Vegetable Gardening and Health Outcomes in Older Cancer Survivors: A Randomized Clinical Trial.

Introduction

The benefits of gardening were first reported by Cicero in 46 BC. 1 More recently, a meta-analysis of 1046 older adults from 13 randomized clinical trials (RCTs) and 2 cohort studies found that gardening was associated with improved physical functioning, increased physical activity, reduced body mass index (BMI), and improved quality of life. 2 A systematic review also found that access to one’s own vegetable garden was an environmental factor associated with vegetable and fruit intake. 3 Thus, gardening, especially vegetable gardening, may be associated with holistic health benefits and address the wide-ranging needs of cancer survivors at risk for a second malignant tumor, obesity, cardiovascular disease, other comorbidities, and functional impairment. 4 , 5 , 6

Over a decade ago, the Harvest for Health feasibility trial was conducted among 12 survivors of breast, prostate, and childhood cancers residing in Jefferson County, Alabama. 7 It was a partnered effort between an academic medical center (which recruited and enrolled participants and assessed changes in diet, physical activity, and physical functioning) and the Auburn University Horticulture Extension Office cooperative extension (which deployed extension-certified master gardeners [EMGs] to provide biweekly mentorship on planning, planting, and caring for a spring, summer, and fall raised-bed garden at participants’ homes). The trial demonstrated full accrual within a week, 83.3% retention over 1 year, and safety. 7 Preintervention and postintervention assessments suggested improvements in physical performance, 8 increased vegetable and fruit intake, and increased moderate-to-vigorous physical activity (MVPA). 7 The study was expanded to 82 breast cancer survivors residing in metropolitan Birmingham, Alabama, and then statewide to 46 older survivors of various cancers using a 2-arm RCT design with comparison against a waitlist control. 9 Both trials evaluated the Harvest for Health intervention, which was modified to supply either raised beds or 4 grow boxes of comparable square footage to increase reach to survivors who resided in condominiums or rental properties. Findings of both trials demonstrated feasibility (ie, 82%-100% accrual, 91%-95% retention, absence of serious attributable adverse events, and 100% “good-to-excellent” satisfaction). The trial among breast cancer survivors found significantly greater improvements in the intervention vs waitlisted arms in MVPA (114 vs −17 minutes per week), 2-minute step test (122 vs 110 steps), and arm curl test (12.7 vs 10.1 repetitions), 10 whereas the RCT among older survivors found significantly attenuated increases in waist circumference (2.30 cm vs 7.96 cm). 11 Both feasibility studies found trends toward higher vegetable and fruit intake.

These positive findings, plus those from survey studies, 12 , 13 observational studies, 14 and gardening interventions in other populations, 2 , 15 , 16 , 17 , 18 , 19 , 20 , 21 reinforce that adequately powered trials to formally test the benefits of gardening interventions among cancer survivors are needed. Herein, we report the results of the fully powered Harvest for Health RCT that was tested among Medicare-eligible cancer survivors across Alabama.

Methods

Harvest for Health was a 2-arm, single-blinded RCT that used a waitlist-controlled, crossover design to test the effects of a home-based gardening intervention on vegetable and fruit intake, physical activity, physical functioning, quality of life, sleep, adiposity, gut microbiome, and inflammatory biomarkers. Individuals assessing outcomes and conducting analyses were blinded to arm assignment. This study was approved by the University of Alabama at Birmingham (UAB) institutional review board ( NCT02985411 ; the trial protocol is given in Supplement 1 ). All study participants provided written informed consent. We followed the Consolidated Standards of Reporting Trials ( CONSORT ) reporting guideline. The study period spanned May 11, 2016, to May 2, 2022, with detailed methods and recruitment results published previously. 22

Participants

Medicare-eligible cancer survivors who were diagnosed with and had completed curative therapy for cancers with 5-year survival of 60% or more 22 were ascertained through UAB or Alabama state cancer registries and approached via direct mail 23 followed by telephone contact. Referrals responding to media also were enrolled after verification of diagnoses by health care practitioners. Interested survivors were screened for suboptimal vegetable and fruit consumption (<5 servings per day), physical activity (<150 MVPA minutes per week), physical functioning (36-Item Short Form Health Survey [SF-36] subscale score ≤90 [score range, 0-100, with higher scores indicating better physical functioning]), and living independently at a residence that could accommodate 1 raised bed (1.2 × 2.4 m) or 4 grow boxes (62.2 × 52.1 cm) with 6 or more hours per day of sunshine and ready access to water. Other criteria were ability to speak, read, and write English and willingness to participate for 2 years with randomization to either study arm. Exclusion criteria included vegetable gardening within the past 2 years or medical conditions that precluded gardening. All study participants were followed up from initial contact until study completion or termination according to the CONSORT guideline. 24

