The knee that has hurt for months: why it is not 'growing pains'
Among adolescents with recurrent knee pain, 72% are girls, the median duration of pain is 24 months, and five years later 40.5% still hurt — with 60% of those having stopped or cut back on sport. Patellofemoral pain does not go away by 'growing out of it': it goes away when someone takes it seriously and doses the load. The data, and what changes in the gym.
There is one injury in young female athletes that appears on no medical report and costs no single match: it costs years. It is the knee that has hurt for months — on stairs, in a squat, after twenty minutes sitting at a school desk — filed away as growing pains, or as something that "will sort itself out once she stops growing". The research says the opposite: in adolescents followed over time, five years later 40.5% still hurt, and 60% of them had stopped or cut back on sport because of the knee (Rathleff et al., 2019).
Key points
- Among adolescents with at least monthly knee pain, 363 out of 504 are girls (72%) and the median duration of pain is 24 months; only 59% had sought healthcare (Rathleff et al., 2013).
- Two years later, 55.9% still have knee pain; with patellofemoral pain the relative risk is 1.26 versus other types of knee pain (Rathleff et al., 2016).
- Five years later: 40.5% still in pain (versus 13.2% of controls), 60% stopped or reduced sport, 15% say the knee influenced their choice of job (Rathleff et al., 2019).
- It is not rare: annual prevalence of patellofemoral pain in adolescents is estimated at 28.9% (Smith et al., 2018).
- What worked was not rest: it was modifying the load and then strengthening. In the 12-week pathway with 151 girls and boys aged 10-14, 86% successful outcomes at 12 weeks and 81% at one year (Rathleff et al., 2019).
How common is knee pain in young female athletes?
Far more common than the number of girls who mention it would suggest. The most cited systematic review with meta-analysis on the topic pooled 23 studies and estimates an annual prevalence of patellofemoral pain of 28.9% in adolescents, versus 22.7% in the general population; in adolescent amateur athletes, incidence over a single season ranges from 5.1% to 14.9% (Smith et al., 2018, PLOS ONE). The authors themselves flag that the estimates rest on few studies and inconsistent definitions: treat the exact figure with caution, but not the order of magnitude.
Population data point the same way. In a sample of 2,953 Danish adolescents aged 12-19, knee pain was reported by 35.0% of girls versus 27.9% of boys, and almost-daily pain at any site by 23.8% versus 13.3% (Rathleff et al., 2013, BMC Pediatrics).
And when you look only at those with recurrent knee pain, the imbalance becomes obvious: among 504 adolescents aged 15-19 with at least monthly knee pain, 363 were girls — 72% (Rathleff et al., 2013, BMC Musculoskeletal Disorders). In the same cohort, 68.3% of the pain had an insidious onset, i.e. no trauma to point to.
This is exactly the kind of pain that enters no statistic in youth sport: no episode, no collision, no day on which it started.
What patellofemoral pain is, and what it is not
It is diffuse pain around or behind the kneecap, triggered by loading the knee in flexion: stairs, squats, jumps, running, and — the detail that often gives it away — sitting for long periods with the knees bent. Whoever has it cannot point to it with one finger: she points with a whole hand, "around here".
It needs to be told apart from two things we have already covered, because management differs:
- It is not an ACL rupture, which is a single event, almost always with a precise moment and a knee giving way.
- It is not a stress fracture, whose hallmark is pinpoint bone pain, indicated with one finger, always at the same spot.
One thing to say plainly: this article is not a diagnostic tool. Knee pain in an adolescent girl has several possible causes — some trivial, some not — and the distinction is made by a healthcare professional. What matters here is something else: that this kind of pain is not a reason to wait.
Does adolescent knee pain go away on its own?
In most cases it does not, and this is the question the literature has answered best. One study followed Danish adolescents with knee pain for two years: at follow-up, 55.9% still reported pain. Those classified with patellofemoral pain had a relative risk of 1.26 of still hurting compared with other types of knee pain. The study's title is a question — is knee pain during adolescence a self-limiting condition? — and the authors conclude that in most cases the pain is still present after two years and therefore may not be self-limiting (Rathleff et al., 2016, Am J Sports Med).
At five years the picture had not straightened out (Rathleff et al., 2019, BMJ Open):
- 40.5% of those with baseline pain still had it (95% CI 35.4-45.6), versus 13.2% of those without (95% CI 8.2-18.1).
- Among those still in pain, 60% had stopped or reduced sport because of the knee.
- 15% (95% CI 12-20) said knee pain had influenced their choice of job or career.
- Much worse functional scores: -13 points on KOOS Function and -30 points on KOOS Sport/recreation; roughly one third used painkillers regularly.
