Intended for healthcare professionals


n of 1 trials comparing a non-steroidal anti-inflammatory drug with paracetamol in osteoarthritis

BMJ 1994; 309 doi: (Published 22 October 1994) Cite this as: BMJ 1994;309:1041
  1. L March,
  2. L Irwig,
  3. J Schwarz,
  4. J Simpson,
  5. C Chock,
  6. P Brooks
  1. Correspondence to: Dr March
  • Accepted 19 August 1994


OBJECTIVE: To evaluate the efficacy of paracetamol and a non-steroidal anti-inflammatory drug for symptom relief in osteoarthritis.

DESIGN: Double blind, randomised, controlled trials in individual patients (n of 1 trials). Three treatment cycles with two weeks' each of paracetamol (1 g twice daily) and diclofenac (50 mg twice daily) prepared in identical gelatin capsules.

SETTING: General practices in metropolitan Sydney, Australia.

SUBJECTS: 25 patients (median age 64 years) with pain of osteoarthritis (median duration of disease eight years) considered by their general practitioners to require regular treatment. 20 were already taking non-steroidal anti-inflammatory drugs.

MAIN OUTCOME MEASURES: Diary of pain and stiffness, function, and side effects.

RESULTS: 15 patients completed the study, five withdrew early but had made a therapeutic decision, and five dropped out very early. Results from 20 patients were analysed. Several patterns of response evolved. Eight of the 20 patients found no clear difference, symptoms being adequately controlled by paracetamol; five indicated a clear preference for the non-steroidal anti-inflammatory drug; two showed control of symptoms after their initial two weeks of the non-steroidal anti-inflammatory drug which continued throughout subsequent treatment changes; in five the non-steroidal anti-inflammatory drug may have been better but neither agent gave satisfactory control. After three months nine of the 20 patients had adequate symptom control with paracetamol alone.

CONCLUSIONS: Of 1 studies--that is, randomised trials in individual patients--are clinically useful in deciding treatment in heterogeneous conditions which require long term symptomatic relief. In osteoarthritis many patients currently receiving or being considered for non-steroidal anti-inflammatory drugs may achieve adequate control with paracetamol.


  • Accepted 19 August 1994