Assessing health impacts of home food gardens with Wind River Indian Reservation families: protocol for a randomised controlled trial

The intervention is a home food garden of at least 80 square feet (approximately 7 square meters), with 2 years of support for designing, installing, and maintaining it. The comparator is a delayed-intervention control group that receives the same gardening support after 2 years of data collection. Outcome measures include:

**Primary outcome:** Adult body mass index (BMI)

**Secondary outcomes:** Child BMI, adult hand strength, self-reported physical and mental health (SF-12v2 survey), diabetes control (haemoglobin A1c), food security (USDA 6-question food security survey), waist circumference, blood pressure, blood serum analysis (standard blood chemistry panel, vitamin D, serum beta carotene)

The study aims to recruit approximately 100 Native American families living within the boundaries of the Wind River Indian Reservation (including Riverton, Wyoming). Eligibility criteria include:

At least one family member enrolled in a federally-recognised tribe

At least two adults (or one adult for single-adult families) willing to participate

Interest in starting and sustaining a home food garden

No current garden over 30 square feet

Children aged 5 and older can enrol for data collection

The population context is critical: approximately 70% of adults on the reservation are obese, over 60% of middle-school-aged children are overweight or obese, and type II diabetes rates are at least 11% (double the state-wide rate). The average age of death for Native Americans in Fremont County is just over 53 years.

Data collection occurs twice per year (February and August) for 2 years, with four total sessions per participant. Measurements include:

**BMI:** Calculated from height and weight (primary outcome for adults)

**Waist circumference:** Measured in centimetres

**Blood pressure:** Systolic and diastolic

**Hand strength:** Measured via dynamometer (a grip strength device)

**Blood serum analysis:** Standard blood chemistry panel, haemoglobin A1c (HbA1c), vitamin D, and serum beta carotene

**Food security:** USDA 6-question food security survey (validated scale)

**Physical and mental wellness:** SF-12v2 survey (validated 12-item health survey, produces physical and mental component summary scores)

**Child measures:** BMI, waist circumference, and HbA1c via finger-prick blood testing

**Fasting status:** Participants are encouraged to fast for 12 hours prior to data collection, and fasting status is recorded

**Study design:** Randomised controlled trial (RCT) with two arms: immediate gardening intervention (2 years of support) versus delayed-intervention control (receives same support after 2 years of data collection). This is a protocol paper, meaning the study is described but results are not yet available.

**Randomisation:** Approximately 100 families are randomly allocated 1:1 to intervention or control. Randomisation is at the household level, not the individual level, because gardening is a household activity. This is important because family members within a household cannot be independent—if one person gardens, others likely benefit from the produce.

**Blinding:** The study cannot be blinded. Participants know whether they are gardening, and researchers collecting data likely know as well. This is a major limitation because expectation effects (the placebo effect) cannot be ruled out. If people believe gardening will improve their health, they may report better health or unconsciously change other behaviours.

**Duration:** 2 years of data collection, with four measurement points (February and August each year). The intervention runs for 2 years for the treatment group; the control group receives the intervention after 2 years.

**Statistical approach:** Primary analysis is intention-to-treat (ITT), meaning all families are analysed in the group they were assigned to, regardless of whether they actually gardened. This preserves the benefits of randomisation and gives a real-world estimate of what happens when you offer gardening support to families. The analysis uses a mixed model (analysis of covariance, ANCOVA) to estimate the effect of the intervention on adult BMI, controlling for baseline weight, gender, age, and tribe, and accounting for household clustering (because multiple family members live together and their outcomes are correlated).

**Secondary analyses:** The study will examine:

Treatment effects using a gardening fidelity measure (how much families actually gardened)

Combined adult and child BMI outcomes using LMS Z-scores (a statistical method that standardises BMI across ages and sexes using NHANES reference data)

Possible mechanisms (e.g., changes in vitamin D or beta-carotene levels)

Qualitative methods including photo narratives, focus groups, and interviews (described elsewhere)

**What this design can and cannot prove:**

**Can prove:** Whether offering home gardening support to families who want it causes changes in BMI and secondary health outcomes compared to not offering it, assuming the randomisation worked and the sample size is adequate.

**Cannot prove:** Why any changes occur (mechanisms are explored but not definitively tested), whether gardening works for people who don't want to garden (the study only includes interested families), or whether effects persist beyond 2 years (the control group gets the intervention after 2 years, so long-term comparison is lost).

**Cannot rule out:** Placebo effects, Hawthorne effects (being observed changes behaviour), or contamination (control families might start gardening on their own or receive produce from intervention families in a tight-knit community).