It is worth being precise about the population: these are Danish adolescents aged 15-19 of both sexes, recruited from the school population rather than from registered athletes. They are not 13-year-olds in an Italian club, and the numbers do not transfer one to one. But the direction — pain that over time takes away the sport rather than itself — is measured, not assumed.
There is a second finding, this one specific to girls, that makes waiting even harder to defend: female adolescents with patellofemoral pain have lower pressure pain thresholds than pain-free peers, including at body sites far from the knee (Rathleff et al., 2013, JOSPT). In other words, pain lasting months does not stay a local problem — it changes how the nervous system processes stimuli, as we described in how pain changes during puberty.
Why almost nobody takes it to a doctor
Because it feels too small to bother anyone with, and too normal to be a problem. In the cohort of 504 adolescents with at least monthly knee pain, the median duration of pain was 24 months and only 59% had sought healthcare; 18% were under treatment at the time of the questionnaire (Rathleff et al., 2013, BMC Musculoskeletal Disorders).
The most instructive detail, though, is another. Among girls, those with traumatic-onset pain sought help in 80% of cases; those with insidious-onset pain in only 55.7%. In boys the difference between onset types was minimal (47.9% versus 49.5%).
Put bluntly: an accident seems to be needed before it feels allowed to ask. It is the same dynamic that keeps urine leakage during jumping and menstruation in front of a coach unspoken — a symptom without an event to justify it does not feel serious enough to name. And pain that is never named is pain nobody can dose.
What works: not rest, but measured load
In the studies, stopping was not what worked: modifying the load and then strengthening was. This is the counter-intuitive part, and the most useful one in the gym.
The study closest to the age band that matters followed 151 adolescents aged 10-14 with patellofemoral pain through a 12-week pathway: activity modification in weeks 1-4, home exercises from week 5 to 8, graded return to sport from week 9 to 12, with only four supervised sessions, each of which a parent also had to attend. Self-reported successful outcomes were 86% at 12 weeks, 77% at 6 months and 81% at 12 months; hip and knee strength rose by 20-33% and return to sport reached 81% at one year (Rathleff et al., 2019, Am J Sports Med).
A limit to state up front: that study has no control group, so on its own it cannot show that the effect came from the intervention rather than from time. The randomised evidence comes from an older population: in a cluster trial of 121 adolescents aged 15-19, adding supervised exercise during school hours to education alone increased the odds of recovery at 12 months (OR 1.73; 95% CI 1.02-2.93; NNT 11), and the effect grew at 24 months (OR 2.52; NNT 5). More exercise sessions per week, higher odds of recovery (Rathleff et al., 2015, Br J Sports Med).
On the type of work, the international consensus statement is clear: exercise therapy is recommended, in particular the combination of hip-focused and knee-focused exercise; joint mobilisations in isolation and electrophysical agents (ultrasound, electrical currents and the like) are not recommended (Collins et al., 2018, Br J Sports Med). One honest note: that consensus concerns the general patellofemoral pain population, largely adult — in adolescents the evidence is the studies above.
The practical point that holds it all together: a knee does not heal by staying still, it heals by getting the right dose. It is the same principle behind early specialisation and overuse: load is not the enemy, unmanaged load is.
And the knee that hurts below the kneecap (Osgood-Schlatter)?
Different picture, same approach. Pain localised over the tibial tuberosity — the bony prominence just below the kneecap, where the patellar tendon attaches — is typical of adolescence during growth and is called Osgood-Schlatter disease. Here too, the idea that you just need to wait has been tested: in a prospective study of 51 adolescents aged 10-14 (51% girls), a 12-week pathway using a load-progression ladder for the patellar tendon, strengthening exercises and a graded return to sport produced 80% self-reported successful outcomes at 12 weeks and 90% at 12 months (Rathleff et al., 2020, Orthop J Sports Med).
This too is a study without a control group and in a small sample: not proof of efficacy, but a consistent signal. And the message is the same, worth repeating: load progression is an intervention; waiting is not.
What a club can do, starting Monday
None of these is a medical prescription. They are organisational and language choices, within reach of anyone who coaches:
- Ask "how long?", not "how much does it hurt?". Today's intensity says little; duration says almost everything. Beyond two or three weeks, pain belongs with a healthcare professional, not with pitch-side management.
- Ask "can you point to it?". A whole hand around the kneecap is one picture; one finger on a precise spot of bone is another, with a different urgency (stress fractures).
- Do not wait for an accident to grant permission. In the data, girls with non-traumatic pain ask for help far less often. Saying it out loud once a year — "pain that keeps coming back does not need an injury to be worth mentioning" — costs thirty seconds.
- Have a "reduced-load" mode, not just "in or out". An athlete with knee pain who can do half a session stays in the group; one sent home to rest completely disappears, and often does not come back. That is precisely the mechanism feeding drop-out during puberty.
