Abstracts

This study analyzed the effect of VNS therapy on direct costs, indirect costs, and utilization of medical services in a large, staff model health maintenance organization. To our knowledge, this is the first cost-effectiveness study of VNS therapy conducted in the United States.
This retrospective analysis accessed databases to capture direct costs and utilization. Telephone interviews provided information on indirect costs. The140 patients were an unselected surgical series receiving VNS therapy [ge]10 months before the beginning of data collection. Typical monthly utilization and costs, direct and indirect, were calculated for each patient during the 12 months before implantation of the VNS therapy device. These pre-VNS statistics were compared with rates during the 10- to 13-month post-VNS follow-up period. Student[rsquo]s t-test was used to compare interval measurements and percentages and McNemar[rsquo]s chi-square for binary measures. Values of [italic][/italic][le]0.05 were considered statistically significant.
A single catastrophic case (motor vehicle crash) during the fourth quarter affected post-VNS costs of inpatient services, length of stay, and attendant hospital costs. Excluding the catastrophic case, overall total direct costs per month post-VNS were 89.0% of the pre-VNS monthly average (=0.25, NS). Including the catastrophic case, overall total direct costs post-VNS were 112.2% of pre-VNS monthly average (=0.630, NS). Total post-VNS outpatient visits were 290.4% ([lt]0.001) of pre-VNS rates during the first quarter and decreased to 70.9% ([lt]0.001) during the fourth quarter. Emergency department visits decreased from 57.2% of baseline in the first quarter to 36.1% ([lt]0.001) during the fourth quarter. Average monthly admissions post-VNS were 56.7% of the pre-VNS rate (=0.021) and length of stay was 68.2% (=0.330, NS). Average monthly hospital costs were 114.1% of the pre-VNS rate (=0.680, NS). Medication costs were unchanged. Assessing indirect costs, the number of days that work was not possible because of health-related concerns decreased from 3.67 to 1.04 days (=0.002). Average time spent caring for health problems decreased from 352.6 to 136.1 minutes per week ([lt]0.001).
During the follow-up period, statistically significant improvements were noted in both direct and indirect costs of care. Among reductions in direct costs were marked decreases in emergency room visits and the number of hospital admissions. Improvements with indirect costs were reflected in decreases in both time missed from work and time spent on health problems. These significant reductions support the cost-effectiveness of VNS therapy as a treatment for refractory epilepsy.
[Supported by: an unrestricted grant from Cyberonics.]