Exercise for knee and hip osteoarthritis

A 63-year-old man comes to the consultation with right knee pain for two years. The pain worsens when climbing slopes and improves with rest. He brings an X-ray that shows reduced joint space and asks whether it is time for surgery. He takes an anti-inflammatory almost every day and stopped walking because he is afraid of "wearing out" the knee.

This scene sums up the three most common mistakes in osteoarthritis: treating the image instead of the patient, basing care on pills and seeing movement as the enemy of the joint. The guidelines say the opposite. The American College of Rheumatology guideline strongly recommends exercise for all people with knee and hip osteoarthritis (Kolasinski, 2020).

This text sums up what reviews and trials show, compares exercise with painkillers and surgery and proposes a practical script for prescribing.

Key points

• Exercise reduced knee osteoarthritis pain by 12 points on a 0 to 100 scale, right after treatment, in a Cochrane review of 54 studies (Fransen, 2015).
• In hip osteoarthritis, exercise reduced pain by 8 points on the same scale, with a number needed to treat of 6 (Fransen, 2014).
• In a network meta-analysis of 152 trials, exercise relieved pain as much as oral anti-inflammatories and paracetamol (Weng, 2023).
• Arthroscopy for degenerative knee brought a gain of 2.4 mm on a 100 mm scale, which disappeared between one and two years (Thorlund, 2015).
• The effect of exercise peaks near 2 months and fades after 9 months if the practice does not continue (Goh, 2019).
• Weight loss added to exercise reduced pain more than exercise alone in adults with overweight (Messier, 2013).

Summary

• Why osteoarthritis matters
• What osteoarthritis is and where exercise fits in
• What the reviews show for the knee
• What the reviews show for the hip
• Exercise, painkillers and surgery
• Weight, load and the IDEA trial
• How to prescribe in practice
• How NEXO helps train this reasoning
• Frequently asked questions
• Take-home messages

Why osteoarthritis matters

Osteoarthritis is the most common form of arthritis in adults and one of the major causes of chronic pain and loss of mobility. The Global Burden of Disease study estimated 595 million people with osteoarthritis worldwide in 2020 (GBD 2021 Osteoarthritis Collaborators, 2023).

A growing problem

The projection from the same group forecasts 642 million people with knee osteoarthritis in 2050 (GBD 2021 Osteoarthritis Collaborators, 2023). For the hip, the projection is 62.6 million people in the same year (GBD 2021 Osteoarthritis Collaborators, 2023).

Population aging explains part of this growth. Excess weight explains another part: high body mass index accounted for about one fifth of the osteoarthritis burden in the Global Burden of Disease analysis (GBD 2021 Osteoarthritis Collaborators, 2023).

Why exercise is central

Osteoarthritis affects the whole joint: cartilage, bone, synovium, ligaments and muscles. Pain leads to inactivity, inactivity weakens the quadriceps and the hip muscles, and weakness increases pain when bearing weight. Exercise interrupts this cycle.

That is why the ACR guideline places exercise, weight loss when indicated, self-care and education at the center of treatment (Kolasinski, 2020). Drugs, injections and surgery come in as a complement or for those who do not respond.

What osteoarthritis is and where exercise fits in

The diagnosis of knee or hip osteoarthritis is clinical in most middle-aged or older adults with mechanical pain and short morning stiffness. The X-ray helps little in deciding the initial treatment, because pain intensity and radiographic grade often do not go together.

Imaging does not define treatment

Ordering a routine MRI for knee pain in the older adult often reveals findings that also exist in people without pain, such as degenerative meniscus tears. The finding becomes a label, and the label pulls procedures. It is the same mechanism described in the text on the drivers of overdiagnosis, and it recalls the reasoning about what an incidental finding on imaging is.

What the guideline strongly recommends

The 2019 ACR guideline strongly recommends exercise, weight loss in people with overweight or obesity, and tai chi for knee and hip osteoarthritis (Kolasinski, 2020). Exercise can include walking, strengthening, neuromuscular training and water exercise, with no hierarchy among them (Kolasinski, 2020).

