The Role of Protein Intake and its Timing on Body Composition and Muscle Function in Healthy Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.

This was a systematic review and meta-analysis that combined data from 65 separate randomized controlled trials to answer two questions:

1. **Does eating more protein than usual improve body composition (more lean mass, less fat) and muscle function (strength and muscle building) in healthy adults?**

2. **Does the timing of when you eat that extra protein — for example, immediately after exercise versus hours later, or spread evenly across meals versus concentrated in one meal — change the results?**

The intervention was any form of oral protein supplementation, including protein powders (whey, soy, casein, milk protein), high-protein foods, or simply a higher-protein diet. The comparators were either a low-protein diet, no protein supplementation, or a non-protein placebo (like a carbohydrate drink). The outcomes measured were:

**Lean body mass (LBM):** measured by DXA scan or bioelectrical impedance analysis.

**Muscle strength:** specifically handgrip strength (measured with a dynamometer) and leg press strength (measured as the maximum weight a person could lift once, called 1-repetition maximum or 1-RM).

**Muscle synthesis:** measured by fractional synthesis rate (how fast the body builds new muscle protein), but data were too scarce and inconsistent to pool.

The meta-analysis included **2,907 participants** across 65 studies (1,514 men, 1,380 women, 13 with sex not reported). Participants were healthy, free-living adults (not hospitalised or in care homes). They were split into two age groups for analysis:

**Adults:** mean age 18–55 years.

**Older adults:** mean age >55 years.

Some studies included people with obesity, metabolic syndrome, hypertension, or high cholesterol, but anyone with major chronic diseases (cancer, kidney disease, diabetes, COPD, HIV) was excluded. People with frailty or sarcopenia (age-related muscle loss) were included. The studies ranged from 2 weeks to over a year in duration, and many included resistance exercise training as part of the protocol.

**Lean body mass (LBM):** Measured using dual-energy X-ray absorptiometry (DXA) or bioelectrical impedance analysis (BIA). DXA is the gold standard — it uses low-dose X-rays to distinguish bone, fat, and lean tissue. BIA sends a small electrical current through the body and estimates lean mass based on how the current travels. DXA is more accurate; BIA can be thrown off by hydration status.

**Handgrip strength:** Measured with a handgrip dynamometer. The participant squeezes the device as hard as possible, usually with their dominant hand. This is a simple, reliable test of overall upper-body strength and is often used as a proxy for general muscle function.

**Leg press strength:** Measured as the 1-repetition maximum (1-RM) — the heaviest weight a person can push through a full leg press movement exactly once with good form. This is a direct measure of lower-body strength.

**Muscle synthesis:** Measured using stable isotope tracers (e.g., labelled amino acids infused into the blood) to calculate the fractional synthesis rate — the percentage of muscle protein that is newly built per hour. This is a lab-based measure, not something you can do at home.

**Study design:** This is a systematic review and meta-analysis of randomized controlled trials (RCTs). The authors searched four major databases (PubMed, Web of Science, CINAHL, Embase) up to March 2019, plus clinical trial registries and reference lists. Two reviewers independently screened titles, abstracts, and full texts against pre-defined inclusion/exclusion criteria. Data were extracted by two reviewers and disagreements were resolved by consensus. The quality of each study was assessed using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach, which rates evidence as high, moderate, low, or very low based on risk of bias, inconsistency, indirectness, imprecision, and publication bias.

**Why this design matters:** A meta-analysis is the most powerful type of evidence because it pools data from many small studies to get a more precise estimate of the true effect. By only including RCTs — the gold standard for testing cause-and-effect — the authors ensured that the individual studies were designed to minimise bias. Randomisation means participants were randomly assigned to protein or control groups, which balances known and unknown confounders (like baseline fitness, diet, or genetics) between groups. Blinding (where participants and/or researchers don't know who is getting the protein vs. placebo) reduces placebo effects and expectation bias. The authors specifically looked for studies with placebo controls (e.g., a carbohydrate drink that looks and tastes like the protein shake) to isolate the effect of protein itself.

**What this design can and cannot prove:** A well-conducted meta-analysis of RCTs can prove that protein supplementation *causes* changes in lean body mass, because the randomisation and blinding rule out most alternative explanations. However, this meta-analysis cannot prove that timing of protein intake *does not* matter — it can only show that, across the existing studies, no consistent timing effect was detected. The authors note that the timing subgroup analyses were limited by small numbers of studies and inconsistent definitions of "timing" (some studies compared before vs. after exercise; others compared morning vs. evening; others compared evenly distributed vs. skewed distribution). So the null result on timing could be a true finding, or it could be due to insufficient data to detect a small effect.

