Continuity of Care Lost In Transition to Carveout

A long-held tenet of treating those with severe psychiatric illness is complete support for and encouragement of a patient’s compliance with his or her medication regimen. A new study indicates that compliance may be directly threatened by the conversion of services to a managed mental health carveout.

In 1996 Tennessee’s Medicaid program, called TennCare, converted the provision of mental health services to a fully capitated, specialty carveout, TennCare Partners. A 1998 report characterized that conversion as one fraught with “serious disruptions of patients’ care” and attributed those problems to intrinsic flaws in new policies and the implementation of those policies. The program itself has since made administrative changes to correct certain problems.

Now, a new report, which appeared in the May 8 New England Journal of Medicine, says that TennCare Partner’s assumption of all mental health services resulted in a loss of continuity in physician care, a decreased duration of antipsychotic therapy, and an increase in the number of days during treatment when patients took no antipsychotic medication. The findings also indicate that the effects were most pronounced in patients who were more severely ill.

“When the system of preventive mental health for fragile patients is changed, care has to be taken to prevent disruption of services, such as community services, that are essential to help patients comply with their medication regimens,” pharmacoepidemiologist Wayne Ray, Ph.D., a professor of preventive medicine at Vanderbilt University, said in a press briefing. “In these patients, a loss of compliance can have severe consequences.”

Ray, along with a data analyst at Vanderbilt and Keith Meador, M.D., M.P.H., a clinical professor of psychiatry and behavioral sciences at Duke University School of Medicine, examined the records of 8,000 patients covered during the TennCare transition, most of whom had schizophrenia.

They studied antipsychotic therapy for one cohort of patients—the “transition cohort”—during a six-month period of care prior to the transition (which occurred July 1, 1996) to establish a baseline, then followed those same patients during the first year of care after the conversion. The team also studied a “pretransition cohort,” also for a total of 18 months, with the study period ending on the day before the transition.

During the baseline six-month study period of both groups, the transition group of 4,507 patients did not significantly differ in any way from the 3,644 patients in the pretransition group.

Compared with the pretransition group, patients in the transition group experienced “a reduction in the continuity of antipsychotic therapy” after the transition. The transition group had a statistically significantly higher percentage of patients who missed more than 60 days of antipsychotic therapy than patients in the pretransition cohort. In addition, they were prescribed fewer days of antipsychotic therapy overall and were less likely to see the same physician as they did before the transition.

The effect was most pronounced in patients termed as most seriously ill, which they defined as those patients who, during the six-month baseline period in each cohort, required depot antipsychotic medication to increase compliance or who were hospitalized with a diagnosis of psychosis during that six-month period. Ray and Meador termed these patients as having an “essential need for continuation of therapy.”

“Essential” patients in the transition cohort were 1.79 times more likely to miss 60 days or more of antipsychotic medication than essential patients in the pretransition cohort (29.4 percent versus 20.3 percent). Overall, patients in the transition cohort were 1.36 times as likely to miss at least 14 days of medication than patients in the pretransition cohort (41.8 percent versus 34.2 percent).

More strikingly, if patients had no risk factors that would classify them as “essential” for continuity of medication therapy, there was no statistically significant difference in the odds of missing 60 or more days of medication between the two cohorts. However, for patients with two risk factors (both the use of depot formulations and a hospitalization during the baseline period, for example), there was a highly statistically significant difference, with these patients being 1.75 times more likely to miss more than 60 days of antipsychotic medication.

Ray noted that for these high-risk patients, compliance with their medication regimen is crucial to their functioning. On medication, “they can maintain fairly normal lives. But if they stop, control over the disease may disappear,” he said.

“It was clear that carveout programs need to include incentives for increasing the quality of care, as well as containing costs. This was missing in the original Tennessee carveout,” Ray concluded.

An abstract of “Effect of a Mental Health ‘Carve-Out’ Program on the Continuity of Antipsychotic Therapy,” is posted on the Web at http://content.nejm.org/cgi/content/abstract/348/19/1885. ▪