Pragmatic Clinical Trials Unit

Research menu

Best Practice in Design, Analysis and Reporting of Trials

Randomised controlled trials (RCTs) are the gold standard for comparing healthcare interventions.  However they require sound design, analysis, and reporting to ensure their results are accurate and robust. The PCTU is involved in a variety of research projects focussing on improving the design, analysis, and reporting of RCTs.

Specific topics of our work include:

Adjustment for covariates 
Adjusting for prognostic covariates in the analysis of RCTs can be beneficial, as it can increase statistical power and precision. Our work has focussed on determining when and how to do this. We have examined whether we need to adjust for stratification or minimisation factors (i.e. variables used to balance treatment groups during randomisation) in the analysis. We have also looked at the best way of adjusting for covariates in the analysis, focussing primarily on how to adjust for centre or site in multicentre trials, and how to adjust for continuous covariates. 

Outcome measures 
Choosing an appropriate outcome measure and ensuring it is measured or assessed in a valid way is an essential part of designing RCTs. An inappropriate outcome measure (e.g. something that is not relevant to patients) may mean RCT results are not useful. Poor measurement or assessment of outcomes may introduce bias to the trial results. Our work has focussed on choosing appropriate primary outcome measures, as well as how to reduce bias in the assessment of the outcome in openlabel (unblinded) trials where blinded outcome assessment is not feasible. 

Accurate reporting is essential in RCTs so that others are able to understand the trial’s design, conduct and analysis, which allows them to assess the validity of the trial results. Incomplete or inaccurate reporting can hinder this process. We have developed reporting standards and guidelines for embedded recruitment trials (studies within a trial which compare different methods of patient recruitment) and for implementation studies of complex interventions. We have also assessed the reporting quality surrounding outcome assessment, method of randomisation, the analysis of RCTs using stratified randomisation, and the analysis of multicentre RCTs.

References to our published papers :

  • Froud R, Ellard D, Patel S, Eldridge S, Underwood M. Primary outcome measure use in back pain trials may need radical reassessment. BMC Musculoskelet Disord. 2015;16:88.
  • Kahan BC, Cro S, Dore CJ, Bratton DJ, Rehal S, Maskell NA, et al. Reducing bias in open-label trials where blinded outcome assessment is not feasible: strategies from two randomised trials. Trials. 2014;15:456.
  • Kahan BC, Rehal S, Cro S. Blinded Outcome Assessment Was Infrequently Used and Poorly Reported in Open Trials. PloS one. 2015;10(6):e0131926.
  • Kahan BC, Jairath V, Dore CJ, Morris TP. The risks and rewards of covariate adjustment in randomized trials: an assessment of 12 outcomes from 8 studies. Trials. 2014;15:139.
  • Kahan BC, Morris TP. Adjusting for multiple prognostic factors in the analysis of randomised trials. BMC medical research methodology. 2013;13:99.
  • Kahan BC, Morris TP. Reporting and analysis of trials using stratified randomisation in leading medical journals: review and reanalysis. BMJ. 2012;345:e5840.
  • Kahan BC, Morris TP. Improper analysis of trials randomised using stratified blocks or minimisation. Statistics in medicine. 2012;31(4):328-40.
  • Kahan BC, Rushton H, Morris TP, Daniel RM. A comparison of methods to adjust for continuous covariates in the analysis of randomised trials. BMC medical research methodology. 2016;16:42.
  • Peacock JL, Sauzet O, Ewings SM, Kerry SM. Dichotomising continuous data while retaining statistical power using a distributional approach. Statistics in medicine. 2012;31(26):3089-103.
  • Kahan BC. Accounting for centre-effects in multicentre trials with a binary outcome - when, why, and how? BMC medical research methodology. 2014;14:20.
  • Kahan BC, Harhay MO. Many multicenter trials had few events per center, requiring analysis via random-effects models or GEEs. Journal of clinical epidemiology. 2015.
  • Kahan BC, Morris TP. Analysis of multicentre trials with continuous outcomes: when and how should we account for centre effects? Statistics in medicine. 2013;32(7):1136-49.
  • Kahan BC, Morris TP. Assessing potential sources of clustering in individually randomised trials. BMC medical research methodology. 2013;13:58.
  • Morris TP, Kahan BC, White IR. Choosing sensitivity analyses for randomised trials: principles. BMC medical research methodology. 2014;14:11.
  • Madurasinghe VW, Sandra Eldridge on behalf of MRCSG, Gordon Forbes on behalf of the SECG. Guidelines for reporting embedded recruitment trials. Trials. 2016;17(1):27.
  • Pinnock H, Epiphaniou E, Sheikh A, Griffiths C, Eldridge S, Craig P, et al. Developing standards for reporting implementation studies of complex interventions (StaRI): a systematic review and e-Delphi. Implement Sci. 2015;10:42.
  • Kahan BC, Rehal S, Cro S. Risk of selection bias in randomised trials. Trials. 2015;16(1):405.
  • Rick J, Graffy J, Knapp P, Small N, Collier DJ, Eldridge S, Kennedy A, Salisbury C, Treweek S, Torgerson D, Wallace P, Madurasinghe V, Hughes-Morley A, Bower P. Systematic techniques for assisting recruitment to trials (START): study protocol for embedded, randomized controlled trials. Trials. 2014;15(1):407.
  • Froud R, Underwood M & Eldridge S. Improving the reporting and interpretation of clinical trial outcomes. Br J Gen Pract 2012;62(603):e729-731.
  • Hooper R, Diaz-Ordaz K, Takeda A & Khan K. Comparing diagnostic tests: trials in people with discordant test results. Stat Med 2013;32(14):2443-2456.
  • Hooper R, Froud RJ, Bremner SA, Perera R, Eldridge S. Cascade diagrams for depicting complex interventions in randomised trials. BMJ 2013;347:f6681
  • Kahan BC. Bias in randomised factorial trials. Stat Med 2013;32(26):4540-4549.
  • Choudhury Y, Hussain I, Parsons S, Rahman A, Eldridge S & Underwood M. Methodological challenges and approaches to improving response rates in population surveys in areas of extreme deprivation. Prim Health Care Res Dev 2012;13(3):211-218.
  • Froud R, Eldridge S, Kovacs F, Breen A, Bolton J, Dunn K, Fritz J, Keller A, Kent P, Lauridsen HH, Ostelo R, Pincus T, van Tulder M, Vogel S & Underwood M. Reporting outcomes of back pain trials: a modified Delphi study. Eur J Pain 2011;15(10):1068-1074.
  • Nadeem NJ, Taylor SJ & Eldridge SM. Withdrawal of inhaled corticosteroids in individuals with COPD - a systematic review and comment on trial methodology. Respir Res 2011;12:107.
  • Eldridge S. Pragmatic trials in primary health care: what, when and how? Fam Pract 2010;27(6):591- 592.
  • Lancaster GA, Campbell MJ, Eldridge S, Farrin A, Marchant M, Muller S, Perera R, Peters TJ, Prevost AT & Rait G. Trials in primary care: statistical issues in the design, conduct and evaluation of complex interventions. Stat Methods Med Res 2010;19(4):349-377.




Return to top