**Major methodological weaknesses:**

**Heterogeneity:** The 65 studies varied wildly in protein dose (from 10g/day to over 40g/day), type of protein (whey, soy, casein, milk, mixed), duration (2 weeks to >1 year), whether they included exercise, and the baseline protein intake of participants. The authors tried to account for this with subgroup analyses and sensitivity analyses, but it's a major source of noise.

**Publication bias:** Studies with positive results are more likely to be published than those with null results. The authors tested for this and found some evidence of publication bias for the LBM outcome, meaning the true effect might be smaller than reported.

**Inconsistent outcome definitions:** Some studies reported "lean body mass," others "fat-free mass," others "skeletal muscle mass." The authors lumped these together, but they are not identical measures.

**Short durations:** Many studies were only 8–12 weeks long. Muscle gain is slow — even with optimal protein and training, you might only gain 0.5–1 kg of muscle in 3 months. Longer studies might show larger effects.

**Lack of blinding in some studies:** Not all studies used a placebo. In studies where participants knew they were getting protein (e.g., eating more meat vs. their usual diet), placebo effects and behaviour changes (like exercising more because you think you're in the "good" group) could inflate the results.

All results are reported as mean difference (MD) with 95% confidence intervals (CI). If the CI does not cross zero, the result is statistically significant at p < 0.05.

**Primary outcome: Lean body mass (LBM)**

**Adults (18–55 years):** Protein supplementation increased LBM by **0.62 kg** (95% CI: 0.36 to 0.88 kg). This was statistically significant.

**Older adults (>55 years):** Protein supplementation increased LBM by **0.46 kg** (95% CI: 0.23 to 0.70 kg). This was statistically significant.

**Sensitivity analysis:** When the authors removed studies that did not include exercise training, the results remained essentially the same — the effect was not driven by exercise.

**Timing subgroup analysis:** There was **no significant difference** between any timing pattern (e.g., protein before vs. after exercise, protein spread across the day vs. concentrated in one meal). The confidence intervals for all timing comparisons overlapped zero.

**Secondary outcome: Handgrip strength**

**Older adults only (data for younger adults were insufficient):** Protein supplementation did not improve handgrip strength. The mean difference was **0.26 kg** (95% CI: -0.51 to 1.04 kg). This was not statistically significant — the CI crosses zero, meaning the effect could be negative, zero, or slightly positive.

**Secondary outcome: Leg press strength**

**Adults (18–55 years):** Protein supplementation showed a trend toward improvement, but it was not statistically significant. Mean difference: **5.80 kg** (95% CI: -0.33 to 11.93 kg). The CI just barely crosses zero.

**Older adults (>55 years):** No significant effect. Mean difference: **1.97 kg** (95% CI: -2.78 to 6.72 kg). Not significant.

**Secondary outcome: Muscle synthesis**

Data were too scarce and inconsistent to pool. The authors concluded that the evidence on muscle protein synthesis was "inconclusive."

**Lean body mass gain:** The average gain of 0.46–0.62 kg (roughly 1 to 1.4 pounds) of lean mass over the course of a study is modest. To put it in perspective, a typical resistance training beginner might gain 2–4 kg of muscle in their first year of training. The extra protein adds about 15–30% on top of that. For someone not exercising, the gain is smaller — about 0.5 kg over several months.

**Leg press strength (adults):** The 5.8 kg trend toward improvement is about 13 pounds on the leg press. For context, a healthy 30-year-old man might leg press 100–150 kg, so this is a 4–6% improvement. But because the result was not statistically significant, we cannot be confident it's a real effect.

**Handgrip strength:** The 0.26 kg effect is tiny — about half a pound. For comparison, normal age-related decline in handgrip strength is about 0.5–1 kg per year after age 50. So protein supplementation does not meaningfully prevent this decline.

**What the authors acknowledge:**

High heterogeneity between studies (different protein types, doses, durations, exercise protocols).

Limited data for timing subgroup analyses — only a handful of studies directly compared different timing patterns.

Inconsistent reporting of outcomes across studies (some reported LBM, others fat-free mass, others muscle cross-sectional area).

