professor, Department of Nutrition Sciences, University of Alabama, Birmingham.