Assessments

Assessments occurred at baseline (before randomization) and 12 months and included anthropometric measures (height [baseline only], weight, and waist circumference), 25 physical performance testing (Senior Fitness Battery 8 ), accelerometry (7-day collection via GTPX3 accelerometer [ActiGraph]), the Community Healthy Activities Model Program for Seniors (CHAMPS) Physical Activity Questionnaire for Older Adults, 26 blood collection via venipuncture (after ≥4 hours of fasting), and collection of toenail clippings and a fecal wipe. Plasma and serum samples and fecal wipes were stored at −80 °C and toenails at room temperature 27 and then batch-analyzed following published methods for α-carotene, 28 cortisol, 29 fecal microbiome, 30 , 31 , 32 , 33 and inflammatory biomarkers. 22 A remote protocol was implemented in April 2020 due to the COVID-19 pandemic. It omitted grip strength, arm curl, and gait speed measures from the performance battery because of cost or failure to prove reliability, 22 , 34 and dried blood spots (found valid and reliable for interleukin-6 [IL-6]) and tumor necrosis factor–α [TNF-α]) supplanted venipuncture. Assessment of plasma α-carotene levels (an objective measure of vegetable and fruit intake) was not possible. Surveys administered online (REDCap) or by mail captured self-reported race, ethnicity, comorbidities, and symptoms 35 ; falls 36 ; perceived stress 37 ; reassurance of worth 38 ; sleep quality 39 ; vegetable and fruit consumption 40 ; physical activity 26 ; quality of life 41 ; and changes in health. Race and ethnicity were included in the analysis as potential effect modifiers; categories were African American or Black, non-Hispanic White, and other (included American Indian, Asian, Pacific Islander, Hispanic, and multiracial; combined into 1 category because <2% of participants identified as a race other than African American or Black or White and/or as Hispanic ethnicity).

Randomization and Intervention

After conclusion of the baseline assessment, a staff member opened a sealed envelope (created by a statistician) to reveal the randomization assignment. Participants were evenly allocated within each county to receive the intervention immediately or to be waitlisted using computer-generated permuted block randomization. In the intervention group, the EMGs were paired with cancer survivors within counties of residence and visited survivors’ homes monthly to help establish and then maintain a spring, summer, and fall garden. Telephone, email, or text messages were scheduled between visits, with social cognitive theory undergirding all communications. 42 The EMGs logged all encounters and provided time-stamped garden photographs. Each participant received a 1.2 × 2.4-m raised bed kit (or 4 grow boxes), soil, fertilizer, mulch, frost cloths, tomato cages, trellises, gardening hose, watering can, trowel, cultivator, soft chemistry insect control products, seeds, transplants, and a gardening journal. Participants and EMGs received hats, sunscreen, and binders with contact information, interaction tracking logs, and publications on gardening, safety, health, and vegetables and fruits (including recipes). Supplies were distributed at EMG-participant meet-and-greets (informational sessions and social events) at local community centers before initial planting. During the COVID-19 pandemic, meetings were conducted via videoconferencing and supplies were distributed in parking lots; all EMG-participant interactions were restricted to masked, outdoor encounters. Waitlisted participants received the identical intervention after 12 months.

Statistical Analysis

A composite dichotomous score (yes or no) served as the primary outcome based on baseline-to-12-month attainment of the following benchmarks: increased consumption of 1 or more vegetable and fruit servings per day, corroborated by an increase of at least 10% in plasma α-carotene level 43 ; 30 or more minutes of MVPA per week, corroborated by accelerometry; and increase of at least 5 SF-36 physical functioning subscale points, corroborated by improved scores in at least 60% of physical performance tests. Power calculations indicated that 185 participants per arm would yield more than 90% power to detect between-arm differences using χ 2 testing (assumptions: 15% attrition and a moderate response difference ≥20%). 44

Descriptive statistics and normality checks were performed on continuous variables (results from transformed analyses were similar). All statistical tests were 2-sided (α < .05) and conducted using SAS, version 9.4 (SAS Institute Inc); an intent-to-treat approach was used for all analyses.

Continuous outcomes were explored using linear mixed-effects models, including fixed effects for group, time, and group × time interaction, with a random intercept for participants estimated using restricted maximum likelihood. The mean change at 12 months vs baseline within each group and the difference in mean change between groups, with corresponding 95% CIs, were reported.

Categorical secondary outcomes (weight, waist circumference, health-related quality of life, reassurance of worth [Revised Social Provision Scale], self-efficacy to garden, social support to garden, sleep quality, perceived stress, cortisol level, IL-6 level, TNF-α level, and microbiome alpha diversity) were explored using logistic mixed-effects models with the same aforementioned model terms and assumed a binary or multinomial response distribution. Odds ratios (ORs) and their corresponding 95% CIs were reported.

Linear mixed-effects models are robust against missing data, as they can accommodate unbalanced data patterns 45 , 46 ; thus, all available observations and participants were included in the analysis. For each outcome, participants with at least 1 nonmissing outcome measure were included in the analysis.

Logistic regression analyses were performed to determine whether garden type (raised beds or grow boxes) and EMG contact of 90% or more vs less than 90% modified the effects. Other potential effect modifiers (cancer type [breast and gynecologic cancer vs others], time elapsed from diagnosis [<5 years vs ≥5 years], age [<70 years vs ≥70 years], sex, race and ethnicity [non-Hispanic White as the reference group], educational attainment [high school graduate or less vs above], household size [1 vs ≥2 members], annual income [<$60 000 vs ≥$60 000], rural or urban county of residence, BMI [<30 vs ≥30; calculated as weight in kilograms divided by height in meters squared], vitality [<70% vs ≥70% of the SF-36 subscale score], 41 garden start season [spring vs others], and trial completion before vs during the COVID-19 pandemic) were explored similarly.

