Evaluating a palliative care intervention for veterans: challenges and lessons learned in a longitudinal study of patients with serious illness

Abstract

Context: Longitudinal studies examining care for seriously ill patients are needed to understand patients' experience of illness, evaluate interventions, and improve quality of care. Unfortunately, such studies face substantial methodological challenges.

Objectives: This article describes such challenges and the strategies used to overcome them in a successfully implemented palliative care intervention trial for veterans.

Methods: Veterans admitted with a physician-estimated moderate-to-high one-year mortality risk were enrolled and followed up to three years, until death or study completion. Study protocols, procedures, and process data were intermittently analyzed to identify and develop strategies to address issues affecting study enrollment and interview completion rates.

Results: Of 561 patients who were eligible, 400 (71%) enrolled in the study; 357 (87%) alive at the end of Month 1 completed interviews; and 254 (88%) alive at Month 6 completed interviews. Of the 208 patients who died during the study and had identified a caregiver, we were able to conduct an after-death interview with 154 (74%) caregivers. A variety of strategies, such as systematic tracking and check-in calls, minimizing respondent burden, and maintaining interviewer-respondent dyads over time, were used to maximize enrollment rates, data collection, and retention.

Conclusion: These data demonstrate that the use of diverse strategies and flexibility with regard to study protocols can result in successful recruitment, data collection, and retention of participants with serious illness. They thus show that longitudinal research can be successfully implemented with this population to evaluate interventions and examine patient experiences.