Hip pain is common, particularly among adults older than 50 years, and intuitively linked to osteoarthritis. Osteoarthritis refers to a heterogeneous group of joint disorders characterised by progressive destruction of cartilage. Patients present with activity related joint pain and stiffness.1 2 Osteoarthritis can be classified as primary or secondary, but in practice and research the diagnosis is more often categorised as radiographic osteoarthritis, self reported osteoarthritis, or clinical (symptomatic) osteoarthritis. These designations have substantially different prevalences and incidences.2 3 4 Since osteoarthritis is a leading cause of pain and disability, we clearly need a better understanding of its epidemiology, and we need better methods for identifying patients at risk of the debilitating consequences of progressive limitations to daily life caused by symptomatic disease.1 2 5 6

In a linked paper, Kim and colleagues (doi:10.1136/bmj.h5983) address these issues in a cross sectional study of the association between self reported hip pain and radiographic findings suggestive of hip osteoarthritis.7 In two large population based cohorts, only a minority of patients with hip pain had radiographic hip osteoarthritis, and only a minority of patients with radiographic hip osteoarthritis had frequent hip pain. The two outcomes were substantially discordant.

This study is the largest to date assessing the association between hip pain and radiographic hip osteoarthritis and draws data from large representative samples. Limitations include the lack either of an objective gold standard—who had “true” (objectified) osteoarthritis and who did not—or of longitudinal follow-up, the lack of any measure of severity or disability due to hip pain, and the absence of data on the subsequent treatment. The study also underestimates the prevalence of radiographic osteoarthritis among adults presenting with hip pain, since adults who did not seek medical attention for hip pain were included.8 9 Nevertheless, the study teaches us an important lesson: many patients with radiographic hip osteoarthritis do not have frequent hip pain, and many patients with frequent self reported groin or anterior hip pain (traditionally considered suspicious of osteoarthritis) do not have radiographic hip osteoarthritis.

Which of these patients should receive a diagnosis of hip osteoarthritis? Generally, a diagnosis requires both symptoms of hip pain consistent with osteoarthritis and radiographic signs consistent with osteoarthritis. Patients with just one or the other are probably better described as having “clinical symptoms suggestive of degenerative hip joint disease” or “asymptomatic degenerative changes.” Currently, a single reliable definition is lacking, which complicates both treatment and research.3 4 9

Given the substantial discordance between symptoms and radiographs, how should we respond to patients presenting with hip pain? In our opinion, hip radiographs should be obtained when a patient reports pain that cannot readily be explained by alternative diagnoses such as trochanteric bursitis, iliotibial band syndrome, or referred pain. Radiographs are crucial to help rule out more serious conditions such as osteonecrosis, (impending) stress fractures, transient osteoporosis, primary neoplasms, or metastatic bone disease. When all are absent the value of hip radiographs in diagnosing hip pain as a symptom of osteoarthritis is difficult to establish, as Kim and colleagues demonstrate.

Moreover, is it essential to add the diagnosis “hip joint osteoarthritis” to the diagnosis “hip pain” (or to “clinical symptoms suggestive of degenerative hip joint disease”)? Of course, for both patients and doctors a “true” diagnosis makes a condition easier to accept and more manageable. Patients can be informed what to expect, and doctors can adhere to treatment protocols. But we must also consider whether the “diagnosis” will make any difference to the treatment. Patients with hip pain may benefit from lifestyle interventions, exercise programmes, or short term drug treatment, whether or not they have osteoarthritis, provided other causes have been ruled out. We currently have no effective method for slowing down the progression of hip osteoarthritis, so treatment regimens would be similar.1 2 10 11 Furthermore, since pathogenic phenotypes of osteoarthritis remain ill defined, the diagnosis “osteoarthritis” may be too general and lead to insufficiently targeted, suboptimal treatment.2

In summary, when patients seek help for hip pain that on clinical examination cannot readily be explained by alternative diagnoses, it is reasonable to obtain hip radiographs to exclude other sources of hip (joint) pain. When no abnormalities other than osteoarthritis are present, conservative management can be initiated and kept under regular review. Although it is important to give conservative management a fair trial, it is equally important to intervene with more invasive treatments (including joint replacement if indicated) in a timely manner if pain and disability progress. Not least because preoperative function is strongly associated with postoperative function.12 Similarly, when radiographic osteoarthritis is diagnosed in patients with few (or no) symptoms, lifestyle advice and a wait and see policy with optional follow-up is preferred. Ultimately, we must always remember to treat patients, not radiographs.