Progressive loading for tendinopathy: what the evidence shows

Why loading is the headline treatment

If you have been told that the fix for a stubborn Achilles, a sore tennis elbow, a painful kneecap tendon, or a nagging hip is to rest it, you have heard advice that most current research pushes back against. For the family of overuse tendon problems known as tendinopathy, the treatment that keeps surfacing at the top of the evidence is not rest, injection, or surgery — it is structured, progressive strengthening of the painful tendon and the muscle attached to it. Two versions get the most study: "eccentric" exercise, which emphasises the lengthening phase of a muscle contraction (think of slowly lowering a heel off a step), and "heavy-slow resistance," which uses heavier loads moved slowly through both directions.

This piece lays out what the best available studies actually found, tendon by tendon — including where the effect looks solid, where it is modest, where the trials disagree, and where the honest answer is "we don't know as precisely as you'd hope." The goal is not to sell you on any one exercise. It is to show you the evidence as it stands so you can have a sharper conversation with a regulated health professional who can examine your specific problem.

What tendinopathy is, and why loading is the rationale

A tendon is the tough band that connects muscle to bone. Tendinopathy is the modern umbrella term for a painful, poorly-healing tendon that has been overloaded over time — it usually shows up as pain and stiffness that is worst when you first load the tendon and eases as it warms up, along with reduced tolerance for the activity that aggravates it. It is a problem of tissue that has failed to adapt to demand rather than a single dramatic injury.

That failure-to-adapt framing is exactly why loading is the leading rationale. Tendon is living tissue that responds to mechanical stress: applied gradually and progressively, load appears to stimulate the cells and matrix of the tendon to remodel and rebuild tolerance [1]. Over roughly the last two decades, isolated eccentric training became the dominant conservative strategy for Achilles and patellar tendinopathy on the strength of early trials — though, as we will see, "dominant" is not the same as "proven best," and a meaningful share of patients do not respond to it [1].

What the evidence shows, by tendon

Tendinopathy is not one condition, and the evidence is not equally strong across sites. Here is the honest picture for the four most-studied tendons.

Achilles (the heel-cord tendon). This is where loading has the deepest track record. An older systematic review of physical therapies found large benefits for eccentric loading in individual trials — for example, a standardised mean difference (SMD) of about −1.26 favouring eccentric exercise over a wait-and-see approach, and −1.67 versus cryotherapy — but cautioned that it could not statistically pool more than two studies for any single analysis, and that trial quality varied widely [2]. A more recent 2023 meta-analysis of eccentric exercise for mid-portion Achilles tendinopathy pooled eight randomised trials (371 patients) and reported a significant reduction in pain, with a mean difference of −1.21 (95% CI −2.72 to −0.30) [3]. That looks impressive, but the same review is candid about why to read it cautiously: statistical heterogeneity was very high (I² = 91%, meaning the trials disagreed with one another a great deal), every included study carried "some concerns" or a "high risk" of bias, and sample sizes were small [3]. The direction of effect is consistent and favourable; the precision behind it is not.

Patellar (the "jumper's knee" tendon). Loading — classically the eccentric decline-squat — is the mainstay here too, but the quantitative evidence is thinner than many assume. A systematic review comparing loading programmes across the Achilles and patellar tendons concluded that, for the patellar tendon specifically, the evidence that eccentric loading beats other loading approaches is conflicting rather than settled [1]. In short, strengthening is widely used and reasonable, but there are no robust, agreed effect-size estimates for it in patellar tendinopathy.

Lateral elbow (tennis elbow). A 2021 meta-analysis of six randomised trials (429 participants) found that adding eccentric exercise to another therapy improved pain more than that therapy alone (SMD −0.63, 95% CI −0.90 to −0.36) and improved grip/muscle strength (SMD 1.05, 95% CI 0.78 to 1.33) [4]. When eccentric exercise was compared directly against other forms of exercise, the extra pain benefit was small (SMD −0.30, 95% CI −0.58 to −0.02) and there was no significant difference in strength or function [4]. The authors were explicit that, given the small number of trials and inconsistent exercise protocols, this amounts to "limited evidence" [4].