- Put hip and knee strength in everyone's programme, not just for those in pain. It is the content the consensus recommends, and in practice it occupies the same slot as the neuromuscular warm-up that protects the ACL.
Where BAB fits
BAB does not diagnose, does not treat and does not decide when someone returns to play. It does something almost nobody does today: it makes duration visible. An athlete privately logs pain, perceived load, energy and recovery, and sees a line form — not the single bad day, but the six weeks in which that knee kept coming back. Clubs receive only aggregated, anonymous signals, never an individual's health data.
Because in this story the decisive number is not pain intensity. It is 24 months — the median time it takes for these knees to finally reach somebody.
Sources
- Smith B.E., Selfe J., Thacker D., Hendrick P., Bateman M., Moffatt F., Rathleff M.S., Smith T.O., Logan P. Incidence and prevalence of patellofemoral pain: A systematic review and meta-analysis. PLOS ONE, 2018;13(1):e0190892. (systematic review of 23 studies; annual prevalence 28.9% in adolescents and 22.7% in the general population; the authors flag few studies and heterogeneous definitions) doi:10.1371/journal.pone.0190892
- Rathleff C.R., Olesen J.L., Roos E.M., Rasmussen S., Rathleff M.S. High prevalence of daily and multi-site pain — a cross-sectional population-based study among 3000 Danish adolescents. BMC Pediatrics, 2013;13:191. (2,953 Danish adolescents aged 12-19, both sexes; knee pain 35.0% in girls versus 27.9% in boys) doi:10.1186/1471-2431-13-191
- Rathleff M.S., Skuldbøl S.K., Rasch M.N.B., Roos E.M., Rasmussen S., Olesen J.L. Care-seeking behaviour of adolescents with knee pain: a population-based study among 504 adolescents. BMC Musculoskeletal Disorders, 2013;14:225. (504 Danish adolescents aged 15-19 with at least monthly knee pain, of whom 363 girls — 72%; median duration 24 months; 68.3% insidious onset) doi:10.1186/1471-2474-14-225
- Rathleff M.S., Rathleff C.R., Olesen J.L., Rasmussen S., Roos E.M. Is Knee Pain During Adolescence a Self-limiting Condition? Prognosis of Patellofemoral Pain and Other Types of Knee Pain. The American Journal of Sports Medicine, 2016;44(5):1165-1171. (2-year follow-up; 55.9% still in pain; relative risk 1.26 for patellofemoral pain) doi:10.1177/0363546515622456
- Rathleff M.S., Holden S., Straszek C.L., Olesen J.L., Jensen M.B., Roos E.M. Five-year prognosis and impact of adolescent knee pain: a prospective population-based cohort study of 504 adolescents in Denmark. BMJ Open, 2019;9(5):e024113. (5-year follow-up; 40.5% still in pain versus 13.2% of controls; 60% stopped or reduced sport; adolescents of both sexes) doi:10.1136/bmjopen-2018-024113
- Rathleff M.S., Roos E.M., Olesen J.L., Rasmussen S. Exercise during school hours when added to patient education improves outcome for 2 years in adolescent patellofemoral pain: a cluster randomised trial. British Journal of Sports Medicine, 2015;49(6):406-412. (cluster randomised trial in 121 adolescents aged 15-19; odds of recovery 1.73 at 12 months and 2.52 at 24 months) doi:10.1136/bjsports-2014-093929
- Rathleff M.S., Graven-Nielsen T., Hölmich P., Winiarski L., Krommes K., Holden S., Thorborg K. Activity Modification and Load Management of Adolescents With Patellofemoral Pain: A Prospective Intervention Study Including 151 Adolescents. The American Journal of Sports Medicine, 2019;47(7):1629-1637. (151 adolescents aged 10-14; prospective study without a control group, therefore not proof of efficacy; 86% successful outcomes at 12 weeks, 81% at 12 months) doi:10.1177/0363546519843915
- Rathleff M.S., Winiarski L., Krommes K., Graven-Nielsen T., Hölmich P., Olesen J.L., Holden S., Thorborg K. Activity Modification and Knee Strengthening for Osgood-Schlatter Disease: A Prospective Cohort Study. Orthopaedic Journal of Sports Medicine, 2020;8(4):2325967120911106. (51 adolescents aged 10-14, 51% girls; prospective cohort without a control group; 80% successful outcomes at 12 weeks, 90% at 12 months) doi:10.1177/2325967120911106
- Rathleff M.S., Roos E.M., Olesen J.L., Rasmussen S., Arendt-Nielsen L. Lower Mechanical Pressure Pain Thresholds in Female Adolescents With Patellofemoral Pain Syndrome. Journal of Orthopaedic & Sports Physical Therapy, 2013;43(6):414-421. (cross-sectional study in female adolescents with patellofemoral pain and age-matched controls) doi:10.2519/jospt.2013.4383
- Collins N.J., Barton C.J., van Middelkoop M., Callaghan M.J., Rathleff M.S., Vicenzino B.T., Davis I.S., Powers C.M., Macri E.M., Hart H.F., de Oliveira Silva D., Crossley K.M. 2018 Consensus statement on exercise therapy and physical interventions (orthoses, taping and manual therapy) to treat patellofemoral pain: recommendations from the 5th International Patellofemoral Pain Research Retreat, Gold Coast, Australia, 2017. British Journal of Sports Medicine, 2018;52(18):1170-1178. (consensus statement on a patellofemoral pain population that is largely adult) doi:10.1136/bjsports-2018-099397
This article is for information and education only and does not constitute medical advice or a diagnostic tool. The cohorts cited involve adolescents of both sexes, mostly Danish and aged 15-19, and do not transfer directly to a 13-14 year-old athlete. Knee pain persisting beyond two to three weeks, or accompanied by swelling, the knee giving way or locking, or following trauma, should be assessed by a paediatrician, sports physician or physiotherapist: decisions on load and return to activity belong with whoever is looking after the athlete clinically.
