Exercise and Type 2 Diabetes: What the Evidence Shows

Almost 4.6 million people in the UK are living with a diabetes diagnosis, and around 90% of those cases are type 2 (Diabetes UK, 2024). A further 1.3 million are estimated to have type 2 diabetes without yet being diagnosed. Behind most of those diagnoses, at some point, a GP has said some version of the same thing: “You should exercise more.”

It’s good advice. But it’s incomplete. What kind of exercise? How much? Is it safe if you’re on medication? What difference will it actually make to your blood sugar levels? These are the questions that matter, and for most people, they go unanswered.

This guide sets out what the clinical evidence shows about exercise and type 2 diabetes, how much is needed, which types of activity the research supports, and what working with a clinical exercise physiologist involves in practice, drawing on peer-reviewed evidence and the clinical framework used at Vitruvian Exercise Physiology in London. For condition-specific detail, see our exercise physiology for diabetes service page.

TL;DR: Exercise reduces HbA1c by 0.5–0.78% in people with type 2 diabetes - comparable to some oral medications (Frontiers in Endocrinology, 2026). Guidelines recommend 150 minutes of moderate aerobic activity plus 2–3 resistance sessions per week. A clinical exercise physiologist can design a programme tailored to your condition, medications, and complications.

How Does Exercise Affect Type 2 Diabetes?

Structured exercise programmes can lower HbA1c - the key marker of long-term blood glucose control - by 0.5–0.78% in people with type 2 diabetes (Frontiers in Endocrinology, 2026). To put that in context, some commonly prescribed oral diabetes medications produce a similar reduction. Exercise isn’t a substitute for medication, but it is a clinical intervention with measurable, dose-dependent effects on glycaemic control.

The mechanism is well understood. When muscles contract during physical activity, they take up glucose from the bloodstream, even without insulin. This acute effect is one reason blood glucose often drops during or shortly after exercise. Over time, regular physical activity also improves the body’s sensitivity to insulin, meaning the insulin that is produced works more effectively. For someone with type 2 diabetes, where insulin resistance is the core problem, that’s a meaningful physiological shift.

Beyond blood glucose, the broader metabolic picture matters too. Type 2 diabetes significantly increases cardiovascular risk. Exercise has been shown to reduce blood pressure, improve cholesterol profiles, and lower systemic inflammation, all of which contribute to long-term cardiovascular outcomes (ACSM, 2022). For many people with type 2 diabetes, the cardiovascular benefits of regular exercise may be as clinically important as the glycaemic ones.

Clinical note: The HbA1c reduction from exercise follows a dose-response curve - it’s not a binary “exercise helps or it doesn’t.” The more structured and consistent the programme, the greater the measurable effect. This is one of the key reasons why a clinical approach to exercise differs from generic advice to “be more active.” The dose matters, just as it does with medication.

A 2026 review in Frontiers in Endocrinology found that structured exercise programmes reduce HbA1c by 0.5–0.78% in people with type 2 diabetes, an effect comparable to some oral diabetes medications. The improvement follows a dose-response relationship, with greater reductions seen in people with more poorly controlled blood glucose at baseline.

How Much Exercise Do You Need with Type 2 Diabetes?

Current clinical guidelines recommend at least 150 minutes of moderate-intensity aerobic activity per week, plus 2–3 resistance training sessions, with no more than two consecutive days without exercise (IDF Global Practice Recommendations, 2025). This is the consensus position across the International Diabetes Federation, the American Diabetes Association, and the American College of Sports Medicine. It’s also broadly consistent with what the NHS recommends for the general adult population - but with diabetes-specific adaptations that matter.

What’s less widely known is that researchers have identified a more precise optimal dose. A 2024 meta-analysis in Diabetes Care, analysing data from 126 studies and 6,718 participants, found that the greatest glycaemic benefits occurred at approximately 1,100 MET-minutes per week, with a significant improvement threshold at 840 MET-minutes per week (Diabetes Care, 2024). Below that threshold, the metabolic gains diminish substantially.

