Your knee is where it hurts. It’s rarely where it started.

If your knee has started hurting when you run, you have almost certainly been told your glutes are weak.

It is the most repeated piece of advice in the whole of running injury, and it comes from everywhere — physios, coaches, training partners, the internet. Strengthen your hips, the knee pain goes away.

The exercises are worth doing. We’ll come back to why. But the explanation attached to them doesn’t survive contact with the evidence, and that matters, because when the story about a problem is wrong, so is everything you do to prevent it.

To be clear about what this post argues: when we say your knee is rarely where the problem started, we don’t mean it started somewhere else in your body. We mean it started in your training.

The advice everyone with runner’s knee has already had

Patellofemoral pain — runner’s knee — is pain at the front of the knee, around or behind the kneecap, that shows up with running, stairs, hills and sitting for a long time with the knee bent.

The standard model goes like this: your hip abductors are weak, so your thigh drops inwards when you land, so your kneecap tracks badly, so it hurts. It’s a satisfying story. It has a villain, a mechanism and a fix.

What happens when you actually test it

The problem is that the story was built almost entirely on cross-sectional research — studies that measure people who already have knee pain and compare them with people who don’t.

Those studies are consistent: adults with patellofemoral pain do have weaker hips.

But that design can’t tell you which came first. To do that you have to measure healthy people, follow them, and see who develops the problem. When researchers have done exactly that, the finding reverses.

A 2014 systematic review and meta-analysis found moderate-to-strong evidence from prospective studies of no association between hip strength and the risk of developing patellofemoral pain. The authors’ conclusion is worth reading twice: reduced hip strength “may be a result of PFP rather than the cause.”

A larger 2019 meta-analysis pooled 18 prospective studies covering 4,818 participants, of whom 483 went on to develop the condition. Hip weakness was not a risk factor there either.

In other words: the weak hips are real, and they probably arrived with the pain rather than before it.

We’ve been assessing the right things the wrong way round.

The other things that don’t predict patellofemoral pain

While that 2019 review was at it, it tested most of the other things runners get told about their knees. Not risk factors for future patellofemoral pain: Q angle, BMI, body fat, height and weight.

Q angle is the one worth dwelling on. It’s a measure of how your thigh meets your shin — the basis of being told your knees are the wrong shape, or that your hips are too wide, or that your alignment was always going to cause this. It does not predict who gets patellofemoral pain.

One strength measure did survive prospective testing: quadriceps weakness — in military recruits. That’s the population it was found in, and it shouldn’t be quietly extended to recreational runners. It’s also notable for pointing at the knee rather than away from it.

And for a sense of how little any of this is settled: a study of 60,997 distance race entrants found patellofemoral pain was independently associated with a higher burden of chronic disease, and with a history of allergies. That’s cross-sectional too, so the same caution applies. But it’s a useful reminder that even enormous datasets throw up associations nobody went looking for.

So why does the rehab work?

Here’s the part that makes this more than a debunking exercise, and it’s the reason to keep doing your exercises.

The 2018 international consensus statement on treating patellofemoral pain — the work of 41 experts reviewing the trial evidence — recommends exercise therapy, and specifically the combination of hip-focused and knee-focused exercises, to improve pain and function.

It also tells you what not to bother with. Not recommended: knee, kneecap or lower back mobilisations on their own, or electrophysical agents. Listed as uncertain: taping, bracing, dry needling, manual soft tissue work.

So the treatment is well supported. The mechanism everyone recites for why it works is not.

There is a better explanation, and it has nothing to do with tracking or alignment.

Loading tissue progressively increases its capacity to absorb and produce force. That happens locally, in the quadriceps and the structures around the joint itself. It also happens remotely — in the calf and plantar fascia below, and in the glutes, hamstrings and adductors above. All of them contribute to what your leg does each time your foot hits the ground.

Raise tolerance in both places and the patellofemoral joint sits further below its threshold of irritability. Same running, same joint, more headroom before symptoms appear.

That also accounts for something the alignment story can’t: why the consensus favours hip-focused and knee-focused exercise in combination rather than either on its own. You aren’t correcting a fault in one location. You’re raising the ceiling in several.

It explains why assessing for a cause up front doesn’t work, either. Raising tissue tolerance helps whether or not that tolerance was unusually low to begin with — so a weakness you could have measured beforehand was never going to predict who develops this. Treating a problem and predicting it are different jobs.

Which is why being told your hips were weak all along isn’t the explanation it sounds like. They may well be weak. That isn’t the same as it being the reason.

What we don’t know

Nobody has a confident cause for patellofemoral pain. Including us.

The honest position is that it looks like a load problem — the tissue being asked for more than it can currently tolerate — and that’s where the broader running injury evidence points. We’ve written about that mechanism in increasing training volume without getting injured and in when you’ll actually get injured in an Ironman build.

