If you coach or train clients who live with knee osteoarthritis (OA), you’ve probably seen the same pattern: They want to move, they know they “should” exercise, but pain and stiffness make every decision feel complicated. What’s safe? What will actually help? And with so many options—strength training, mobility work, tai chi, balance training, aquatic exercise—what should they prioritize when time, energy and confidence are limited?

A major new analysis in The BMJ offers a clearer starting point. After reviewing 217 randomized controlled trials spanning 1990 to 2024 and totaling 15,684 participants, researchers concluded that aerobic exercise is likely the most beneficial exercise modality for improving pain, function, gait performance and quality of life in people with knee OA.

That doesn’t mean other exercise types don’t matter. But it does strengthen the case that when you’re building an exercise plan for knee OA and you need evidence-based guidance, aerobic activity deserves to be the foundation.

Why This Question Matters in the Real World

Knee OA is common, especially as adults age, and it can be deeply limiting. Nearly 30% of people older than 45 show radiological evidence of knee OA, and about half of them also have severe knee symptoms. OA develops as cartilage cushioning the ends of bones wears down, leading to swelling, stiffness and discomfort, and the knee is the joint most commonly affected.

Exercise is already a cornerstone recommendation across guidelines, but the question of what type of exercise to choose has often been answered more by tradition and clinical experience than by research-based evidence. The study’s authors point out that current guidelines provide limited information on what specific exercise types to recommend, leaving much of today’s decision-making based on expert opinion.

This study set out to clarify that gray area.

Inside the Study

The researchers conducted a systematic review and network meta-analysis. This is a method that allows comparisons across multiple interventions even when every option hasn’t been directly tested against every other option in the same trial. Eligible studies were randomized controlled trials involving participants with symptomatic knee OA, diagnosed through clinical findings and/or imaging. Trials compared different exercise modalities—aerobic, flexibility, mind-body, neuromotor, strengthening and mixed programs—against either control groups or against other exercise approaches.

Importantly, the authors did not impose restrictions on dose, duration or intensity of the exercise interventions included in the analysis. Outcomes focused on the things your clients care about: pain, function, gait performance and quality of life. They looked at these outcomes at approximate follow-ups of four weeks, 12 weeks and 24 weeks.

Across the full network of trials, aerobic exercise consistently rose to the top, as it likely resulted in a large reduction in pain at both short-term and mid-term follow-ups. Additionally, aerobic exercise was associated with improvements in function at multiple time points, including a large increase at mid-term follow-up.

The analysis also found meaningful benefits for gait and quality of life. Aerobic exercise improved gait performance at mid-term follow-up and quality of life at short-term follow-up. And when the researchers ranked treatments by the probability of being “best” across outcomes, aerobic exercise had the highest overall probability.

Other modalities did show strengths. In the abstract, mind-body exercise was associated with a large increase in function at short-term follow-up, and neuromotor exercise resulted in a large increase in gait performance at short-term follow-up. The authors also describe strengthening and mixed exercise as improving function at mid-term follow-up, and flexibility exercise as potentially reducing pain at long-term follow-up.

In other words: aerobic exercise was found to be the most consistently strong “all-around” option, while other categories offered targeted boosts that could be considered useful additions when you’re building a complete program.

Safety: No Clear Differences in Adverse Events

When clients are wary of worsening pain or “making it worse,” safety data matters.

Adverse events were reported in a minority of studies (40 trials, 18%), and the authors found no clear differences in safety outcomes between exercise interventions and control groups. In other words, none of the exercise types resulted in more adverse events than controls, suggesting these approaches are generally safe.

Of course, that shouldn’t be seen as a green light to ignore pain signals, poor technique or individual limitations, but it is a helpful reassurance that exercise, broadly, belongs in the conversation.

A Few Important Caveats

Because this was a network meta-analysis, not every conclusion rests on direct, head-to-head comparisons. The authors note that many results came from indirect comparisons and should be interpreted with caution. They also cite potential “small study effects” for some outcomes at early time points, along with inconsistent reporting of socioeconomic characteristics that limited deeper analysis of who benefits most.

They also highlight a practical research gap: Incomplete reporting of exercise parameters such as intensity, frequency and duration can make it harder to connect “what was done” to “what worked.”

Still, the conclusion is clear: Aerobic exercise is likely the most beneficial exercise modality for knee OA outcomes.

Practical Application: How to Apply the Study Findings to Your Program Design

So, what do you do with this as a health and exercise professional?

Start by letting the evidence simplify—not complicate—your decision-making. The authors explicitly recommend aerobic exercise as a first-line intervention for knee OA management, particularly when the aim is to improve functional capacity and reduce pain. They also emphasize tailoring: exercise programming should be based on preferences, client goals, comorbidities and mobility levels to maximize adherence and long-term benefit. And, of course, input from the client’s healthcare team should be considered.

Here are some tips for translating these findings into safe and effective program design without turning it into a rigid formula.

When you’re choosing where to begin, make the “base” of the plan aerobic. This can be as straightforward as walking, cycling or swimming. That foundation can help you address pain and function simultaneously, which is often the difference between a client who sticks with movement and one who stops after a flare.

If you need to lower the barrier to entry, consider that the researchers found no clear differences among types of aerobic exercise. That gives you flexibility to match the mode to your client—what they can tolerate, what they enjoy and what they’ll actually do. For a client who hates the treadmill but will ride a bike while watching TV, that’s not a “compromise.” It’s an evidence-supported strategy.

When aerobic work isn’t possible—or isn’t possible yet—use the study’s guidance to keep clients moving in other structured ways. The authors note that if aerobic exercise isn’t possible due to individual limitations, alternative forms of structured physical activity may still be beneficial. That could mean leaning more heavily on mind-body, neuromotor, strengthening or mixed programs in the short term, then revisiting aerobic options as tolerance improves.

And when clients ask, “Should I also do strength work?” you don’t have to choose sides. The study suggests other modalities can provide complementary benefits. For example, mind-body exercise can provide short-term function gains, while neuromotor work can be used to achieve short-term gait improvements. The takeaway for programming is straightforward: Keep aerobic at the core when you can, and layer in other training elements to address specific needs like balance, confidence and movement quality.

Finally, talk about safety in a way that builds confidence. This analysis found no clear safety differences between exercise types and control, though data were limited. Use that to reinforce a key message: The goal isn’t to avoid movement—it’s to find the right movement, delivered in a way your client can sustain.

Remember, for knee OA, the “best” plan isn’t the most complicated one. It’s the one your client will do consistently, and this study makes a strong case that the most reliable first step is aerobic exercise.