Summary

Doctors at an academic medical center referred 10 times as many patients to a clinical trial, and enrolled twice as many, when they were alerted by means of a computerized medical record system that their patients might be eligible to take part in the trial.

Source

Archives of Internal Medicine, October 24, 2005 (see the journal abstract).

Background

Advances in treatment and care for patients with cancer and other serious diseases come about as a result of findings from clinical trials. However, the number of patients who participate in clinical trials is very small. For example, only about three percent of adult patients with cancer take part in clinical trials. It often takes longer than planned to complete a trial because of difficulty recruiting enough patients in a timely fashion.

Previous studies have shown that patients are more likely to enroll in a clinical trial if their doctor recommends it. Few doctors, however, offer their patients the opportunity to take part in a trial. Doctors often cite limited time, lack of support staff, lack of awareness of trials, and complex enrollment procedures as reasons for not recruiting patients for trials.

The Study

Researchers at the Cleveland Clinic in Cleveland, Ohio, devised a system that used the clinic’s computerized medical record system to alert doctors to the existence of a clinical trial that some of their patients might be eligible for.

The alert popped up whenever doctors accessed the computerized medical records of patients with characteristics that matched criteria related to a specific clinical trial underway at the Cleveland Clinic. The doctor could ignore the alert or respond to it in one of three ways: refer the patient to the trial (with the patient’s permission), indicate that the patient wasn’t eligible, or indicate that the patient wasn’t interested in taking part in the trial.

When the doctor referred a patient, the system sent a secure message to the person in charge of recruitment for the trial. The message included confirmation that the patient had given his or her permission for the trial coordinator to view the patient’s electronic chart, and also provided a link to the chart. The trial coordinator telephoned all referred patients and invited those who were potentially eligible to come in for further screening.

The researchers compared doctors’ recruitment patterns during a one-year period before the computerized alert system was introduced and a four-month period after it was introduced. During the initial one-year period, doctors used electronic medical charts but were only informed about the trial by the usual means: memos, posted flyers, and discussions at meetings.

The study’s lead author is Peter J. Embi, M.D., M.S., who is now at the University of Cincinnati College of Medicine.

Results

After the introduction of the computerized alert system, doctors referred an average of 59.5 patients per month to the clinical trial. This was up from 5.7 patients per month before the computerized system was introduced - a 10-fold increase. The number of patients who enrolled in the trial more than doubled from an average of 2.9 per month before the electronic alert system was in place to 6.0 per month after.

Limitations

This study looked at how well a computerized clinical trial alert system performed at recruiting patients for a single clinical trial at one medical center. The study authors caution that their findings could be due to features unique to that trial or that medical center or to specific features of the computerized medical record system involved. The trial that was the focus of the study involved patients with type 2 diabetes, not cancer.

Comments

The findings from this study serve “as an invitation to incorporate the evolving use of technology in the clinical practice setting to enhance all phases of the clinical research process,” wrote Al B. Benson III, M.D., of Northwestern University in Chicago, Ill., in an accompanying editorial. “Because important obstacles to the successful maintenance of a given clinical trials program include recruitment strategies, costs, and time commitments, the integration of an institution’s electronic records system with the research record could help to reduce some of the barriers to the accrual of patients to clinical trials.”

Medical oncologist Ted Trimble, M.D., of the National Cancer Institute’s Cancer Therapy Evaluation Program, agrees that the study has merit, saying, “This study shows that reminding doctors that a patient could be eligible for a clinical trial had a strong impact on referrals,” adding that the findings also suggest there is scope for using computerized medical record systems to improve patient recruitment into cancer clinical trials.