Results

This trial randomized 381 survivors of a variety of cancers with dates of diagnosis preexisting enrollment by 2 to 43 years and ages spanning 50 to 95 years (mean [SD] age, 69.8 [6.4] years) to Harvest for Health (194 participants) or the waitlist control (187 participants) ( Table 1 ). A total of 263 participants (69.0%) were female, and 118 (31.0%) were male; 67 (17.6%) were African American or Black; 296 (77.7%), non-Hispanic White; 7 (1.8%), other race and ethnicity; and 11 (2.9%), unknown race and ethnicity. Most were married, urban dwelling, and currently retired or unemployed. A total of 218 (57.2%) were college graduates, and 94 (24.7%) reported an annual income less than $30 000. While few survivors were current smokers, 310 (81.4%) had a BMI of 25 or greater, and participants ate a mean (SD) of 2.1 (1.7) servings of fruits and vegetables a day, less than half of the recommended amount. 47 While participants were screened with 1 item to verify suboptimal physical activity, results of accelerometry and the longer CHAMPS Physical Activity Questionnaire at the baseline assessment revealed higher physical activity levels. 26

Details on enrollment and postrandomization events are shown in Figure 1 . Among the older cancer survivors participating in this trial during the COVID-19 pandemic, 14 (7.2%) in the intervention arm and 13 (7.0%) in the waitlisted arm were excluded before receiving their full assigned study condition. These postrandomization exclusions were evenly distributed between arms, with approximately one-quarter of these participants being excluded due to cancer progression (intervention, 6 [3.1%]; control, 3 [1.1%]), another quarter due to death (intervention, 5 [2.6%]; control, 2 [1.1%]), and another quarter due to subsequently divulging that they were gardening at baseline (intervention, 1 [0.5%]; control, 6 [3.2%]). Two participants (1.1%) assigned to the waitlist refused their randomization assignment. Three participants in each arm (intervention, 1.5%; control, 1.6%) withdrew or were lost to follow-up; thus, the intent-to-treat analysis included a sample of 180 in the intervention arm and 174 who were waitlisted. No between-arm differences in dropouts or adverse events were detected (the only attributable adverse event was nonserious minor bruising from a fall occurring while harvesting vegetables).

Approximately one-half of the intervention arm elected to receive raised beds (99 [51.0%]), whereas 95 (49.0%) preferred grow boxes; amounts of all other supplies were identical. Mean (SD) monthly adherence to EMG contact was 83.7% (20.2%).

Figure 2 depicts percentages of survivors in each arm who demonstrated benchmark improvements in primary outcomes; the intent-to-treat analysis did not detect a significant improvement in the composite index of vegetable and fruit intake, physical activity, and physical function (intervention arm vs waitlisted arm, 4.5% vs 3.1%; P = .53). With the exception of self-reported MVPA, greater baseline-to-12-month improvements were observed in the intervention arm compared with the waitlisted arm. These data showed consistent directionality, although no significant between-arm differences were detected. Table 2 provides continuous data on each of these variables. There were several significant within- and between-group differences. For example, a significant improvement in daily servings of vegetables and fruits was detected in the intervention arm (mean change, 0.3 servings; 95% CI, 0.0-0.6 servings; P = .04), whereas no difference was observed among controls (mean change, 0.0 servings; 95% CI, −0.3 to 0.2 servings; P = .76); plasma α-carotene levels paralleled these values. While no significant between-arm differences were detected in vegetable and fruit consumption (mean difference, 0.3 servings per day; 95% CI, −0.1 to 0.7 servings per day; P = .10), between-arm differences were found for both the 2-minute step test (significant decline in waitlisted arm: −5.4 steps [95% CI, −9.1 to −1.7 steps]; mean between-group difference, 6.0 steps [95% CI, 0.8-11.2 steps]; P = .03) and the 30-second chair-stand test (significant improvement in intervention arm: 0.6 repetitions [95% CI, 0.2-1.1 repetitions]; mean between-group difference, 0.8 repetitions [95% CI, 0.1-1.5 repetitions]; P = .02). These results were consistent regardless of EMG contact. However, there was evidence of significant effect moderation. In analyses of the intervention arm alone, participants who were more proximal to diagnosis demonstrated greater gains in physical functioning (odds ratio [OR], 0.37; 95% CI, 0.16-0.86; P = .02) and those reporting higher baseline vitality showed larger improvements in MVPA (OR, 2.21; 95% CI, 1.05-4.66; P = .04). The only variable that significantly moderated effects within the entire sample was the COVID-19 pandemic, for which significantly greater odds of improvements in self-reported physical functioning (difference of 5 points) were observed among participants who completed the study before vs during the pandemic (OR, 2.17; 95% CI, 1.12-4.22; P = .02).

Table 3 provides secondary outcome data. A modest weight loss of 0.8 kg over 12 months was observed with the gardening intervention but not among controls (mean between-arm difference, −1.0 kg; 95% CI, −2.2 to 0.1 kg; P = .08). Significant between-arm differences were noted for perceived health, which improved in the intervention arm but not among controls (8.4 points [95% CI, 3.0-13.9 points] on a 100-point scale, with higher scores indicating better health; P = .003). Other quality-of-life outcomes (eg, physical and emotional role and general health) improved in both arms, with no between-arm differences. Inflammatory markers (cortisol, IL-6, and TNF-α) remained fairly stable over the study period in both groups. Likewise, no significant between-arm differences were found for gardening self-efficacy (a hypothesized intervention mediator); however, significant improvements in social support to garden (another hypothesized mediator) were detected in the intervention arm (mean [SD] days of support per month, 4.5 [6.0] at baseline vs 9.0 [7.7] at 12 months; mean change, 4.4 days [95% CI, 3.3-5.6 days]; P < .001). Between-arm differences also were detected for microbiome alpha diversity (observed species: 84.1 [95% CI, 20.5-147.6] more in the intervention group; P = .01), with significant declines observed among waitlisted survivors across 3 other indices (whole tree phylogeny, Shannon, and Simpson indices).