Gluteal (lateral hip / "greater trochanteric" pain). Here the strongest single study is the LEAP randomised trial, which compared an education-plus-exercise (load-management) programme against a corticosteroid injection and against a wait-and-see approach in 204 people [5]. At eight weeks, 77% of the exercise-and-education group rated themselves at least "moderately better," versus 58% for injection and 29% for wait-and-see, and the exercise group reported less pain [5]. At one year, the exercise group still reported better global improvement than the injection group — but by then there was no significant difference in pain intensity between exercise and injection [5]. So loading looked meaningfully better than an injection or doing nothing in the short term, with the advantage narrowing over the longer haul.

How big is "better"? Reading the numbers honestly

Effect sizes like "SMD −0.6" are hard to feel. It helps to know that the usual textbook rules of thumb (Cohen's 0.2 / 0.5 / 0.8 for small/medium/large) appear to understate what counts as a meaningful change in tendinopathy trials. A 2023 meta-analysis that derived thresholds specifically from tendon-rehabilitation studies suggested that, for patient-reported pain, a "small" effect is closer to an SMD of 0.5, "medium" around 0.9, and "large" around 1.4 [6]. Read against that yardstick, the elbow pain benefit of adding eccentrics (−0.63) sits in the small-to-moderate range, while some of the Achilles figures reach into the large range — with the heavy caveats about heterogeneity and bias noted above. The same paper flagged that most tendinopathy trials are short (median eight weeks), even though tendon remodelling is thought to need longer, and that benefits tended to be larger with supervised programmes and when symptoms had been present for less time [6].

Eccentric vs heavy-slow resistance vs other loading: no clear winner

A natural next question is which loading protocol to do. On current evidence, no single protocol has been shown to be reliably superior. A head-to-head randomised trial in Achilles tendinopathy compared 12 weeks of eccentric training against heavy-slow resistance in 58 patients: both produced large, sustained improvements in pain and function, and — critically — none of those improvements differed between the two groups at 12 or 52 weeks [7]. The main differences were practical, not clinical: the heavy-slow group had better adherence (92% vs 78%) and somewhat higher satisfaction [7]. Broader reviews reach the same place: the evidence that eccentric loading outperforms other loading programmes is limited for the Achilles and conflicting for the patellar tendon, and up to 45% of patients may not respond to the eccentric protocol at all — which is why authors increasingly suggest that eccentric-concentric or heavy-slow loading are reasonable alternatives, not clearly inferior ones [1]. One 2023 meta-analysis did title its finding as eccentric being "more effective than other exercises" for the Achilles [3], but that conclusion sits against very high heterogeneity and universal risk-of-bias concerns in its own trials, and against the head-to-head and review evidence pointing to rough equivalence [1][7]. The honest reading is that the category — progressive loading — is what has support; the exact recipe is not settled.

Limitations worth taking seriously

Several features of this literature should keep everyone's confidence measured:

• Small, heterogeneous trials. Many studies enrol a few dozen participants, and they differ so much in protocol and population that pooling them produces wide, uncertain estimates — the Achilles meta-analysis reached I² = 91% [3], and other reviews could not pool more than two studies at a time [2].
• Risk of bias. You cannot blind someone to whether they are exercising, and much of this research carries "some concerns" or "high" risk-of-bias ratings, which tends to inflate apparent benefit [3][4].
• Protocol variation. "Loading" spans eccentric, heavy-slow, isometric, and combined programmes at different doses, making it hard to say precisely what to prescribe [1][4].
• Short follow-up and adherence. Trials are typically short relative to how long tendons take to adapt, and benefits depend on people actually doing the exercise over time [6][7].

What this means for you

Put together, the evidence supports a measured, encouraging summary rather than a promise. Progressive loading is the best-supported conservative approach across the main tendinopathies, the benefits on pain and function are real and range from modest to substantial depending on the tendon and the comparison, and — for the lateral hip at least — a structured programme outperformed a corticosteroid injection and doing nothing in the short term [1][2][3][4][5]. At the same time, the trials are often small and inconsistent, no single loading recipe has proven superior, a meaningful minority of people don't respond, and results hinge on sticking with the programme long enough for the tendon to adapt [1][3][6][7].

What the research averages cannot do is tell you which tendon is actually driving your pain, whether something else is masquerading as tendinopathy, how heavy to load, how fast to progress, and how to fit it around your work, sport, or other conditions. That is a job for a licensed, regulated health professional — a physiotherapist, physician, or comparable clinician — who can examine you directly, confirm the diagnosis, and build and adjust a loading programme for your situation. Use this summary to ask better questions; let a qualified professional individualise the plan.