Computer-based simulation as a teaching tool for residents treating patients with cancer-related pain crises

Residents often fail to escalate narcotics to ensure pain relief in patients with cancer because of fear of overdose. A computer simulation of patients in pain may provide a way to improve management without fear of harm. We developed a pain care simulation to train residents. STUDY DESIGN/MEASURES: Thirty-one residents trained on 2 to 3 consecutive simulated patients. Simulated cases were assigned variable tolerances to narcotics and starting pain scores. The goals of training were as follows: (1) rapid induction of pain relief, (2) measurement of pain response at appropriate times, and (3) early institution and escalation in care long-acting pain medication to ensure stable pain control for 48 hours. Seven reviewers judged graphical summaries of care and assessed if pain improved, worsened, or stayed the same.

Thirty-one residents participated: 25 completed 3 simulations; 6 completed only 2. Sixty-eight percent improved from case 1 to 2; 90% improved by 3. The average pain score declined from 2.5 in case 1 to 1.9 in case 3 (P = .04). Rescue medication use declined from 37% for the first case to 23% by the third (P = .18). Reviewers' intraclass correlation for improved pain care was 0.821 (95% CI, 0.721-0.895).

Residents improved using a pain treatment simulator. A graphical representation of pain scores more than 48 hours of care that provides a useful way to assess pain control. Lessons learned may translate into improved patient care.