Discussion

To our knowledge, this study is the largest RCT of a vegetable gardening intervention to date. While the Harvest for Health intervention did not have a significant effect on the composite primary outcome, it resulted in significant differences in individual components of the composite score and other secondary outcomes. For example, while the between-arm difference of 0.3 vegetable and fruit servings consumed per day in the current study falls below the difference of 0.5 to 1.2 servings per day reported in prior gardening interventions, including Harvest for Health pilot studies, 10 , 21 , 48 it falls within the range of 0.1 to 1.4 servings per day reported in a systematic review of 44 intervention studies among adults that were focused exclusively on increasing vegetable and fruit intake. 49 This variation could be due to the instruments used (vegetable and fruit screener vs food frequency questionnaires or dietary recalls) but more likely is attributable to shorter study periods and measures taken prior to planting vs harvest time as well as influences of seasonality, which have proven effects on vegetable and fruit consumption. 50 While large meta-analyses indicated that overall mortality decreased with vegetable and fruit intakes of 5 servings per day, 51 , 52 dose-response analyses suggested that larger health benefits may occur for increases from low to moderate intakes, 52 , 53 as in the current study, in which baseline intakes averaged only 2.1 servings per day. Another community gardening intervention documented increases of 0.46 servings per day but did so in populations with baseline intakes of 4.95 servings per day 48 ; in that study, Litt and colleagues 48 excluded data collected after the COVID-19 pandemic on the premise that they may have differed from prepandemic values. Instead, the current study included all data, but as in the study by Litt et al, 48 the lack of data on a primary outcome (α-carotene) was limiting, and results yielded hard evidence that another primary outcome (ie, physical functioning) was moderated by the COVID-19 pandemic.

Participants who completed the intervention prior to the pandemic had twice the likelihood of demonstrating a clinically meaningful physical functioning difference of 5 points. 54 Changes during the study in factors associated with functional decline in older adults, such as pain, fatigue, or depression, 55 , 56 , 57 , 58 whether or not related to the pandemic, could have affected an individual’s physical functioning, as reported in the Southwest Harvest for Health intervention. 18 Also, physical activity may have contributed to COVID-19 pandemic–related moderation of physical functioning, 58 even though the data suggested that physical activity was not moderated directly and there were no between-arm differences (given high heterogeneity). The absence of a statistically significant effect on MVPA mirrors findings of other studies 17 , 22 , 48 and may relate to the need to assess populations with greater baseline vitality to observe postulated effects that the light-to-moderate activities of gardening serve as a gateway to more intensive exercise (eg, joining a gym); this also may have been affected by the pandemic.

Significant differences in the 2-minute step and 30-second chair-stand performance tests have been observed in other gardening studies, with increases in scores ranging from 4 to 35 steps and 0.6 to 1.4 stands, respectively. 10 , 21 , 22 Our results for these 2 tests represent small, meaningful changes according to the effect size method. 44 , 59 , 60 Importantly, these outcomes serve as significant factors associated with cardiorespiratory fitness, 61 mobility, 62 and frailty. 63 Some physical performance tests associated with the Senior Fitness Battery that were used in prepandemic assessments, such as grip strength and arm curls (measures of upper-arm strength that improved in our group’s pilot studies 7 , 10 ), were not possible during the COVID-19 pandemic and likely impaired the overall composite score.

Despite the pandemic and changes in protocol, perceived social support increased significantly across the study period among intervention arm participants. Therefore, despite restrictions placed on EMG-participant contact, the number of days per month that intervention arm participants reported receiving support doubled compared with unchanged levels among controls. Data were not as strong for self-efficacy, another key construct of social cognitive theory differences that might be further explored in future research. 64

Finally, the intervention was found to have significant effects on key subjective (perceived health) and objective (gut microbiome alpha diversity) health-related outcomes by mitigating declines that typically occur with aging and chronic disease. 65 , 66 Given associations of perceived health with mortality 66 , 67 and of sustained higher levels of alpha diversity with decreasing risk of heart disease, 68 diabetes, 69 sarcopenia, 70 and obesity 71 (conditions that are prevalent among cancer survivors 4 , 5 , 6 , 65 ), these effects are noteworthy.

Strengths and Limitations

This study has strengths. It enrolled a diverse study sample in terms of cancer type, household income, and geography (rural and urban). Moreover, participants were fairly representative of older individuals in Alabama with respect to race and ethnicity (proportion of non-Hispanic White individuals in the study sample, 77.7%; Alabama, 73.9%) and income (Alabama’s median income among older adults is $47 114), though the proportions of female participants and college graduates in the study were higher (ie, 69.0% female [Alabama, 55%] and 57.2% college graduates [Alabama, 26.1%]). 72 The trial also successfully enrolled and retained older cancer survivors up to 95 years of age with multiple comorbidities and functional limitations. This theoretically grounded study also included rigorous measures of health behaviors and outcomes that are important to cancer survivors, clinicians, and researchers alike.

