Intended for healthcare professionals

Endgames Statistical Question

What are randomised consent designs?

BMJ 2014; 349 doi: (Published 25 July 2014) Cite this as: BMJ 2014;349:g4727
  1. Philip Sedgwick, reader in medical statistics and medical education1,
  2. Carwyn Hooper, senior lecturer in medical ethics and law1
  1. 1Institute for Medical and Biomedical Education, St George’s, University of London, London, UK
  1. p.sedgwick{at}

Researchers investigated whether an off-site transitional care facility provided a safe alternative to usual care for elderly patients in acute hospital beds awaiting transfer to a residential long term care bed.1 When elderly patients are admitted to hospital, they often need time to recover. Therefore, if the transitional care facility provided a safe alternative, hospital bed space could be freed. A randomised controlled trial study design that incorporated a randomised consent design was used. The intervention was transfer from an acute hospital bed to a transitional care facility to await a long term care bed, with patients receiving a single assessment from a specialist elderly care team and appropriate ongoing treatment. The control treatment was for the patient to remain in hospital with usual care provided.

Eligible patients were invited to participate in a four month follow-up study that involved routine observation of their healthcare. In total, 320 patients (mean age 83 years) agreed to participate. After recruitment, participants were randomised without their consent to treatment groups in a 2:1 ratio (212 to the intervention, 108 to the control). Once randomised, patients allocated to the intervention group were approached again for consent to transfer to the transitional care facility to await long term care. A total of 44 (20.8%) patients declined the intervention and received usual care instead. Those allocated to receive usual care were not approached again.

The main outcome measures included length of stay in hospital, rates of readmission, and death. Analysis was by the principle of intention to treat. It was reported that the intervention group spent less time than the control group in hospital (median 32.5 v 43.5 days, 95% confidence interval for difference 6 to 16). The intervention group took longer than the control group to be admitted to permanent care (median difference 21 …

View Full Text

Log in

Log in through your institution


* For online subscription