What the guideline strongly advises against

The same guideline strongly recommends against glucosamine, chondroitin, bisphosphonates, hydroxychloroquine and methotrexate in knee and hip osteoarthritis (Kolasinski, 2020). Transcutaneous electrical stimulation also received a strong recommendation against (Kolasinski, 2020). This list helps in conversations with patients who already spend money on supplements with no benefit.

What the knee reviews show

The Cochrane review by Fransen and colleagues gathered 54 land-based exercise studies for knee osteoarthritis (Fransen, 2015). It is the most cited synthesis on the topic and the basis of many guidelines.

Pain, function and quality of life

In 44 trials with 3,537 participants, exercise reduced pain with a standardized mean difference of 0.49 right after treatment (Fransen, 2015). This is equivalent to a 12-point reduction on a 0 to 100 scale, starting from 44 points in the control group (Fransen, 2015).

Physical function improved by the equivalent of 10 points on the same scale, with moderate-quality evidence (Fransen, 2015). Quality of life improved by 4 points, with high-quality evidence (Fransen, 2015).

The effect lasts for a few months

Twelve studies with 1,468 participants measured pain two to six months after the end of the supervised program (Fransen, 2015). The reduction persisted, equivalent to 6 points on the 0 to 100 scale (Fransen, 2015).

The authors describe the effect as moderate right after treatment and small later, but comparable to estimates for nonsteroidal anti-inflammatory drugs (Fransen, 2015).

Safety

Eight studies reported adverse events, all linked to increased knee or low back pain attributed to exercise (Fransen, 2015). No study reported a serious adverse event (Fransen, 2015).

The dropout rate was similar between exercise and control, with an odds ratio of 0.93 (Fransen, 2015). This finding dismantles the idea that patients with osteoarthritis cannot tolerate exercise.

Which modality to choose

A 2025 network meta-analysis gathered 217 trials with 15,684 participants with knee osteoarthritis (Yan, 2025). Aerobic exercise had the largest short-term effect on pain, with a standardized mean difference of 1.10 compared with control (Yan, 2025).

Strengthening and mixed exercise improved function at 12 weeks, with standardized mean differences of 0.86 and 1.07 (Yan, 2025). Mind and body exercises, such as tai chi and yoga, improved function in the short term (Yan, 2025).

The practical message is simple: several modalities work, and the best one is the one the patient does regularly. The ACR guideline itself avoids setting a hierarchy among them (Kolasinski, 2020).

What the hip reviews show

The evidence in the hip is smaller, but it points the same way. The Cochrane review of exercise in hip osteoarthritis included 10 randomized trials (Fransen, 2014).

Pain and function

In nine trials with 549 participants, exercise reduced pain with a standardized mean difference of 0.38 (Fransen, 2014). The reduction is equivalent to 8 points on a 0 to 100 scale, with a number needed to treat of 6 (Fransen, 2014).

Physical function improved by 7 points on the same scale, also with a number needed to treat of 6 (Fransen, 2014). Quality of life, measured in only three small studies with 183 participants, showed no demonstrated benefit (Fransen, 2014).

The effect persists

The pain reduction persisted three to six months after the end of supervised treatment, in five trials with 391 participants (Fransen, 2014). The reported adverse events were few and linked to increased pain during the program (Fransen, 2014).

Knee and hip together

A meta-analysis of 77 trials with 6,472 participants assessed knee and hip together (Goh, 2019). Exercise reduced pain with an effect size of 0.56 near 8 weeks (Goh, 2019).

The gain was greater in trials with younger participants, with knee osteoarthritis and without planned prosthesis surgery (Goh, 2019). This pattern suggests that exercise should start early in the course of the disease, and not as a last attempt before the prosthesis.

Exercise, painkillers and surgery

The patient in the case uses an anti-inflammatory daily and is thinking about surgery. The natural question is how exercise compares with these options.

Exercise versus anti-inflammatories and paracetamol

A network meta-analysis of 152 trials with 17,431 participants compared exercise with oral anti-inflammatories and paracetamol in knee and hip osteoarthritis (Weng, 2023). Pain and function were similar between exercise and drugs at 4, 8 and 24 weeks (Weng, 2023).

The authors recall the much better safety profile of exercise, especially in older adults with comorbidities and a higher risk of adverse events from the drugs (Weng, 2023). In a patient with hypertension, kidney disease or risk of gastrointestinal bleeding, this weighs heavily.

