Use of Mental Health Services in Rural Areas

To the Editor: The effect that managed care will have on service access, utilization, and clinical outcome for Medicaid beneficiaries who live in rural areas is unknown. Before statewide implementation of Medicaid managed care in Iowa in March 1995, Medicaid recipients who had used fee-for-service mental health services in fiscal year 1993 were surveyed at the request of the Iowa Department of Human Services (1).

One purpose of the study was to determine if residents of rural counties had less access to the same types of services than their urban counterparts. ("Urban" was defined by the presence of a U.S. Census Bureau metropolitan statistical area within the county.) If rural services were less accessible, then services for vulnerable populations such as rural residents with serious and persistent mental illness would need to be protected or enhanced as the fiscal constraints of managed care began to influence the delivery system.

Out of a total population of 16,579 persons who met criteria for inclusion in the study, survey instruments were sent to 2,520 persons and were returned by 815, for a response rate of 32 percent. Responses were analyzed by the Cochran-Mantel-Haenszel chi square to test the study hypothesis that persons with residence in rural counties used fewer community-based and inpatient mental health services than persons living in urban counties.

The results showed that Medicaid beneficiaries living in a rural county did not report less overall use of mental health services than beneficiaries living in urban counties; however, they were less likely to use group therapy (22 percent versus 31 percent, c²=4.75, df=1, p=.03). Rural Medicaid recipients were more likely to report use of home services than urban residents (36 percent versus 27 percent, c²=5.64, df=1, p=.02).

Persons who did not have schizophrenia and who had been hospitalized in fiscal year 1993 were more likely to report a greater lifetime cumulative number of hospital days if they lived in urban versus rural areas (c²=14.10, df= 1, p<.001). Cumulative hospitalization of persons with schizophrenia, however, did not differ in rural and urban counties.

From these results, we conclude that rural residents may be less likely to use group therapy because it is less available in rural areas. The lower availability may be a reflection of fewer participants and fewer group facilitators, but it is also possible that rural residents use group therapy less because anonymity in participation is less assured.

One explanation for more frequent use of home services in rural areas is that they are a mechanism for providing outreach to persons who otherwise do not have access to services due to problems with transportation. It is also possible that home services provide a substitute for clinic- or hospital-based providers in rural areas.