Resistance Training Lowers HbA1c Without Weight Loss

For older adults with type 2 diabetes, the bathroom scale can miss a meaningful treatment response. A 2025 meta-analysis of 19 randomized controlled trials found that resistance exercise training lowered HbA1c by 0.51 percentage points versus control in older adults with established type 2 diabetes (P<0.0001), while total body weight and total fat mass did not change significantly. Feng et al.'s meta-analysis also found increased lean mass and a smaller waist circumference.

That is a useful corrective to the idea that metabolic progress must first appear as weight loss. HbA1c is a blood marker that reflects average glucose exposure over roughly the prior two to three months. A 0.51-point reduction is a biomarker result, not proof that lifting prevents heart attacks, disability, or diabetes complications. Still, it is clinically relevant enough to take seriously—especially in an older population that may be losing muscle while trying, often unsuccessfully, to lose weight.

What the 19 trials found

Feng and colleagues pooled randomized trials of resistance exercise training in older adults with type 2 diabetes. Compared with control conditions, resistance training reduced HbA1c by 0.51 percentage points. It also improved several lipid measures: triglycerides fell by a mean difference of 0.32 (P=0.03), total cholesterol by 7.08 (P=0.005), and LDL cholesterol by 1.91 (P=0.05). The units for those lipid estimates are not specified in the available summary, so they should not be converted or compared with a patient's lab report without checking the full publication.

The review reported a fasting-blood-glucose mean difference of −1.43 (P=0.04), but the published abstract appears to label its units inconsistently. The figure may be mmol/L rather than mg/dL. That distinction changes the clinical meaning substantially, so the number is best treated as a direction-of-effect finding until the full paper clarifies it.

Fasting insulin did not change significantly across the trials, despite the HbA1c improvement. That result limits any tidy mechanistic story. More muscle could improve peripheral glucose disposal, but these trials did not establish that an individual participant's gain in lean mass caused that participant's HbA1c reduction. They show two favorable changes occurring together at the group level.

The muscle-and-scale split matters

Resistance training increased lean mass and reduced waist circumference without significantly changing weight or total fat mass. In practical terms, a person can improve a diabetes marker and gain or preserve muscle without producing the before-and-after scale photo that diet culture has trained everyone to expect.

This does not make BMI useless. BMI is quick, inexpensive, and often useful for population screening. But it cannot distinguish a person with relatively high fat mass and low muscle from someone at the same BMI with more lean tissue. The editorial argument for looking beyond BMI is therefore reasonable: body composition can add information that a height-and-weight ratio cannot provide.

Evidence for that broader clinical framing is less firm than the exercise evidence. A 2026 observational report on sarcopenic obesity argues that diagnosing the combination of excess adiposity and low muscle may add clinically relevant information beyond BMI-based obesity severity. Because this study is observational and lower quality than the randomized-trial reviews, it should generate better questions rather than settle them. No prospective trial cited here shows that prescribing resistance training based on body composition instead of BMI improves long-term outcomes.

Why resistance training belongs in the plan

The appeal of resistance training is not that it owns glucose control. It addresses glucose, muscle, strength, and physical function in the same session. That combination matters when aging and diabetes raise the odds of sarcopenia and loss of independence.

Outside the specific older-diabetes trials, the pattern is consistent. A 2026 meta-analysis in older women with sarcopenia found that resistance training improved muscle mass, strength, and physical function. A 2024 meta-analysis of secondary sarcopenia also found structural and functional benefits across different disease-related causes. In older adults with osteosarcopenic obesity, a particularly complicated mix of bone loss, low muscle, and excess fat, a 2022 review found improvements in body composition and physical function with resistance training.

These findings support muscle-focused exercise in older adults, but they do not prove that every person with type 2 diabetes needs the same program. The primary diabetes meta-analysis does not resolve the best load, frequency, session duration, equipment, or training intensity. It also does not establish whether the HbA1c effect persists beyond the generally short trial periods, typically under 12 months.

Combine modalities when the goal is broader

For many people, the evidence favors a mixed exercise plan rather than a resistance-only identity. A 2025 meta-analysis of adults with type 2 diabetes and overweight or obesity found that concurrent aerobic and resistance training improved body composition, lipid metabolism, and physical function. The population differs from Feng et al.'s older-adult focus, but the result fits a practical approach: resistance work can preserve or build muscle while aerobic activity targets cardiorespiratory fitness and adds its own metabolic effects.

Nutrition can contribute too. A 2023 network meta-analysis in adults near retirement age found that nutrition-plus-exercise interventions improved body composition beyond exercise alone. It did not establish one universal diet prescription, and it should not be read as permission to treat protein powder as a substitute for training. Muscles are annoyingly literal about this.

Resistance training is effective, not exclusive

Aerobic exercise can also improve metabolic outcomes without resistance-driven lean-mass gains. A 2023 review of low-volume high-intensity interval training in type 2 diabetes reported significant metabolic and cardiorespiratory improvements. A broader meta-analysis found high-intensity interval exercise and moderate-intensity continuous training produced comparable improvements in cardiometabolic risk factors across health and disease populations. That 2021 analysis does not make either approach automatically suitable for every older adult, but it rules out claims that muscle gain is the only route to better metabolic markers.

The same caution applies to prediabetes. A 2022 network meta-analysis found combined exercise programs outperformed resistance training alone for glycemic control in people with prediabetes. Prediabetes is not established type 2 diabetes, so it cannot be directly substituted for Feng et al.'s population. It does, however, strengthen the case for combining modes when feasible.

Exercise may also reach beyond glucose and lipids. A 2021 meta-analysis found that exercise training improved heart-rate variability in people with type 2 diabetes, a marker of cardiac autonomic regulation. The evidence there concerns exercise training broadly, predominantly aerobic training, not resistance training specifically. A separate review of resistance exercise and immune cells found changes in circulating immune-cell outcomes relevant to infection and vaccine responses, but it did not establish a clinical benefit specifically for older adults with type 2 diabetes.

A realistic verdict

Resistance training deserves a place in lifestyle treatment for older adults with type 2 diabetes. Across 19 randomized trials, it lowered HbA1c by 0.51 percentage points while weight and fat mass remained statistically unchanged, and it improved lean mass and waist circumference. That is a meaningful answer to the question many patients ask after months of training: "Why is the scale barely moving?"

It is not a replacement for glucose-lowering medication when medication is indicated, and it is not evidence that lifting beats aerobic exercise for glycemia. Semaglutide, for example, can produce substantial weight loss and metabolic improvements in adults with obesity and type 2 diabetes, though weight loss can include lean-mass loss. A 24-week study of semaglutide reported about 9.4% to 9.5% weight loss, depending on formulation. No head-to-head trial here compares resistance training with metformin or GLP-1 drugs in older adults.

The evidence supports a less glamorous but more useful scorecard: follow HbA1c and other clinical markers, track strength and function, and consider waist or body composition when available. Weight may change. It may not. Either way, muscle is doing work the scale cannot see.