Reviewing continuous use of an anti-inflammatory is an opportunity to decide with the patient what to stop. In chronic pain, it is also worth knowing the difference between physical dependence, tolerance and substance use disorder before escalating to opioids.

Arthroscopy in the degenerative knee

A meta-analysis of nine trials assessed arthroscopy with partial meniscectomy, debridement or both in middle-aged and older adults with a degenerative knee (Thorlund, 2015). The benefit on pain was 2.4 mm on a 100 mm visual scale (Thorlund, 2015).

This small gain appeared at three and six months and disappeared between one and two years after surgery (Thorlund, 2015). Physical function showed no significant benefit (Thorlund, 2015).

The harms included symptomatic deep vein thrombosis, with 4.13 events per 1,000 procedures, as well as pulmonary embolism, infection and death (Thorlund, 2015). Taken together, these data show a surgery with little gain and real risk.

Exercise versus meniscectomy

A Norwegian trial randomized 140 middle-aged adults with degenerative medial meniscus injury seen on MRI to 12 weeks of supervised exercise or arthroscopic partial meniscectomy (Kise, 2016). Mean age was 49.5 years (Kise, 2016).

After two years, the difference in the KOOS4 score between the groups was 0.9 points, without clinical relevance (Kise, 2016). Thigh strength improved more in the exercise group at three months (Kise, 2016).

About 19 of every 100 participants in the exercise group crossed over to surgery during follow-up, without additional benefit (Kise, 2016). No serious adverse event occurred in either group (Kise, 2016).

The evidence side by side

Study	Design and population	Main result
Fransen, 2015	Cochrane review of 54 studies in knee osteoarthritis (Fransen, 2015)	Pain 12 points lower on a 0 to 100 scale (Fransen, 2015)
Fransen, 2014	Cochrane review of 10 trials in hip osteoarthritis (Fransen, 2014)	Pain 8 points lower, number needed to treat of 6 (Fransen, 2014)
Goh, 2019	Meta-analysis of 77 trials, 6,472 participants (Goh, 2019)	Peak effect near 2 months, loss after 9 months (Goh, 2019)
Weng, 2023	Network meta-analysis of 152 trials, 17,431 participants (Weng, 2023)	Pain and function equal to those of anti-inflammatories and paracetamol (Weng, 2023)
Yan, 2025	Network meta-analysis of 217 trials, 15,684 participants (Yan, 2025)	Aerobic exercise with the greatest effect on pain in the short term (Yan, 2025)
Thorlund, 2015	Meta-analysis of 9 arthroscopy trials (Thorlund, 2015)	Gain of 2.4 mm in pain, absent at one to two years (Thorlund, 2015)
Kise, 2016	Trial with 140 adults with degenerative meniscus injury (Kise, 2016)	Difference of 0.9 points in KOOS4 at two years (Kise, 2016)
Messier, 2013	IDEA trial with 454 overweight adults (Messier, 2013)	Diet plus exercise reduced pain more than exercise alone (Messier, 2013)

And what about the prosthesis?

Total arthroplasty remains a good option for advanced osteoarthritis with pain and limitation that persist after well-conducted conservative treatment. Structured exercise prepares the patient for surgery and, in many cases, postpones the indication. The decision is shared and depends on how much pain limits the person's life, not on the appearance of the radiograph.

Weight, load and the IDEA trial

Excess weight increases the load on the knee with every step and feeds low-grade inflammation. That is why the guideline strongly recommends weight loss for people with knee or hip osteoarthritis and excess weight (Kolasinski, 2020).

The study design

The IDEA trial randomized 454 adults aged 55 or older, with a body mass index between 27 and 41 and painful knee osteoarthritis (Messier, 2013). The groups received intensive diet, diet plus exercise or exercise alone for 18 months (Messier, 2013).

At the end of the study, 399 participants completed follow-up (Messier, 2013). Mean weight loss was 10.6 kg in the diet plus exercise group and 1.8 kg in the exercise group (Messier, 2013).

