Effects of Plant-Based Protein Interventions, with and without an Exercise Component, on Body Composition, Strength and Physical Function in Older Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.

This is a systematic review and meta-analysis of randomised controlled trials (RCTs). The authors pooled data from multiple studies to answer two main questions:

Does plant-based protein supplementation (soy, pea, rice, or mixed plant proteins) improve body composition, muscle strength, or physical function in adults aged 50+?

Does adding resistance exercise to plant-based protein supplementation produce better results than either intervention alone?

The comparators varied across studies but included:

Plant-based protein vs. placebo (usually maltodextrin or isocaloric carbohydrate)

Plant-based protein vs. animal-based protein (whey, casein, or milk)

Plant-based protein + resistance exercise vs. resistance exercise alone

Plant-based protein + resistance exercise vs. placebo + resistance exercise

Outcome measures were grouped into three categories:

**Body composition:** lean body mass (LBM), fat mass, body fat percentage

**Strength:** handgrip strength, leg press 1-repetition maximum (1-RM), knee extension strength

**Physical function:** gait speed (usual and fast), timed up-and-go (TUG), short physical performance battery (SPPB), chair stand test

The meta-analysis included 22 RCTs with a total of 1,381 participants. Individual study sample sizes ranged from 20 to 220 participants.

**Population characteristics:**

Age range: 50 to 85 years (mean age across studies approximately 67 years)

Sex: mixed populations, with some studies enrolling only women and others only men

Health status: generally healthy community-dwelling older adults, though some studies included sarcopenic or pre-frail individuals

Exclusion criteria common across studies: chronic kidney disease, liver disease, cancer, recent surgery, use of hormone replacement therapy or corticosteroids, BMI > 35, allergy to soy or pea protein

Setting: all studies were conducted in outpatient or community settings, with supervised exercise sessions in the exercise arms

**Important note:** The majority of participants were Caucasian, and most studies were conducted in high-income countries (USA, Canada, Australia, Japan, Europe). This limits generalisability to other ethnic and socioeconomic groups.

The authors extracted data from studies that used the following instruments and protocols:

**Body composition:**

Dual-energy X-ray absorptiometry (DXA) — the gold standard for body composition assessment, used in most studies

Bioelectrical impedance analysis (BIA) — used in a minority of studies, less precise than DXA

Computed tomography (CT) — used in one study to measure thigh muscle cross-sectional area

**Strength:**

Handgrip dynamometry (Jamar or similar hydraulic dynamometer) — standardised protocol with participants seated, elbow at 90 degrees, best of three trials

Leg press 1-repetition maximum (1-RM) — measured on plate-loaded or selectorised machines, with incremental loading until failure within 3–5 attempts

Isokinetic knee extension (Biodex or Cybex dynamometer) — measured peak torque at 60–180 degrees per second

**Physical function:**

Gait speed: timed walk over 4 or 6 metres at usual and fast pace, measured in metres per second

Timed up-and-go (TUG): time to rise from a chair, walk 3 metres, turn, and return to seated position

Short physical performance battery (SPPB): composite score (0–12) combining balance tests, gait speed, and chair stand time

Chair stand test: time to complete 5 or 10 repetitions of standing from a seated position without using arms

**Dietary compliance:**

Most studies used daily protein supplement logs or returned supplement containers

Some studies included 3-day food diaries or 24-hour dietary recalls to assess background protein intake

**Study design:** This is a systematic review and meta-analysis of randomised controlled trials. The authors searched five databases (PubMed, Embase, CINAHL, Scopus, and Web of Science) from inception to May 2023. Two reviewers independently screened titles, abstracts, and full texts, extracted data, and assessed risk of bias using the Cochrane Risk of Bias 2 (RoB 2) tool.

**Inclusion criteria:**

Randomised controlled trials (parallel or crossover designs)

Participants aged 50 years or older

Intervention: plant-based protein supplement (soy, pea, rice, hemp, or mixed plant proteins) given for at least 6 weeks

Comparator: placebo, no supplement, or animal-based protein

Outcomes: at least one measure of body composition, strength, or physical function

Published in English

**Exclusion criteria:**

Studies where protein was provided only as part of a meal replacement or total diet intervention

Studies where exercise was the only intervention (no protein comparison)

Observational studies, case reports, reviews

**Statistical approach:**

Random-effects meta-analysis using the DerSimonian and Laird method

Effect sizes expressed as standardised mean differences (SMD) or mean differences (MD) with 95% confidence intervals

