Abstract
Background and Aims: Large clinical trials involving common chronic conditions may benefit from recruitment using an electronic health record (EHR). We developed a novel recruitment approach using our EHR to systematically identify all patients who appear eligible for such trials.
Methods: Patients satisfying broad inclusion and exclusion criteria, defined primarily by ICD and procedure codes, were identified from the EHR and invited by letter to call to learn more about the study. The letter also informed patients that those who did not respond would be contacted by telephone 2 weeks later. The EHR of patients expressing interest were reviewed (pre-screened), the study was described in detail, and those who still appeared to be eligible were invited to come in for a screening visit. Patients were interviewed and examined (screened) to assure that they satisfied all inclusion and exclusion criteria and remained interested. Eligible and interested subjects were enrolled. The disposition of each potential patient was tracked through each phase (response to letter and/or follow-up phone call, result of pre-screening, number enrolled).
Results: In STABILITY, a placebo-controlled trial of darapladib (a drug that lowers LP-PLA2) in patients with severe vascular disease, 5,300 patients satisfied broad inclusion/ exclusion criteria and 1,687 were sent letters inviting them to call and express their interest in possibly participating in the trial. Of the 1,687 who received letters, 587 (35%) expressed interest (either by calling or in response to our follow-up call) and were pre-screened. Of the patients who did not respond, 341 were unable to be reached. Study enrollment closed before we were able to contact 92 patients who expressed interest. Overall, 148 patients were screened, and we enrolled 101 patients, more than any of the 158 other participating institutions in the USA and more than were enrolled in 41 of 45 participating countries.
Conclusions: These data support the ability of an EHR to effectively identify large numbers of patients interested in and eligible to enroll in a research study. The ability to more accurately identify patients who also satisfy the more “minor” inclusion/exclusion criteria would improve the efficiency and automation of this approach.
- Received May 27, 2010.
- Accepted May 27, 2010.