Does adolescent knee pain go away on its own with growth?
Often it does not, and this is the question the research has answered most clearly. In a Danish cohort of adolescents followed over time, 55.9% of those reporting knee pain at baseline still reported it two years later; those classified with patellofemoral pain had a relative risk of 1.26 of still having pain compared with other types of knee pain. The authors explicitly conclude that adolescent knee pain, and patellofemoral pain in particular, 'may not be self-limiting' (Rathleff et al., 2016). At five years, 40.5% (95% CI 35.4-45.6) still had pain versus 13.2% of those pain-free at baseline, and among those still in pain, 60% had stopped or reduced sport because of the knee (Rathleff et al., 2019). Population: Danish adolescents aged 15-19 of both sexes, not specifically 13-14 year-old athletes.
What is patellofemoral pain and how do you recognise it?
It is diffuse pain around or behind the kneecap, triggered by activities that load the joint in flexion: stairs up and down, squatting, jumping, running, and also sitting for a long time with the knees bent. Unlike a stress fracture it cannot be pointed to with one finger on a precise spot of bone, and unlike an ACL rupture it does not start with a single traumatic episode: in the Danish cohort, 68.3% of adolescent knee pain had an insidious onset, i.e. no trauma (Rathleff et al., 2013). One caveat: this article is not a diagnostic tool. Knee pain in an adolescent girl has several possible causes, and the distinction is made by a healthcare professional, not by a symptom list.
Is rest the treatment?
Complete rest is not what worked in the studies: what worked was modifying the load and then strengthening. In a prospective study of 151 adolescents aged 10-14 with patellofemoral pain, a 12-week pathway built on activity modification in weeks 1-4, home exercises from week 5 and a graded return to sport from week 9 produced 86% self-reported successful outcomes at 12 weeks and 81% at 12 months, with hip and knee strength up by 20-33% (Rathleff et al., 2019). Note that this study had no control group, so on its own it does not prove the intervention works. The randomised evidence comes from an older group: in 121 adolescents aged 15-19, adding supervised exercise to education increased the odds of recovery at 12 months (OR 1.73; 95% CI 1.02-2.93) and more so at 24 months (OR 2.52), with a number needed to treat of 5 (Rathleff et al., 2015).
Why does knee pain seem to affect girls more?
Because in population data they report it more often. In a sample of 2,953 Danish adolescents aged 12-19, knee pain was reported by 35.0% of girls versus 27.9% of boys, and almost-daily pain at any site by 23.8% versus 13.3% (Rathleff et al., 2013). In the cohort of adolescents with at least monthly knee pain, 363 out of 504 were girls — 72% (Rathleff et al., 2013). One further finding, measured in girls only: female adolescents with patellofemoral pain have lower pressure pain thresholds than pain-free peers, including at sites far from the knee (Rathleff et al., 2013, JOSPT). That is not evidence of fragility: it is a sign that pain lasting months changes how the nervous system processes stimuli, which makes waiting even harder to justify.
When should she see a doctor?
Knee pain lasting more than two or three weeks, returning at every session, making her avoid stairs, squats or jumps, or costing her training sessions, should be assessed by a healthcare professional: a paediatrician, sports physician or physiotherapist. Some signs warrant assessment without waiting: swelling, the knee giving way or locking, pain after trauma, night pain, or pain that can be pointed to with one finger on a precise spot of bone. The reason not to wait is not alarmism, it is statistics: the median duration of pain in adolescents who eventually seek help is 24 months, and only 59% had sought care at all (Rathleff et al., 2013). Time is not an ally here.