Intended for healthcare professionals

Letters Treatment effects in routine data and trials

Authors’ reply to Pérol and colleagues

BMJ 2016; 355 doi: (Published 16 December 2016) Cite this as: BMJ 2016;355:i6747
  1. Lars G Hemkens, senior researcher1 2,
  2. Despina G Contopoulos-Ioannidis, clinical associate professor3 4,
  3. John P A Ioannidis, professor1 4 5 6
  1. 1Stanford Prevention Research Center, Department of Medicine, Stanford University School of Medicine, Stanford, CA 94305, USA
  2. 2Basel Institute for Clinical Epidemiology and Biostatistics, University Hospital Basel, Basel, Switzerland
  3. 3Department of Pediatrics, Division of Infectious Diseases, Stanford University School of Medicine, Stanford, California, USA
  4. 4Meta-Research Innovation Center at Stanford (METRICS)
  5. 5Department of Health Research and Policy, Stanford University School of Medicine, Stanford, California, USA
  6. 6Department of Statistics, Stanford University School of Humanities and Sciences, Stanford, California, USA
  1. lars.hemkens{at}

We thank Pérol and colleagues for their reflections on our work on how routinely collected data might complement randomised controlled trials.1 2 We agree that routinely collected data may be used to analyse any number of associations for combinations of medical interventions, patient characteristics, settings, and concurrent treatments. They may indeed …

View Full Text

Log in

Log in through your institution


* For online subscription