Heterogeneity assessed using I² statistic (low: <25%, moderate: 25–50%, high: >50%)

Subgroup analyses by: protein source (soy vs. pea vs. mixed), exercise component (yes vs. no), sex, baseline protein intake, and intervention duration

Sensitivity analyses excluding studies with high risk of bias

**What this design can prove:**

A meta-analysis of RCTs provides the highest level of evidence for causal inference when the individual trials are well-conducted

The random-effects model accounts for between-study variability, making results more generalisable than a fixed-effects model

Subgroup analyses can identify effect modifiers (e.g., whether exercise is necessary for benefit)

**What this design cannot prove:**

Meta-analyses inherit all limitations of the included studies — if individual studies are poorly designed, the pooled result is still unreliable

Publication bias is possible: studies with null or negative results may be unpublished or published in non-English journals

The "apples and oranges" problem: combining studies with different protein doses, durations, exercise protocols, and outcome measures may obscure real effects

Cannot determine optimal dose or timing of protein supplementation — only that some dose within the studied range (typically 20–40 g/day) produced the observed effects

**Major methodological weaknesses flagged by the authors:**

Only 8 of 22 studies had low risk of bias across all domains; 10 had some concerns, and 4 were rated high risk

Blinding was a common issue: protein supplements (especially soy vs. whey) have different tastes and textures, making true blinding difficult

Many studies did not control for background dietary protein intake, which could dilute or amplify the intervention effect

Exercise protocols varied widely in frequency (2–5 days/week), intensity (low to high), and type (resistance only vs. combined resistance and aerobic)

Only 3 studies included long-term follow-up beyond 24 weeks

**Primary outcomes (body composition):**

**Lean body mass (LBM):** Plant-based protein supplementation combined with resistance exercise produced a small but statistically significant increase in LBM compared to placebo + exercise (SMD = 0.18, 95% CI: 0.04 to 0.32, p = 0.01, I² = 12%). This translates to approximately 0.5–0.8 kg additional lean mass gain over 12–24 weeks.

**Plant-based protein alone (no exercise):** No significant effect on LBM compared to placebo (SMD = 0.05, 95% CI: -0.10 to 0.20, p = 0.51, I² = 0%). Protein without exercise did not increase muscle mass.

**Comparison with animal protein:** Plant-based protein was non-inferior to whey or milk protein for LBM when combined with exercise (SMD = -0.03, 95% CI: -0.22 to 0.16, p = 0.76, I² = 0%). Plant and animal proteins produced equivalent gains.

**Fat mass:** No significant change in fat mass with plant-based protein vs. placebo, regardless of exercise (SMD = -0.06, 95% CI: -0.20 to 0.08, p = 0.40, I² = 0%).

**Secondary outcomes (strength):**

**Leg press 1-RM:** Plant-based protein + exercise significantly improved leg press strength compared to placebo + exercise (MD = 5.2 kg, 95% CI: 1.8 to 8.6 kg, p = 0.003, I² = 34%). This represents approximately 5–8% greater improvement.

**Handgrip strength:** No significant difference between plant-based protein + exercise and placebo + exercise (MD = 0.3 kg, 95% CI: -0.8 to 1.4 kg, p = 0.59, I² = 0%). Handgrip strength did not respond to protein supplementation.

**Knee extension strength:** Plant-based protein + exercise showed a trend toward improvement but did not reach statistical significance (SMD = 0.15, 95% CI: -0.02 to 0.32, p = 0.08, I² = 22%).

**Tertiary outcomes (physical function):**

**Gait speed:** No significant improvement with plant-based protein + exercise vs. placebo + exercise (MD = 0.02 m/s, 95% CI: -0.03 to 0.07 m/s, p = 0.44, I² = 0%). The minimal clinically important difference for gait speed is 0.05 m/s, so even if statistically significant, the effect would be borderline meaningful.