Possible publication bias for LBM — the funnel plot (a statistical test for publication bias) was asymmetrical, suggesting small studies with negative results may be missing.

The GRADE assessment rated the quality of evidence as "moderate" for LBM and "low" for strength outcomes, mainly due to risk of bias and imprecision.

**What a critical reader would note:**

**Industry funding:** Many of the included studies were funded by dairy or supplement companies (e.g., Nestlé, FrieslandCampina, Glanbia). While the authors declare no conflicts of interest, industry-funded studies are more likely to report positive results. The authors did not test whether funding source affected the results.

**Short duration:** The minimum study duration was 2 weeks, which is far too short to see meaningful changes in muscle mass. Most studies were 8–12 weeks. Muscle protein synthesis rates are ~1–2% per day, so measurable changes take months.

**Baseline protein intake:** The authors did not consistently report or control for participants' habitual protein intake. Someone already eating 1.6 g/kg/day (the upper end of recommendations) will get less benefit from extra protein than someone eating 0.8 g/kg/day.

**Exercise confound:** Many studies included resistance training, which itself increases muscle mass and strength. The sensitivity analysis showed the protein effect was independent of exercise, but the magnitude of the effect might differ between exercisers and non-exercisers.

**Body composition measurement:** BIA (used in some studies) is less accurate than DXA and can be influenced by hydration, food intake, and time of day. This adds measurement error.

**Healthy volunteer bias:** People who volunteer for nutrition studies tend to be more health-conscious and have better diets than the general population. The results may not apply to people with poor baseline diets.

For someone running their own n=1 experiment:

**What to test:**

**Intervention:** Add 20–40 grams of extra protein per day, either as a supplement (whey, casein, or plant-based protein powder) or as whole foods (e.g., 150g chicken breast = ~35g protein; 3 eggs = ~18g; 200g Greek yoghurt = ~20g). Aim for a total daily intake of 1.6–2.2 g per kg of body weight (e.g., 112–154 g/day for a 70 kg person).

**Dose:** 20–40g per serving, consumed at any time of day. The meta-analysis found no benefit of timing, so don't stress about the "anabolic window" — just get the total amount in.

**Minimum meaningful duration:**

**At least 8–12 weeks.** Muscle gain is slow. The average effect in the meta-analysis was ~0.5 kg over 8–12 weeks. To see a clear signal above the noise of daily weight fluctuations (which can be ±1 kg due to hydration and glycogen), you need at least 2–3 months. For a more robust test, run it for 16–24 weeks.

**What to measure:**

**Lean body mass:** Use a DXA scan (gold standard) or a bioelectrical impedance scale (less accurate but usable if you measure at the same time of day, under the same conditions — e.g., first thing in the morning, after voiding, before eating or drinking). Do not rely on a regular bathroom scale — it cannot distinguish muscle from fat.

**Strength:** Pick one or two exercises you can progressively overload (e.g., leg press, bench press, or squat). Measure your 1-repetition maximum (1-RM) or your 5-repetition maximum (5-RM) at baseline and at the end. Use the same equipment and technique both times.

**Body fat percentage:** If you have access to DXA or calipers, measure this too. Protein may help you lose fat while preserving muscle, especially if you're also in a calorie deficit.

**Subjective measures:** Track energy levels, recovery from workouts, and hunger. These are not primary outcomes but are useful for personal insight.

**Key confounds to control for:**

**Total calorie intake:** If you add protein without adjusting calories, you'll gain weight (some muscle, some fat). If you want to isolate the effect of protein, keep total calories constant by replacing some carbs or fat with protein.

**Exercise:** If you start a new exercise program at the same time as the protein, you won't know which caused any changes. Either keep your exercise routine constant throughout the experiment, or, if you want to test protein + exercise, do a phased experiment (e.g., 8 weeks of exercise alone, then 8 weeks of exercise + protein).

**Sleep and stress:** Both affect muscle protein synthesis and recovery. Track your sleep quality and subjective stress levels weekly.

**Hydration and measurement timing:** For body composition measurements, always measure at the same time of day, under the same conditions (morning, fasted, after voiding, before exercise). Dehydration can make you look leaner; overhydration can make you look heavier.

**Baseline protein intake:** Measure your usual protein intake for 1–2 weeks before starting. If you're already eating >1.6 g/kg/day, you may see little to no benefit from adding more.

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

**Lean body