The results

Pain, on a scale from 0 to 20, was 3.6 in the diet plus exercise group and 4.7 in the exercise alone group (Messier, 2013). Function was also better in the combined group (Messier, 2013).

Knee compressive force fell more in the diet group than in the exercise group (Messier, 2013). Interleukin 6, an inflammatory marker, was lower in the diet groups (Messier, 2013).

What this changes in the consultation

For the patient with excess weight, the message is to combine movement and weight loss, not to choose one of the two. When weight loss is difficult, obesity drugs can enter the conversation, as discussed in the text on what the tirzepatide studies showed.

How to prescribe in practice

Knowing that exercise works is the easy part. The hard part is getting the patient to start and keep going.

The Danish GLA:D model

The Danish GLA:D program combines three education sessions over two weeks with 12 supervised neuromuscular exercise sessions, twice a week, for six weeks (Skou, 2017). Trained physiotherapists lead the sessions (Skou, 2017).

In data from 9,825 participants in the national registry, pain fell 12.4 points on a scale from 0 to 100 at three months and 13.7 points at 12 months (Skou, 2017). Fewer patients were using analgesics and fewer were on sick leave after one year (Skou, 2017).

Despite the good result, only about 20 out of every 100 Danish municipalities offered the program at the time of the report (Skou, 2017). Even where the treatment exists, reaching the patient is a challenge.

Education that changes beliefs

Many patients think that pain when moving means damage to the joint. Explain that some discomfort during and shortly after exercise is expected and does not indicate structural worsening. A useful rule is to accept mild pain that returns to the usual level within one day.

The Cochrane review showed similar dropout between exercise and control, which helps reassure those who fear they cannot cope (Fransen, 2015). Motivational interviewing techniques such as the OARS skills help turn the recommendation into the patient's own decision.

How to build the program

Start with strengthening of the quadriceps, glutes and hip abductors, plus low-impact aerobic activity such as walking, and balance training. The ACR guideline accepts walking, strengthening, neuromuscular training and aquatic exercise (Kolasinski, 2020).

Progress the load gradually and adjust intensity by pain. Supervised programs at the start tend to give better results; the Cochrane review noted more effect with individual programs than with group classes or home exercise (Fransen, 2015).

Keeping it up is the challenge

The effect of exercise peaks at around 2 months and falls until it equals usual care between 9 and 18 months (Goh, 2019). This pattern reflects stopping the practice, not a failure of exercise.

The conclusion is that the program needs to become a habit. Activities the person enjoys, groups and periodic follow-ups help maintain it. The same reasoning appears in the text on how to prescribe exercise in depression.

When to look for another cause

Severe pain at rest, night pain, fever, weight loss, a large joint effusion or true locking call for assessment before attributing everything to osteoarthritis. Knee pain with a normal knee examination may come from the hip.

Avoid cascades

Treating the patient conservatively, with method and follow-up, protects against tests and procedures that add little. The idea is the same one described in the text about cascades of care.

Back to the case

For the 63-year-old man, the plan is to explain that the radiograph confirms osteoarthritis but does not indicate surgery by itself. Propose a supervised program of strengthening and walking for a few weeks, review daily anti-inflammatory use and discuss weight loss if there is excess weight. Schedule a return visit to adjust the program and reinforce adherence.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by looking for the ICD-10 code.

The limits of the game's use are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information under curation and under creators, and institutions can learn about the proposal on the partnerships page. For the most frequent reasons for consultation, see the most common codes in primary care.

Frequently asked questions

Does exercise make osteoarthritis worse?

Cochrane reviews show reduced pain and improved function, with no serious adverse events reported in the knee studies (Fransen, 2015). Mild discomfort during the program is expected and passes.

How much does exercise reduce knee pain?

About 12 points on a 0 to 100 scale right after treatment, according to the Cochrane review (Fransen, 2015). The effect falls to about 6 points two to six months later (Fransen, 2015).

Does exercise work for the hip?

It does, with an 8-point reduction in pain and a number needed to treat of 6 in the Cochrane review (Fransen, 2014). The evidence is weaker than for the knee.

Is exercise as good as anti-inflammatories?

In a network meta-analysis of 152 trials, pain and function were similar between exercise and oral anti-inflammatories or paracetamol (Weng, 2023). Exercise has a better safety profile (Weng, 2023).

