A Comparative Cost Analysis of Participating versus Non-Participating Somatizing Patients Referred to a Behavioral Medicine Group in a Health Maintenance Organization

Aim To determine whether participation in a six-session behavioral medicine group program was associated with a post-intervention decrease in health costs among participants. Methods A retrospective study conducted in a convenience sample of 295 high utilizers of healthcare at a health maintenance organization in northeast USA. High utilizers were considered to be those patients with at least $US1500 in utilization costs (excluding eye care, dental services and pharmacy services) in the 12 months prior to the course. Five patients with $US20 000 or more in utilization costs for any 6-month period were excluded from the analysis because of the occurrence of severe acute medical illnesses. The intervention group comprised 114 patients who completed at least four of six behavioral medicine group sessions in the Personal Health Improvement Program (PHIP). The comparison group (n = 176) consisted of those patients who were referred but declined participation. Healthcare utilization for both groups was measured for two epochs: the 12 months before the referral and the 12 months following the referral for the control group, or the 12 months following completion of the program for the PHIP group. Results The PHIP course significantly decreased utilization from an average of $US4079 prior to course participation to an average of $US2462 in the 12-month period after the course, a decrease of $US1616 (p < 0.0001). Utilization in the comparison group decreased by $US608 (from $US4347 before referral to $US3739 12 months after referral). Post-intervention health costs were $US1008 less than those observed in the control group during the same time period. There was a mean decrease in costs from baseline of 25% for the PHIP group and less than 1% for the control group (p = 0.031, one-tailed). Conclusion This cost saving, if attributable to a direct impact of PHIP on morbidity and a subsequent reduction in healthcare utilization, would represent roughly a 25% saving in health costs. The study was limited by the non-random assignment to condition and the resulting potential for selection bias, as well as other possible confounds. However, the present finding of lower health costs after PHIP participation is consistent with earlier studies showing reductions in ambulatory visit rates following PHIP. Taken together, these findings suggest that the integration of behavioral medicine group programs into primary care will benefit patients and clinicians as well as help to control health costs.