**Major methodological weaknesses:**

No blinding whatsoever

Small sample size (100 families, though individual participants may number several hundred)

Contamination risk in a small community

Self-report measures (food security, mental health) are subject to bias

BMI as primary outcome may not capture meaningful health changes in 2 years

The protocol does not specify how missing data will be handled

This is a protocol paper, so no results are reported. However, the paper describes what the study aims to find and the rationale for the design. Key points from the background and methods:

This is the first full-scale RCT of home food gardening health impacts identified by the authors

Observational studies suggest gardening improves food security, fruit and vegetable intake, physical activity, and reduces stress

A pilot RCT with older cancer survivors found a trajectory toward positive outcomes

An RCT of community gardens (not home gardens) is currently underway

The study is community-based participatory research, co-designed with Wind River Indian Reservation partners

No results are available because this is a protocol paper. The study is designed to detect changes in adult BMI over 2 years. For context, a meaningful BMI change would be approximately 1–2 kg/m² (roughly 5–15 pounds for an average-height adult), which would represent a clinically significant weight change. The study also aims to detect changes in HbA1c (diabetes control), hand strength, and self-reported health.

**Acknowledged by authors:**

Not all families will wish to garden, limiting the reach of the intervention

The study cannot be blinded

Intervention benefits may "contaminate" control families (e.g., sharing produce, gardening knowledge spreading)

The protocol is described; results are not yet available

**Additional critical limitations:**

**No blinding:** Participants and researchers know group assignment, introducing expectation bias

**Small sample size:** 100 families may not provide enough statistical power to detect modest but meaningful health changes, especially for subgroup analyses (e.g., diabetes control among diabetics)

**Short duration:** 2 years may be insufficient to see changes in BMI or diabetes outcomes, which often require longer-term behaviour change

**Self-report measures:** Food security and mental health are self-reported and subject to social desirability bias

**BMI limitations:** BMI does not distinguish fat from muscle, and changes in body composition may occur without BMI change

**Generalisability:** Results may not apply to non-Native populations, urban settings, or people who are not interested in gardening

**Contamination risk:** In a tight-knit reservation community, control families may start gardening or receive produce from intervention families, diluting the treatment effect

**No washout period:** The delayed-intervention control design means long-term comparison is impossible

**Multiple comparisons:** Many secondary outcomes are tested, increasing the risk of false positives

**Fasting compliance:** Participants are encouraged but not required to fast, which could affect blood measures

**Attrition:** 2 years is a long commitment; families may drop out, especially in the control group

For someone running their own n=1 experiment:

**What to test:**

Start a home food garden of at least 80 square feet (7 square meters) – roughly 8 feet by 10 feet, or a few large raised beds

Grow vegetables, fruits, and herbs that you actually eat

Aim for at least one growing season of consistent gardening (the study uses 2 years, but you can start with one season)

**Minimum meaningful duration:**

At least one full growing season (4–6 months) to see any changes

Two growing seasons (12–18 months) would be more convincing, especially for weight or diabetes outcomes

Measure before planting (early spring) and at peak harvest (late summer/early fall)

**What to measure (specific metrics):**

**Body weight and BMI:** Weigh yourself weekly at the same time of day, in the same clothing

**Waist circumference:** Measure at the narrowest point (or at navel level) monthly

**Hand grip strength:** Use a dynamometer if available; otherwise, track how many push-ups or pull-ups you can do

**Fruit and vegetable intake:** Keep a simple daily log of servings (aim for 5–9 servings per day)

**Food security:** Note whether you ever run out of food or worry about running out

**Mental health:** Use a free validated scale like the PHQ-9 (depression) or GAD-7 (anxiety) weekly

**Blood sugar:** If diabetic or pre-diabetic, check fasting blood glucose weekly

**Physical activity:** Track minutes of gardening and other exercise daily

**Key confounds to control for:**

**Seasonal effects:** Measure at the same time of year (e.g., compare August to August, not August to February)

**Diet changes outside the garden:** Keep a food diary to see if you're eating more or less of other foods

**Exercise changes:** Gardening is physical activity; track whether you're doing more or less other exercise

**Stress and sleep:** Major life events, sleep changes, and stress affect weight and health

**Medication changes:** Note any changes in dosage or type of medications

**Weather:** A bad growing season (drought, flood, pests) will affect garden yield

**Social support:** If family members help garden, you may eat more vegetables; if they don't, you may eat fewer

**What a positive result would look like:**

Weight loss of 2–5 pounds (1–2 kg) over one growing season, or 5–10 pounds over two seasons

Waist circumference decrease of 1–2 inches (2.5–5 cm)

Hand grip strength increase of 2–5 kg (if you're doing the physical work of gardening)

Fruit and vegetable intake increase of 2–4 servings per day

Improved food security (less worry about running out of food)

Improved mental health scores (lower PHQ-9 or GAD-7 scores)

Lower fasting blood glucose (if diabetic or pre-diabetic)

Better sleep quality and more physical activity

**Important caveat for self-experimenters:** This study is designed for families and includes social, cultural, and community components that may be as important as the gardening itself. If you garden alone, you may not see the same benefits. Also, the study population has very high baseline rates of obesity and diabetes, so results may not generalise to healthier populations. Start small, measure carefully, and be patient – meaningful health changes take time.