The results should be interpreted in light of potential study limitations, which include bias resulting from a study sample with a higher educational level that was predominantly female, thus limiting generalizability. Moreover, the unblinded nature of the intervention may have influenced various behaviors or reporting. The COVID-19 pandemic also necessitated changes in intervention delivery and assessment and affected the results.

Conclusions

In this RCT of a home-based vegetable gardening intervention among older cancer survivors, a vegetable gardening intervention did not significantly improve a composite index of diet, physical activity, and physical function; however, survivors assigned to the intervention had significantly increased vegetable and fruit consumption and, compared with waitlisted survivors, experienced significant improvements in perceived health, physical performance, and gut microbiome diversity. However, trial outcomes were significantly moderated by the COVID-19 pandemic. Thus, future studies are needed, especially those that use rigorous measures to evaluate the physical, mental, and social benefits of vegetable gardening and that explore potential mechanisms. Continued use and testing of an extant infrastructure for delivering the intervention (eg, a cooperative extension master gardener program) is warranted given the potential for widespread dissemination and sustainability.

References

1 Cicero Letter to Varro . Cic. Fam. 9.4; 46 BC.

2 Wang Z , Zhang Y , Lu S , . Horticultural therapy for general health in the older adults: a systematic review and meta-analysis . PLoS One . 2022 ; 17 ( 2 ): e0263598 . doi: 10.1371/journal.pone.0263598 35143551 PMC8830630

3 Kamphuis CB , Giskes K , de Bruijn GJ , Wendel-Vos W , Brug J , van Lenthe FJ . Environmental determinants of fruit and vegetable consumption among adults: a systematic review . Br J Nutr . 2006 ; 96 ( 4 ): 620 - 635 . 17010219

4 DeSantis CE , Lin CC , Mariotto AB , . Cancer treatment and survivorship statistics, 2014 . CA Cancer J Clin . 2014 ; 64 ( 4 ): 252 - 271 . doi: 10.3322/caac.21235 24890451

5 Sanft T , Day A , Peterson L , . NCCN Guidelines insights: Survivorship, version 1.2022 . J Natl Compr Canc Netw . 2022 ; 20 ( 10 ): 1080 - 1090 . doi: 10.6004/jnccn.2022.0052 36240847

6 Siegel RL , Miller KD , Wagle NS , Jemal A . Cancer statistics, 2023 . CA Cancer J Clin . 2023 ; 73 ( 1 ): 17 - 48 . doi: 10.3322/caac.21763 36633525

7 Blair CK , Madan-Swain A , Locher JL , . Harvest for health gardening intervention feasibility study in cancer survivors . Acta Oncol . 2013 ; 52 ( 6 ): 1110 - 1118 . doi: 10.3109/0284186X.2013.770165 23438359 PMC3718632

8 Rikli RE , Jones CJ . Development and validation of criterion-referenced clinically relevant fitness standards for maintaining physical independence in later years . Gerontologist . 2013 ; 53 ( 2 ): 255 - 267 . doi: 10.1093/geront/gns071 22613940

9 Cases MG , Frugé AD , De Los Santos JF , . 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 . doi: 10.1016/j.cct.2016.08.014 27565830 PMC5055381

10 Bail JR , Frugé AD , Cases MG , . A home-based mentored vegetable gardening intervention demonstrates feasibility and improvements in physical activity and performance among breast cancer survivors . Cancer . 2018 ; 124 ( 16 ): 3427 - 3435 . doi: 10.1002/cncr.31559 29932460 PMC6108896

11 Demark-Wahnefried W , Cases MG , Cantor AB , . 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 ( 4 ): 689 - 704 . doi: 10.1016/j.jand.2017.11.001 29305129 PMC5869079

12 Rogers LQ , Markwell SJ , Courneya KS , McAuley E , Verhulst S . Physical activity type and intensity among rural breast cancer survivors: patterns and associations with fatigue and depressive symptoms . J Cancer Surviv . 2011 ; 5 ( 1 ): 54 - 61 . doi: 10.1007/s11764-010-0160-8 21110134 PMC3041842

13 Johnson BL , Trentham-Dietz A , Koltyn KF , Colbert LH . Physical activity and function in older, long-term colorectal cancer survivors . Cancer Causes Control . 2009 ; 20 ( 5 ): 775 - 784 . doi: 10.1007/s10552-008-9292-9 19123055 PMC2716661

14 Veldheer S , Tuan WJ , Al-Shaar L , . Gardening is associated with better cardiovascular health status among older adults in the United States: analysis of the 2019 Behavioral Risk Factor Surveillance System Survey . J Acad Nutr Diet . 2023 ; 123 ( 5 ): 761 - 769.e3 . doi: 10.1016/j.jand.2022.10.018 36323395 PMC10752423

15 Spees CK , Hill EB , Grainger EM , . Feasibility, preliminary efficacy, and lessons learned from a garden-based lifestyle intervention for cancer survivors . Cancer Control . 2016 ; 23 ( 3 ): 302 - 310 . doi: 10.1177/107327481602300314 27556671 PMC5097863