What does that look like in practice? A MET-minute is a unit of measurement that accounts for both duration and intensity. Walking briskly for 30 minutes five days a week is roughly 500 MET-minutes. Adding two 30-minute resistance training sessions brings the total closer to 800. To reach 1,100, you’d need to add some higher-intensity activity, or simply do more of what’s already working. A clinical exercise physiologist can help work out what makes sense for each individual, given their current capacity and any complications.

The dose-response data are striking because they vary by baseline glycaemic control. A person with severely uncontrolled diabetes (HbA1c above 9%) can expect a reduction of −0.66% to −1.02% at the optimal dose. Someone with controlled diabetes (HbA1c 7–8%) might see a decrease of −0.40% to −0.47%. Even people with prediabetes benefit, though the effect is smaller at −0.24% to −0.38%.

Baseline control	HbA1c reduction at optimal dose
Severely uncontrolled (HbA1c >9%)	−0.66% to −1.02%
Uncontrolled (HbA1c 8–9%)	−0.49% to −0.64%
Controlled (HbA1c 7–8%)	−0.40% to −0.47%
Prediabetes	−0.24% to −0.38%

Source: Frontiers in Endocrinology, 2026; Diabetes Care, 2024

A 2024 meta-analysis in Diabetes Care found that the optimal physical activity dose for glycaemic benefit in type 2 diabetes is approximately 1,100 MET-minutes per week, with a significant improvement threshold at 840 MET-minutes per week. Below that threshold, metabolic gains diminish substantially, suggesting exercise for diabetes management has a clinically meaningful minimum effective dose.

What Types of Exercise Are Most Effective?

A 2025 network meta-analysis - one of the largest to date, pooling 158 studies and 17,059 participants - found that HIIT produced the greatest HbA1c reduction at −0.61%, with combined aerobic and resistance training close behind at −0.58%, and concluded that combined training delivers the most comprehensive metabolic benefits across blood glucose, cardiovascular markers, and body composition (ScienceDirect, 2025). A separate 2026 review found HIIT reductions as high as −0.78% in some populations, and combined training in the range of −0.48% to −0.74% (Frontiers in Endocrinology, 2026).

The distinction between exercise types matters clinically. They don’t all do the same thing.

Aerobic activity - walking, cycling, swimming - primarily improves cardiovascular fitness, lowers blood glucose during and after sessions, and reduces blood pressure. It’s the most widely recommended form of exercise for diabetes, and the most accessible. A 10-minute walk after meals, for example, has been shown to measurably improve post-meal blood glucose levels.

Resistance training - working against external load - produces a different set of benefits. A 2024 meta-analysis found that resistance training alone reduces HbA1c by 0.50% (95% CI: −0.67 to −0.34%) and produces 10–15% improvements in strength, bone mineral density, and insulin sensitivity (Wan & Su, 2024; ACSM, 2022). These adaptations are particularly relevant for people with type 2 diabetes, who are at increased risk of sarcopenia (age-related muscle loss) and osteoporosis.

Combined training — aerobic and resistance together — is increasingly regarded as the gold standard. HbA1c reductions of −0.48% to −0.74% have been reported, alongside improvements that neither modality achieves alone (Frontiers in Endocrinology, 2026).

HIIT is time-efficient and produces the largest HbA1c reduction in head-to-head comparisons. But it isn’t appropriate for everyone, particularly those with cardiovascular complications or very low baseline fitness. Whether HIIT is a good fit is a clinical decision, not a general recommendation.

Exercise type	HbA1c reduction	Source
HIIT	−0.61% to −0.78%	ScienceDirect, 2025; Frontiers in Endocrinology, 2026
Combined (aerobic + resistance)	−0.48% to −0.74%	ScienceDirect, 2025; Frontiers in Endocrinology, 2026
Resistance training	−0.50%	Wan & Su, 2024
Aerobic training	0.5–0.7%	ACSM, 2022

A 2025 network meta-analysis of 158 studies and 17,059 participants found HIIT produced the greatest HbA1c reduction at −0.61%, with combined training at −0.58% and the most comprehensive overall metabolic benefits (ScienceDirect, 2025). A 2026 review reported HIIT reductions up to −0.78% in some populations. Resistance training alone reduced HbA1c by 0.50% and improved strength, bone density, and insulin sensitivity by 10–15%.