**Timed up-and-go:** No significant difference (MD = -0.3 seconds, 95% CI: -0.9 to 0.3 seconds, p = 0.33, I² = 15%).

**Chair stand test:** No significant improvement (SMD = 0.10, 95% CI: -0.08 to 0.28, p = 0.28, I² = 0%).

**Subgroup analyses:**

**Protein source:** Soy and pea protein showed similar effects; rice protein had insufficient data for analysis

**Dose:** Studies using ≥30 g/day of plant protein showed larger effects on LBM than those using <30 g/day (p for interaction = 0.04)

**Duration:** Interventions lasting ≥12 weeks showed greater effects than those lasting 6–11 weeks (p for interaction = 0.03)

**Baseline protein intake:** Participants with baseline protein intake <0.8 g/kg/day showed greater benefit than those with adequate intake (p for interaction = 0.02)

**Sex:** No significant difference between men and women (p for interaction = 0.61)

To put these numbers in plain English:

**Lean body mass gain:** The additional 0.5–0.8 kg of lean mass from adding plant protein to resistance exercise over 12–24 weeks is roughly equivalent to the amount of muscle a 70-year-old might lose in 2–3 years of normal ageing. In other words, it's a meaningful offset of sarcopenia, but not a dramatic transformation. For context, a typical resistance training program alone in older adults produces about 1.0–1.5 kg of lean mass gain over 12 weeks; adding plant protein adds about 30–50% more gain on top of that.

**Leg press strength:** The 5.2 kg improvement in leg press 1-RM is approximately a 5–8% increase from baseline. This is enough to make daily activities like rising from a chair or climbing stairs noticeably easier for a frail older adult, but would be barely perceptible for a healthy, active person.

**Handgrip strength:** The null result here is important. Handgrip strength is a proxy for overall muscle health and is strongly predictive of mortality and disability. The fact that plant protein did not improve it suggests that the benefits are specific to lower-body, large-muscle-group exercises — not a generalised effect on all muscles.

**Physical function:** The lack of improvement in gait speed, TUG, and chair stand is disappointing but not surprising. These tests are relatively crude measures of function, and the participants in these studies were generally healthy and already functioning well at baseline. A ceiling effect likely masked any improvements.

**What the authors acknowledge:**

High heterogeneity in exercise protocols across studies (frequency, intensity, type, supervision)

Incomplete reporting of background dietary intake in many studies

Short intervention durations (most studies were 12–16 weeks; only 3 exceeded 24 weeks)

Lack of long-term follow-up to assess sustainability of gains

Possible publication bias, though Egger's test was non-significant for most outcomes

Inability to blind participants to protein type due to taste/texture differences between plant and animal proteins

**What a critical reader would add:**

**Industry funding:** Several included studies were funded by protein supplement manufacturers (soy protein associations, pea protein companies). While the authors conducted sensitivity analyses excluding these studies and found similar results, the potential for subtle bias in study design or reporting remains.

**Dose-response not established:** The meta-analysis could not determine the optimal dose, timing (pre- vs. post-exercise), or pattern (bolus vs. spread across meals) of plant protein supplementation. The subgroup analysis suggesting ≥30 g/day is better is based on a small number of studies.

**Amino acid profile differences:** Plant proteins are typically lower in leucine (the key amino acid for stimulating muscle protein synthesis) than animal proteins. The fact that plant proteins still produced equivalent gains to whey is surprising and may reflect the higher total protein doses used in plant protein studies (often 30–40 g vs. 20–25 g for whey).

**Absorption kinetics:** Plant proteins are digested more slowly than whey, which may affect the timing of muscle protein synthesis stimulation. Most studies provided supplements immediately post-exercise, which may be suboptimal for plant proteins.

**Gastrointestinal tolerability:** Soy and pea protein can cause bloating, gas, or digestive discomfort in some individuals, particularly older adults with reduced digestive enzyme production. Only 2 studies reported adverse events systematically.

**Population homogeneity:** Nearly all participants were Caucasian, healthy, and community-dwelling. Results may not apply to frail, hospitalised, or institutionalised older adults, or to non-Caucasian populations.

**No data on muscle quality:** Lean mass measured by DXA does not distinguish between muscle tissue and intramuscular fat or connective tissue. Changes in muscle quality (e.g., fibre type, mitochondrial density) may occur without detectable changes in mass.

For someone running their own n=1 experiment to test whether plant-based protein supplementation improves their body composition and strength:

### What to test

**Intervention:** 30–40 g of plant-based protein powder (soy isolate, pea isolate, or a blend) taken immediately after resistance exercise. Soy and pea appear equally effective; rice protein has less evidence. Avoid "complete" plant protein blends that include hemp or chia, as these have not been tested in this context.

**Comparator:** An isocaloric placebo (maltodextrin or rice flour) matched for calories and appearance, taken at the same time. Alternatively, compare plant protein to whey protein to see if you respond differently.

**Exercise component:** This is critical. Based on the meta-analysis, protein alone without resistance exercise will not produce measurable gains. You must be doing progressive resistance training at least 2–