16 Forbes CC , Swan F , Greenley SL , Lind M , Johnson MJ . Physical activity and nutrition interventions for older adults with cancer: a systematic review . J Cancer Surviv . 2020 ; 14 ( 5 ): 689 - 711 . doi: 10.1007/s11764-020-00883-x 32328828 PMC7473955

17 Blair CK , Adsul P , Guest DD , . Southwest Harvest for Health: an adapted mentored vegetable gardening intervention for cancer survivors . Nutrients . 2021 ; 13 ( 7 ): 2319 . doi: 10.3390/nu13072319 34371829 PMC8308636

18 Sharma H , Pankratz VS , Demark-Wahnefried W , Pestak CR , Blair CK . Association between quality of life and physical functioning in a gardening intervention for cancer survivors . Healthcare (Basel) . 2022 ; 10 ( 8 ): 1421 . doi: 10.3390/healthcare10081421 36011078 PMC9407773

19 Spees CK , Braun AC , Hill EB , . Impact of a tailored nutrition and lifestyle intervention for overweight cancer survivors on dietary patterns, physical activity, quality of life, and cardiometabolic profiles . J Oncol . 2019 ; 2019 : 1503195 . doi: 10.1155/2019/1503195 31871455 PMC6906801

20 Winkels RM , Artrip R , Tupinio M , Veldheer S , Dandekar SC , George DR . Opportunities for growth: evaluating the feasibility of a community gardening intervention pairing adolescent and young adult cancer survivors with experienced gardeners . J Adolesc Young Adult Oncol . 2020 ; 9 ( 1 ): 115 - 119 . doi: 10.1089/jayao.2019.0035 31580754

21 Bail JR , Wilson J , Tipre M , . Harvest for Health in survivors of chronic disease: a demonstration project in the Alabama Black Belt and Mississippi Delta region . Transl Behav Med . 2024 ; 14 ( 2 ): 127 - 137 . doi: 10.1093/tbm/ibad062 37824851 PMC10849179

22 Bail JR , Blair CK , Smith KP , . 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 . J Acad Nutr Diet . 2022 ; 122 ( 9 ): 1629 - 1643 . doi: 10.1016/j.jand.2022.05.005 35533876 PMC10656755

23 Warnock AC , Kaur H , Buckman JR , Hoenemeyer T , Demark-Wahnefried W . A comparison of two mail-based strategies to recruit older cancer survivors into a randomized controlled trial of a lifestyle intervention . J Cancer Surviv . 2022 ; 16 ( 5 ): 998 - 1003 . doi: 10.1007/s11764-021-01091-x 34342814 PMC9489558

24 Moher D , Hopewell S , Schulz KF , . CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials . BMJ . 2010 ; 340 : c869 . doi: 10.1136/bmj.c869 20332511 PMC2844943

25 Lohman TG , Roche AF , Martorell F . Anthropometric Standardization Reference Manual . Human Kinetics ; 1988 .

26 Stewart AL , Mills KM , King AC , Haskell WL , Gillis D , Ritter PL . CHAMPS Physical Activity Questionnaire for Older Adults: outcomes for interventions . Med Sci Sports Exerc . 2001 ; 33 ( 7 ): 1126 - 1141 . doi: 10.1097/00005768-200107000-00010 11445760

27 Frugé AD , Cases MG , Howell CR , . Fingernail and toenail clippings as a non-invasive measure of chronic cortisol levels in adult cancer survivors . Cancer Causes Control . 2018 ; 29 ( 1 ): 185 - 191 . doi: 10.1007/s10552-017-0989-5 29170880 PMC5764195

28 Bieri JG , Brown ED , Smith JC . Determination of individual carotenoids in human plasma by high performance liquid chromatography . J Liq Chromatogr . 1985 ; 8 ( 3 ): 473 - 484 . doi: 10.1080/01483918508067094

29 Warnock F , McElwee K , Seo RJ , . Measuring cortisol and DHEA in fingernails: a pilot study . Neuropsychiatr Dis Treat . 2010 ; 6 : 1 - 7 . 20169040 PMC2951060

30 Caporaso JG , Bittinger K , Bushman FD , DeSantis TZ , Andersen GL , Knight R . PyNAST: a flexible tool for aligning sequences to a template alignment . Bioinformatics . 2010 ; 26 ( 2 ): 266 - 267 . doi: 10.1093/bioinformatics/btp636 19914921 PMC2804299

31 Caporaso JG , Lauber CL , Walters WA , . Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample . Proc Natl Acad Sci U S A . 2011 ; 108 ( suppl 1 ): 4516 - 4522 . doi: 10.1073/pnas.1000080107 20534432 PMC3063599

32 Kumar R , Eipers P , Little RB , . Getting started with microbiome analysis: sample acquisition to bioinformatics. Curr Protoc Hum Genet . 2014 ;82: 18.8.1 - 18.8.29 . 10.1002/0471142905.hg1808s82 PMC4383038 25042718

33 Van Der Pol WJ , Kumar R , Morrow CD , . In silico and experimental evaluation of primer sets for species-level resolution of the vaginal microbiota using 16S ribosomal RNA gene sequencing . J Infect Dis . 2019 ; 219 ( 2 ): 305 - 314 . doi: 10.1093/infdis/jiy508 30535155 PMC6941453

