Article Text

other Versions

Extended report
Does flare trial design affect the effect size of non-steroidal anti-inflammatory drugs in symptomatic osteoarthritis? A systematic review and meta-analysis
  1. Toby O Smith1,
  2. Kun Zou2,
  3. Natasya Abdullah3,
  4. Xi Chen3,
  5. Sarah R Kingsbury4,
  6. Michael Doherty3,
  7. Weiya Zhang5,
  8. Philip G Conaghan4
  1. 1Faculty of Medicine and Health Sciences, University of East Anglia, Norwich, UK
  2. 2Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital, Affiliated Hospital of University of Electronic Science and Technology, Sichuan, China
  3. 3Division of Rheumatology, Orthopaedics and Dermatology, University of Nottingham, Nottingham, UK
  4. 4Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds & NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds, UK
  5. 5Faculty of Medicine & Health Sciences, University of Nottingham, Nottingham, UK
  1. Correspondence to Professor Philip G Conaghan, NIHR Leeds Musculoskeletal Biomedical Research Unit, Section of Musculoskeletal Disease, Chapel Allerton Hospital, Chapeltown Road, Leeds LS7 4SA, UK; p.conaghan{at}


Objectives It is thought that the clinical trial benefits of oral non-steroidal anti-inflammatory drugs (NSAIDs) may relate to flare designs. The aim of this study was to examine the difference in NSAID (including cyclooxygenase-2 (COX-2) inhibitors) response in osteoarthritis (OA) trials based on different designs.

Methods Systematic review was undertaken of the databases MEDLINE, EMBASE, AMED, CINAHL and the Cochrane library till February 2015. Randomised controlled trials assessing pain, function and/or stiffness following commencement of NSAIDs in flare and non-flare designs were eligible. Trials were assessed using the Cochrane Risk of Bias tool. Meta-analyses were conducted to assess the effect sizes (ES) of NSAIDs for OA with flare versus non-flare trial designs.

Results Fifty-seven studies including 33 263 participants assessing 26 NSAIDs were included. Twenty-two (39%) were flare design, 24 (42%) were non-flare designs, 11 (19%) were possible flare designs. On meta-analysis, there was no statistically significant difference in ES of NSAIDs versus placebo between flare and non-flare trial designs for absolute pain and function or stiffness at immediate-term (1 week), short-term (2–4 week) or longer-term (12–13 week) follow-up periods (p>0.05). However there was a lower ES for mean change in pain in flare and possible flare trials compared with non-flare trials at short-term follow-up (0.36 vs 0.69; p=0.05).

Conclusions Contrary to previous understanding, flare trial designs do not result in an increased treatment effect for NSAIDs in people with OA compared with non-flare design. Whether flare design influences other outcomes such as joint effusion remains unknown.

  • NSAIDs
  • Osteoarthritis
  • Health services research

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.