| Sign In to gain access to subscriptions and/or personal tools. |
Constructing common cohorts from trials with overlapping eligibility criteria: implications for comparing effect sizes between trialsWake Forest University School of Medicine, Winston-Salem, NC, USA
Wake Forest University School of Medicine, Winston-Salem, NC, USA
University of Pennsylvania, Philadelphia, PA, USA
The University of Tennessee Health Science Center, Memphis, TN, USA
Wake Forest University School of Medicine, Winston-Salem, NC, USA
Wake Forest University School of Medicine, Winston-Salem, NC, USA
The Miriam Hospital, Providence, RI, USA
Wake Forest University School of Medicine, Winston-Salem, NC, USA
Diabetes Epidemiology and Clinical Research Section, NIDDK, Phoenix, AZ, USA
St. Joseph's Health Care, London, ON, Canada
Ohio State University School of Medicine, Columbus, OH, USA
Emory University School of Medicine/Atlanta VAMC, Atlanta, GA, USA
Wake Forest University School of Medicine, Winston-Salem, NC, USA, mespelan{at}wfubmc.edu Background Comparing findings from separate trials is necessary to choose among treatment options, however differences among study cohorts may impede these comparisons. Purpose As a case study, to examine the overlap of study cohorts in two large randomized controlled clinical trials that assess interventions to reduce risk of major cardiovascular disease events in adults with type 2 diabetes in order to explore the feasibility of cross-trial comparisons Methods The Action for Health in Diabetes (Look AHEAD) and The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trials enrolled 5145 and 10,251 adults with type 2 diabetes, respectively. Look AHEAD assesses the efficacy of an intensive lifestyle intervention designed to produce weight loss; ACCORD tests pharmacological therapies for control of glycemia, hyperlipidemia, and hypertension. Incidence of major cardiovascular disease events is the primary outcome for both trials. A sample was constructed to include participants from each trial who appeared to meet eligibility criteria and be appropriate candidates for the other trial’s interventions. Demographic characteristics, health status, and outcomes of members and nonmembers of this constructed sample were compared. Results Nearly 80% of Look AHEAD participants were projected to be ineligible for ACCORD; ineligibility was primarily due to better glycemic control or no early history of cardiovascular disease. Approximately 30% of ACCORD participants were projected to be ineligible for Look AHEAD, often for reasons linked to poorer health. The characteristics of participants projected to be jointly eligible for both trials continued to reflect differences between trials according to factors likely linked to retention, adherence, and study outcomes. Limitations Accurate ascertainment of cross-trial eligibility was hampered by differences between protocols. Conclusions Despite several similarities, the Look AHEAD and ACCORD cohorts represent distinct populations. Even within the subsets of participants who appear to be eligible and appropriate candidates for trials of both modes of intervention, differences remained. Direct comparisons of results from separate trials of lifestyle and pharmacologic interventions are compromised by marked differences in enrolled cohorts. Clinical Trials 2009; 6: 416—429. http://ctj.sagepub.com
This version was published on October
1, 2009 Clinical Trials, Vol. 6, No. 5,
416-429 (2009) |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
Sign In to gain access to subscriptions and/or personal tools.