34 Hoenemeyer TW , Cole WW , Oster RA , Pekmezi DW , Pye A , Demark-Wahnefried W ; In-Person Anthropometric and Physical Performance Assessments in Cancer Survivors and Supportive Partners . Test/retest reliability and validity of remote vs in-person anthropometric and physical performance assessments in cancer survivors and supportive partners . Cancers (Basel) . 2022 ; 14 ( 4 ): 1075 . doi: 10.3390/cancers14041075 35205823 PMC8869803

35 Fillenbaum G . Multidimensional Functional Assessment of Older Adults . Lawrence Erlbaum Associates ; 1988 .

36 Chen TY , Janke MC . Predictors of falls among community-dwelling older adults with cancer: results from the health and retirement study . Support Care Cancer . 2014 ; 22 ( 2 ): 479 - 485 . doi: 10.1007/s00520-013-2000-7 24122385

37 Cohen S , Kamarck T , Mermelstein R . A global measure of perceived stress . J Health Soc Behav . 1983 ; 24 ( 4 ): 385 - 396 . doi: 10.2307/2136404 6668417

38 Cutrona CE . The Provisions of Social Relationships and Adaptation to Stress . Vol 1 . JAI Press ; 1987 . Advances in Personal Relationships .

39 Buysse DJ , Reynolds CF III , Monk TH , Berman SR , Kupfer DJ . The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research . Psychiatry Res . 1989 ; 28 ( 2 ): 193 - 213 . doi: 10.1016/0165-1781(89)90047-4 2748771

40 Division of Cancer Control & Population Sciences. Fruit & vegetable screeners in the Eating at America’s Table Study (EATS): overview. National Cancer Institute. Accessed April 5, 2024. https://appliedresearch.cancer.gov/diet/screeners/fruitveg/

41 Ware JEKM . Physical & Mental Health Summary Scales: A Manual for Users . Version 2. QualityMetric, Inc ; 2007 .

42 Bandura A . Human agency in social cognitive theory . Am Psychol . 1989 ; 44 ( 9 ): 1175 - 1184 . doi: 10.1037/0003-066X.44.9.1175 2782727

43 Brevik A , Andersen LF , Karlsen A , Trygg KU , Blomhoff R , Drevon CA . Six carotenoids in plasma used to assess recommended intake of fruits and vegetables in a controlled feeding study . Eur J Clin Nutr . 2004 ; 58 ( 8 ): 1166 - 1173 . doi: 10.1038/sj.ejcn.1601945 15054430

44 Cohen J . Statistical Power Analysis for the Behavioral Sciences . 2nd ed. Hove and London; 1988 .

45 Detry MA , Ma Y . Analyzing repeated measurements using mixed models . JAMA . 2016 ; 315 ( 4 ): 407 - 408 . doi: 10.1001/jama.2015.19394 26813213

46 Diggle PJHP , Liang KY , Zeger SL . Analysis of Longitudinal Data . 2nd ed . Oxford University Press ; 2013 .

47 Clinton SK, Giovannucci EL, Hursting SD. The World Cancer Research Fund/American Institute for Cancer Research Third Expert Report on Diet, Nutrition, Physical Activity, and Cancer: impact and future directions. J Nutr . 2020;150(4):663-671. doi: 10.1093/jn/nxz268 PMC7317613 31758189

48 Litt JS , Alaimo K , Harrall KK , . Effects of a community gardening intervention on diet, physical activity, and anthropometry outcomes in the USA (CAPS): an observer-blind, randomised controlled trial . Lancet Planet Health . 2023 ; 7 ( 1 ): e23 - e32 . doi: 10.1016/S2542-5196(22)00303-5 36608945 PMC9936951

49 Pomerleau J , Lock K , Knai C , McKee M . Interventions designed to increase adult fruit and vegetable intake can be effective: a systematic review of the literature . J Nutr . 2005 ; 135 ( 10 ): 2486 - 2495 . doi: 10.1093/jn/135.10.2486 16177217

50 Alaimo K , Beavers AW , Coringrato E , . Community gardening increases vegetable intake and seasonal eating from baseline to harvest: results from a mixed methods randomized controlled trial . Curr Dev Nutr . 2023 ; 7 ( 5 ): 100077 . doi: 10.1016/j.cdnut.2023.100077 37215644 PMC10196338

51 Leenders M , Sluijs I , Ros MM , . Fruit and vegetable consumption and mortality: European prospective investigation into cancer and nutrition . Am J Epidemiol . 2013 ; 178 ( 4 ): 590 - 602 . doi: 10.1093/aje/kwt006 23599238

52 Wang DD , Li Y , Bhupathiraju SN , . Fruit and vegetable intake and mortality: results from 2 prospective cohort studies of US men and women and a meta-analysis of 26 cohort studies . Circulation . 2021 ; 143 ( 17 ): 1642 - 1654 . doi: 10.1161/CIRCULATIONAHA.120.048996 33641343 PMC8084888

53 Stanaway JD , Afshin A , Ashbaugh C , . Health effects associated with vegetable consumption: a burden of proof study . Nat Med . 2022 ; 28 ( 10 ): 2066 - 2074 . doi: 10.1038/s41591-022-01970-5 36216936 PMC9556321