Can Exercise Put Type 2 Diabetes into Remission?

Diabetes UK confirms that exercise, combined with dietary changes, can help some people with type 2 diabetes achieve remission, defined as HbA1c below 48 mmol/mol (6.5%) without diabetes medication for at least three months. That’s a meaningful clinical outcome. It’s also not guaranteed for everyone, and it’s worth being realistic about what it involves.

The strongest UK evidence comes from the DiRECT trial, which demonstrated that intensive weight management - combining calorie reduction with supported physical activity - achieved remission in a significant proportion of participants. Weight loss was the primary mediator: the more weight lost, the more likely remission became. Exercise alone, without dietary change, has not been shown to produce remission at the same rate. But exercise plays a critical role in sustaining the metabolic improvements that make remission possible, and in maintaining them over time.

For many people with type 2 diabetes, the more relevant question isn’t “Can I come off medication?” - it’s “Can I get my blood glucose under better control, reduce my cardiovascular risk, and feel better day-to-day?” The answer to all of those is well-supported by the evidence, regardless of whether remission is the outcome.

What Should You Consider Before Starting?

Exercise is safe for most people with type 2 diabetes, but certain complications require adaptations that should be considered before beginning or changing an exercise programme (ACSM, 2022). The updated NICE NG28 guideline (February 2026) reinforces the recommendation that lifestyle advice, including physical activity, should be provided at all stages of type 2 diabetes management (NICE, 2026).

Some practical considerations are worth flagging.

Blood glucose monitoring. Exercise typically lowers blood glucose, which is the desired effect. But for people taking insulin or sulfonylureas, this creates a hypoglycaemia risk that needs to be managed. Checking blood glucose before and after exercise, and understanding what the numbers mean in the context of activity, is an important part of exercising safely.

Peripheral neuropathy. Reduced sensation in the feet - common in longer-standing diabetes - means foot injuries can occur without being noticed. Appropriate footwear and regular foot checks matter more when physical activity increases.

Retinopathy. Certain types of high-intensity exercise may be inadvisable for people with active proliferative retinopathy. This is a conversation for your clinical team - not something to assess on your own.

Cardiovascular screening. Type 2 diabetes significantly increases cardiovascular risk. For people who have been sedentary for a long time, or who have known cardiovascular disease, a baseline assessment before starting a structured programme is clinically prudent.

None of these are reasons to avoid exercise. They're simply reasons to approach it within a clinical framework rather than without one. A clinical exercise physiologist is trained to screen for these factors and adapt a programme accordingly, which is the point of clinical supervision in the first place.

How Does a Clinical Exercise Physiologist Help?

A clinical exercise physiologist is a healthcare professional trained to assess, prescribe, and deliver evidence-based exercise for people living with chronic conditions, including type 2 diabetes. As of 2024, there are 94 formally registered clinical exercise physiologists in the UK, with projections reaching approximately 1,678 by 2029 (PMC11191777, 2024). It’s a small but rapidly growing profession - and one that most people with type 2 diabetes haven’t been told about.

What does a clinical exercise physiologist actually do that’s different? The starting point is an individual assessment: functional capacity, current medications (and their implications for exercise), any existing complications, physical activity history, and personal goals. From that assessment, a structured programme is designed - not a generic plan, but one built for this person’s clinical picture.

For someone with type 2 diabetes, that might involve accounting for hypoglycaemia risk if they’re on insulin, adapting for peripheral neuropathy, or designing a combined aerobic and resistance programme that fits realistically into their week. The programme is then monitored, progressed, and adapted as the individual’s condition changes.

This is fundamentally different from a gym instructor giving general fitness advice, or a GP saying “try to walk more.” It’s also different from physiotherapy, which typically focuses on rehabilitation from a specific injury or event. For a fuller explanation of the distinction, see our guide on how a clinical exercise physiologist differs from a physiotherapist.