But patellofemoral pain specifically has thinner evidence than you’d expect. One prospective study followed 586 runners for 24 weeks with GPS-tracked training and tested whether increasing weekly volume beyond 10% combined with increasing pace predicted injury. It didn’t reach significance. And a 2025 evidence-gap map of patellofemoral pain found only two of twenty-two prognosis studies were conducted in runners at all.

So when we say it’s about load, we’re reasoning from the general running injury evidence and from what we see in clinic — not from a body of research that has tested that claim in this condition. That’s a real distinction and it belongs in the open.

Cadence: useful, but not the way you think

One thing that does have good biomechanical evidence behind it is step rate.

Increasing running cadence by 7.5% reduced peak patellofemoral joint contact force by half a body weight. That’s a substantial change for a small adjustment, and it’s part of why “take shorter, quicker steps” has become common advice.

The caveat in that same study is the part that rarely gets repeated. Raising cadence reduced knee load per step, but only partially reduced cumulative knee load per kilometre. You take smaller bites — you also take more of them.

Which is the whole point in miniature. Changing your form lowers the cost of each step. It doesn’t make the distance free.

As for formal gait retraining, the 2018 consensus listed it as uncertain. Two small randomised trials since have been more encouraging — one found both impact-focused and cadence-focused retraining beat no intervention for running pain at six months, the other favoured retraining over foot orthoses. Both are small. It’s promising, not proven.

What we do know: it’s slow

If there’s one number in this post worth remembering, it’s this one.

In a prospective study of 254 injured novice runners, patellofemoral pain accounted for 10% of all injuries. Across that whole injured group, the median time to recovery was 71 days, and half still could not run 2×500m without pain after ten weeks. The study didn’t break those timeframes out by diagnosis — but patellofemoral pain is not the quick one on that list.

That’s the real argument for dealing with this early, and it isn’t that early treatment is magic. It’s that the treatment that works takes time to work, and the clock starts when you begin, not when you decide it’s gone on long enough.

If your knee has been grumbling for a fortnight and isn’t trending down, it isn’t settling on its own.

---

On Form Physiotherapy is Perth’s endurance sport clinic. If your knee has been bothering you for more than a couple of weeks, book an assessment or call (08) 6385 8029.

---

Sources

According to PubMed:

• Rathleff MS, Rathleff CR, Crossley KM, Barton CJ. Is hip strength a risk factor for patellofemoral pain? A systematic review and meta-analysis. British Journal of Sports Medicine 2014;48(14):1088. DOI
• Neal BS, Lack SD, Lankhorst NE, Raye A, Morrissey D, van Middelkoop M. Risk factors for patellofemoral pain: a systematic review and meta-analysis. British Journal of Sports Medicine 2019;53(5):270–281. DOI
• Collins NJ, Barton CJ, van Middelkoop M, et al. 2018 Consensus statement on exercise therapy and physical interventions to treat patellofemoral pain. British Journal of Sports Medicine 2018;52(18):1170–1178. DOI
• Marandure TT, Schwellnus MP, Grant C, et al. Patellofemoral pain syndrome is associated with chronic disease and allergies in 60 997 distance runner race entrants: SAFER XXX. Clinical Journal of Sport Medicine 2023;33(6):603–610. DOI
• Ramskov D, Rasmussen S, Sørensen H, Parner ET, Lind M, Nielsen R. Interactions between running volume and running pace and injury occurrence in recreational runners. Journal of Athletic Training 2022;57(6):557–563. DOI
• Neal BS, Bolgla LA, Lack SD, et al. Prognosis of patellofemoral pain: a systematic review with evidence- and gap-map. Journal of Orthopaedic & Sports Physical Therapy 2025;55(10):661–670. DOI
• Willy RW, Willson JD, Clowers K, Baggaley M, Murray N. The effects of body-borne loads and cadence manipulation on patellofemoral and tibiofemoral joint kinetics during running. Journal of Biomechanics 2016;49(16):4028–4033. DOI
• de Souza Júnior JR, Rabelo PHR, Lemos TV, Esculier JF, Barbosa GMP, Matheus JPC. Effects of two gait retraining programs on pain, function, and lower limb kinematics in runners with patellofemoral pain: a randomized controlled trial. PLoS One 2024;19(1):e0295645. DOI
• Bonacci J, Hall M, Saunders N, Vicenzino B. Gait retraining versus foot orthoses for patellofemoral pain: a pilot randomised clinical trial. Journal of Science and Medicine in Sport 2018;21(5):457–461. DOI
• Nielsen RO, Rønnow L, Rasmussen S, Lind M. A prospective study on time to recovery in 254 injured novice runners. PLoS One 2014;9(6):e99877. DOI