Consecutive screening and enrollment in clinical trials: the way to representative patient samples?

BACKGROUND

Trials usually do little or nothing to ensure random samples of patients representative of the disease under investigation.

METHODS AND RESULTS

To evaluate the effects of consecutive screening and enrollment, we compared the demographic characteristics of patients from three interventional trials, applying consecutive screening with those from nine similar survival trials of risk-stratified patients with either myocardial infarct or congestive heart failure. Administrative information was also compared. Patients in the consecutive studies were 5-10 years older and more often female. Six of the nine conventional studies enrolling high-risk patients had a 1-year mortality of < or =10% compared with 22-28% in the consecutive studies. The conventional studies that accounted for screenings tended to screen four to six times more patients, but typically enrolled fewer of those screened: 4-7% versus 17-26% in the consecutive studies. The conventional studies excluded the majority of those eligible, up to 85% versus 35% in consecutive studies. The conventional studies enrolled less than 1 patient per center per month compared with two to four times as many in the consecutive studies.

CONCLUSIONS

Patients in conventional cardiovascular trials, not using consecutive screening are selected and not representative of the disease under investigation. They are younger, more often male, and have lower risk. This makes such trials less reliable and useful. It prolongs the length of a trial study and makes it more expensive. Consecutive screening and enrollment are feasible and offer a detailed description of the patient selection The consecutive principles contribute to representative patient samples and should be mandatory in future clinical trials.