Is it worth operating on a degenerative meniscus?

In middle-aged adults, 12 weeks of exercise had the same result as partial meniscectomy after two years (Kise, 2016). Arthroscopy in the degenerative knee brings a small, temporary gain, with risks (Thorlund, 2015).

Which type of exercise should be chosen?

Walking, strengthening, neuromuscular training, exercise in water and tai chi are options accepted by the ACR guideline (Kolasinski, 2020). The best modality is the one the patient keeps up.

Does losing weight help?

Yes. In the IDEA trial, diet plus exercise reduced pain more than exercise alone in adults with excess weight and knee osteoarthritis (Messier, 2013).

Take-home messages

• Exercise is first-line treatment in knee and hip osteoarthritis, with a strong recommendation from the ACR.
• In the knee, exercise reduces pain and improves function, with an effect comparable to that of anti-inflammatories.
• In the hip, the evidence is weaker, but it points to benefit in pain and function.
• Imaging confirms osteoarthritis but does not define the initial treatment.
• Arthroscopy in the degenerative knee brings a small, temporary gain, with risks.
• Weight loss added to exercise works better in people with excess weight.
• The effect of exercise is lost when practice stops; the program needs to become a habit.
• Glucosamine, chondroitin and transcutaneous electrical stimulation have no place in treatment.

Play the case of the day on NEXO.

References

1. Fransen M, McConnell S, Harmer AR, et al. Exercise for osteoarthritis of the knee. Cochrane Database Syst Rev. 2015. PMID 25569281. DOI 10.1002/14651858.CD004376.pub3.
2. Fransen M, McConnell S, Hernandez-Molina G, et al. Exercise for osteoarthritis of the hip. Cochrane Database Syst Rev. 2014. PMID 24756895. DOI 10.1002/14651858.CD007912.pub2.
3. Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis Rheumatol. 2020. PMID 31908163. DOI 10.1002/art.41142.
4. GBD 2021 Osteoarthritis Collaborators. Global, regional, and national burden of osteoarthritis, 1990-2020 and projections to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Rheumatol. 2023. PMID 37675071. DOI 10.1016/S2665-9913(23)00163-7.
5. Yan L, Li D, Xing D, et al. Comparative efficacy and safety of exercise modalities in knee osteoarthritis: systematic review and network meta-analysis. BMJ. 2025. PMID 41093618. DOI 10.1136/bmj-2025-085242.
6. Goh SL, Persson MSM, Stocks J, et al. Efficacy and potential determinants of exercise therapy in knee and hip osteoarthritis: A systematic review and meta-analysis. Ann Phys Rehabil Med. 2019. PMID 31121333. DOI 10.1016/j.rehab.2019.04.006.
7. Weng Q, Goh SL, Wu J, et al. Comparative efficacy of exercise therapy and oral non-steroidal anti-inflammatory drugs and paracetamol for knee or hip osteoarthritis: a network meta-analysis of randomised controlled trials. Br J Sports Med. 2023. PMID 36593092. DOI 10.1136/bjsports-2022-105898.
8. Thorlund JB, Juhl CB, Roos EM, et al. Arthroscopic surgery for degenerative knee: systematic review and meta-analysis of benefits and harms. BMJ. 2015. PMID 26080045. DOI 10.1136/bmj.h2747.
9. Kise NJ, Risberg MA, Stensrud S, et al. Exercise therapy versus arthroscopic partial meniscectomy for degenerative meniscal tear in middle aged patients: randomised controlled trial with two year follow-up. BMJ. 2016. PMID 27440192. DOI 10.1136/bmj.i3740.
10. Messier SP, Mihalko SL, Legault C, et al. Effects of intensive diet and exercise on knee joint loads, inflammation, and clinical outcomes among overweight and obese adults with knee osteoarthritis: the IDEA randomized clinical trial. JAMA. 2013. PMID 24065013. DOI 10.1001/jama.2013.277669.
11. Skou ST, Roos EM. Good Life with osteoArthritis in Denmark (GLA:D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide. BMC Musculoskelet Disord. 2017. PMID 28173795. DOI 10.1186/s12891-017-1439-y.