At Vitruvian Exercise Physiology, clinical exercise physiology for type 2 diabetes is delivered by an ESSA-accredited practitioner (Exercise and Sports Science Australia) - the internationally recognised professional body for the discipline. Sessions are available in London (W8), at home, or remotely. For a broader overview of how clinical exercise physiology works across different conditions, see our complete guide to clinical exercise physiology for chronic disease.

Clinical exercise physiology is a regulated UK healthcare profession with 94 registered practitioners as of 2024, projected to reach approximately 1,678 by 2029 (PMC11191777, 2024). A clinical exercise physiologist assesses, prescribes, and delivers evidence-based exercise for people with chronic conditions - accounting for medications, complications, and individual capacity in a way that generic fitness advice does not.

Frequently Asked Questions

Can exercise replace medication for type 2 diabetes?

Exercise is not a replacement for medication, but it works alongside it to improve glycaemic control. Structured exercise can reduce HbA1c by 0.5–0.78% — comparable to some oral medications (Frontiers in Endocrinology, 2026). Some patients may be able to reduce medication under medical supervision. Always consult your GP before making changes.

Is it safe to exercise with high blood sugar?

Generally, moderate exercise helps lower blood glucose. However, if blood glucose is above 16.7 mmol/L (300 mg/dL) with ketones present, exercise should be postponed (ACSM, 2022). A clinical exercise physiologist can advise on safe thresholds based on your individual situation and medication.

How quickly will exercise improve my blood sugar levels?

Acute effects can be seen within a single session — blood glucose often drops during and after exercise. Long-term HbA1c improvements typically appear after 8–12 weeks of consistent structured exercise (ACSM, 2022). Consistency matters more than intensity in the early stages.

What if I haven’t exercised in years?

A clinical exercise physiologist will assess your current capacity and design a programme that starts at your level. Many patients begin with short walks and progress gradually. The evidence supports starting gently and building up — even 10-minute walks after meals have been shown to improve post-meal blood glucose.

Does walking count as exercise for diabetes?

Yes. Walking is one of the most accessible and effective forms of exercise for people with type 2 diabetes. Diabetes UK recommends it as a starting point, and research confirms that regular walking reduces blood glucose and improves cardiovascular health. The key is consistency and gradual progression.

Key Takeaways

• Exercise reduces HbA1c by 0.5–0.78% in people with type 2 diabetes, an effect comparable to that of some oral medications.

• The evidence-based target is 150 minutes of moderate aerobic activity plus 2–3 resistance training sessions per week, with an optimal dose of approximately 1,100 MET-minutes per week.

• Combined aerobic and resistance training produces the most comprehensive metabolic benefits. HIIT achieves the greatest HbA1c reduction, but it isn’t appropriate for everyone.

• Exercise, combined with dietary change, can help some people achieve remission, but this is best pursued within a clinical framework.

• Certain diabetes-related complications require specific adaptations to exercise programming. A clinical exercise physiologist is trained to assess for these and design a programme accordingly.

• Clinical exercise physiology provides the structured, individualised bridge between “exercise more” and a plan that accounts for your condition, medications, and goals.

If you’re living with type 2 diabetes and want to understand how structured exercise could help, contact Vitruvian Exercise Physiology to discuss your situation with a clinical exercise physiologist.

Matt Butterworth is a clinical exercise physiologist and founder of Vitruvian Exercise Physiology. He holds ESSA accreditation (Exercise and Sports Science Australia) and works with people living with metabolic conditions, neurological conditions, cancer, and musculoskeletal presentations at the practice’s London (W8) base and via home visits and remote appointments. Vitruvian Exercise Physiology has been in practice since 2018.

Sources: Frontiers in Endocrinology (2026) · Diabetes Care (2024) · Diabetes UK (2023) · GOV.UK Diabetes Profile (2025) · NICE NG28 (2026) · IDF Global Practice Recommendations (2025) · PMC11191777 (2024) · Wan & Su meta-analysis (2024) · Network meta-analysis of 158 studies (2025) · ACSM Consensus Statement (2022)