54 Ware JE Jr , Bayliss MS , Rogers WH , Kosinski M , Tarlov AR . Differences in 4-year health outcomes for elderly and poor, chronically ill patients treated in HMO and fee-for-service systems: results from the Medical Outcomes Study . JAMA . 1996 ; 276 ( 13 ): 1039 - 1047 . doi: 10.1001/jama.1996.03540130037027 8847764

55 Carroll JE , Bower JE , Ganz PA . Cancer-related accelerated ageing and biobehavioural modifiers: a framework for research and clinical care . Nat Rev Clin Oncol . 2022 ; 19 ( 3 ): 173 - 187 . doi: 10.1038/s41571-021-00580-3 34873313 PMC9974153

56 Loh KP , Pandya C , Zittel J , . Associations of sleep disturbance with physical function and cognition in older adults with cancer . Support Care Cancer . 2017 ; 25 ( 10 ): 3161 - 3169 . doi: 10.1007/s00520-017-3724-6 28455547 PMC5660663

57 Muhandiramge J , Orchard SG , Warner ET , van Londen GJ , Zalcberg JR . Functional decline in the cancer patient: a review . Cancers (Basel) . 2022 ; 14 ( 6 ): 1368 . doi: 10.3390/cancers14061368 35326520 PMC8946657

58 Schmitz KH , Cappola AR , Stricker CT , Sweeney C , Norman SA . The intersection of cancer and aging: establishing the need for breast cancer rehabilitation . Cancer Epidemiol Biomarkers Prev . 2007 ; 16 ( 5 ): 866 - 872 . doi: 10.1158/1055-9965.EPI-06-0980 17507607

59 Kwon S , Perera S , Pahor M , . What is a meaningful change in physical performance? findings from a clinical trial in older adults (the LIFE-P study) . J Nutr Health Aging . 2009 ; 13 ( 6 ): 538 - 544 . doi: 10.1007/s12603-009-0104-z 19536422 PMC3100159

60 Perera S , Mody SH , Woodman RC , Studenski SA . Meaningful change and responsiveness in common physical performance measures in older adults . J Am Geriatr Soc . 2006 ; 54 ( 5 ): 743 - 749 . doi: 10.1111/j.1532-5415.2006.00701.x 16696738

61 Ibrahim A , Mat Ludin AF , Shahar S , Hamzah NH , Chin AV , Singh DKA . Association between maximal oxygen consumption and physical performance tests among older adults with cognitive frailty . Exp Gerontol . 2023 ; 184 : 112326 . doi: 10.1016/j.exger.2023.112326 37967590

62 Gordt K , Mikolaizak AS , Taraldsen K , . Creating and validating a shortened version of the community balance and mobility scale for application in people who are 61 to 70 years of age . Phys Ther . 2020 ; 100 ( 1 ): 180 - 191 . doi: 10.1093/ptj/pzz132 31581286

63 Chiu TY , Yu HW . Associations of multicomponent exercise and aspects of physical performance with frailty trajectory in older adults . BMC Geriatr . 2022 ; 22 ( 1 ): 559 . doi: 10.1186/s12877-022-03246-6 35790911 PMC9258123

64 Bandura A . Health promotion by social cognitive means . Health Educ Behav . 2004 ; 31 ( 2 ): 143 - 164 . doi: 10.1177/1090198104263660 15090118

65 Strasser B , Ticinesi A . Intestinal microbiome in normal ageing, frailty and cognition decline . Curr Opin Clin Nutr Metab Care . 2023 ; 26 ( 1 ): 8 - 16 . doi: 10.1097/MCO.0000000000000878 36102345

66 Latham K , Peek CW . Self-rated health and morbidity onset among late midlife US adults . J Gerontol B Psychol Sci Soc Sci . 2013 ; 68 ( 1 ): 107 - 116 . doi: 10.1093/geronb/gbs104 23197340 PMC3605944

67 DeSalvo KB , Bloser N , Reynolds K , He J , Muntner P . Mortality prediction with a single general self-rated health question: a meta-analysis . J Gen Intern Med . 2006 ; 21 ( 3 ): 267 - 275 . doi: 10.1111/j.1525-1497.2005.00291.x 16336622 PMC1828094

68 Simadibrata DM , Auliani S , Widyastuti PA , . The gut microbiota profile in heart failure patients: a systematic review . J Gastrointestin Liver Dis . 2023 ; 32 ( 3 ): 393 - 401 . doi: 10.15403/jgld-4779 37774217

69 Qin J , Li Y , Cai Z , . A metagenome-wide association study of gut microbiota in type 2 diabetes . Nature . 2012 ; 490 ( 7418 ): 55 - 60 . doi: 10.1038/nature11450 23023125

70 Nikkhah A , Ejtahed HS , Ettehad Marvasti F , . The critical role of gut microbiota dysbiosis in skeletal muscle wasting: a systematic review . J Appl Microbiol . 2023 ; 134 ( 1 ): lxac014 . doi: 10.1093/jambio/lxac014 36626794

71 Pinart M , Dötsch A , Schlicht K , . Gut microbiome composition in obese and non-obese persons: a systematic review and meta-analysis . Nutrients . 2021 ; 14 ( 1 ): 12 . doi: 10.3390/nu14010012 35010887 PMC8746372

72 US Census Bureau . Alabama: 2020 Census. Accessed March 14, 2024. https://www.census.gov/library/stories/state-by-state/alabama-population-change-between-census-decade.html

Trial Protocol

Data Sharing Statement