MANAGEMENT OF MAJOR AFFECTIVE DISORDER

Working With Vendors to Improve Coordination of Behavioral Health Services

Management of patients with a major affective disorder is an issue for many health plans. Quality improvement can always be challenging, but when care is delegated to multiple vendors, even seemingly straightforward issues can become complex. This plan determined that there was considerable variation in the performance of its three behavioral health vendors and that their approaches to the management of the treatment of depression needed standardization.

Major affective disorders were responsible for almost half of all behavioral health admissions for this plan, and antidepressants were the most commonly prescribed class of medicine. Preliminary data indicated that the plan was not meeting its goals for posthospital ambulatory follow-up for these patients.

The plan decided to work closely with its contracted behavioral health delegates to improve the management of members hospitalized with a major affective disorder.

The plan targeted the population of all members, hospitalized for a major affective disorder.

The plan used three measures for this quality initiative:

• Rate of follow-up after hospitalization for major affective disorder within 30 days of discharge
• Rate of readmission within 30 days of discharge
• Rate of medication review with a psychiatrist within 60 days of discharge (for patients discharged on medications)

For the readmission measures, the plan redefined "readmission" as any unplanned and unexpected admission to the same, or higher, level of care within 30 days of discharge from a behavioral health inpatient facility. (If the member's treatment plan upon discharge dictated a readmission within 30 days, then it was not counted as part of the numerator in the rate calculation.)

For the medication review with a psychiatrist, the plan limited the eligible population to members discharged on psychotropic medications. Since members are given a maximum of a 60-day supply of medication on discharge, the plan's quality improvement committee set a 60-day time period for this measure.

STANDARDIZED PERFORMANCE MEASUREMENT Only one of these three measures - the 30-day ambulatory follow-up rate - was based on HEDIS 2.5 technical specifications. Use of standardized measures such as HEDIS can enhance comparability, allow the discovery of meaningful benchmarks, and indicate the need for stronger interventions.

The plan was able to find a benchmark for only one of its measures - ambulatory follow-up after hospitalization for major affective disorder. This benchmark of 75% for the state was selected from NCQA's 1996 Quality Compass.

The behavioral health vendors were capitated but submitted encounter data and quarterly data on readmissions to the plan. The plan analyzed the encounter data both quarterly and annually to track progress. Using 1996 data, the plan measured the baseline performance for all indicators, for each of the three vendors, and calculated a weighted average for the plan-wide rate. The baseline performances for all three measures are outlined in Table 1 on the following page.

The plan set three goals for this quality initiative. The goals were the same for all three vendors:

• To increase ambulatory follow-up rates to 95% in 1997 and to 100% in 1998
• To lower readmission rates to less than 6%
• To achieve a 95% rate of medication review with psychiatrists

The plan met with personnel from each of its behavioral health vendors to discuss the potential barriers to improvement:

• Lack of member awareness of transportation resources
• Poor coordination in transition from inpatient care
• Member nonadherence with discharge medication(s) and aftercare plans
• Psychosocial environmental issues such as the lack of a familial support system
• Coordination issues preventing the making of follow-up appointments while member was still hospitalized
• Poor access to psychiatric care in some areas

PARTNERING WITH VENDORS When dealing with contracted vendors, it is essential to get buy-in early in a quality improvement activity. Because this plan obtained buy-in from the vendors on its goals, it was able to get the vendors to commit resources for needed interventions. The vendors were also invaluable sources of information for barrier analysis and program improvements.

The plan established a behavioral health advisory committee, composed of representatives from plan staff and representatives from the three behavioral health vendors. To avoid past experiences with variation in vendor reporting, the behavioral health advisory committee first adopted good operational definitions - including procedure codes -for the numerators and denominators of each measure.

The vendors implemented interventions aimed at three issues:

• Policies/procedures
• Access
• Education

They revised policies and procedures for communicating with members' primary care physicians (PCPs) during hospitalization and for prescribing psychotropic medications. Case managers were required to ask members if transportation was a barrier to follow-up visits, and to offer solutions. Following hospitalization, the revised policies and procedures also required the case managers to specifically ask members about aftercare plans, to place appointment reminder calls to the members, and to require that behavioral health providers notify the vendors if a member did not keep an appointment (examples of admission, discharge, and follow-up tracking tools are included in QP Tool ).

In response to access barriers, the vendors added step-down units, partial hospital and intensive outpatient programs to their network. They also added psychiatrists and child and adolescent providers in certain areas. They increased psychiatric home-care services for members who had problems keeping scheduled office appointments.

The vendors taught behavioral health providers how to coordinate members' care with their PCPs. Behavioral health providers were shown how PCPs could help care for difficult-to-manage patients. The providers were trained about the importance of achieving member adherence with medication plans. The plan then trained case managers how to better manage the members' psychosocial environmental issues. Guidelines were distributed to the behavioral health network on the use of antidepressant medications.

The health plan concentrated on supporting the vendors' interventions and rethinking policy matters. The plan provided the vendors with quarterly data on the initiative.

The first remeasurement, using 1997 data, showed a statistically significant improvement in the plan-wide ambulatory follow-up rate after hospitalization for major affective disorder, as well as substantial improvements in the rates for two of the three vendors. Plan-wide success in lowering readmission rates was also statistically significant, and the plan's largest vendor had similar success. With two of the vendors showing improvement in the rate of medication review, and the third showing a substantial decrease, the plan-wide change for medication review was not statistically significant. See Table 2 on the following page for the results.

In July 1998, the plan required behavioral health providers to coordinate care with the PCPs, making authorization of further care contingent on feedback received by the health plan. At the vendor's request, the health plan also developed letters that the vendors could send to their practitioners. These letters defined the plan's expectations for cooperation in meeting goals, including communication with PCPs when a member was hospitalized or prescribed psychotropic medications.

The second remeasurement, based on 1998 data, showed slight improvement plan-wide in the 60-day medication review, a slight decrease in ambulatory follow-up, and a slight increase in readmission rates.

Although the second remeasurement is a bit disappointing, all measures were statistically better than baseline. This initiative has progressed with the health plan continuing to work together with its behavioral health vendors.

This activity has helped the plan understand issues related to continuity and coordination of care between general medical and behavioral health providers. They have established a track record of collaboration and cooperation. Great persistence is necessary in this sort of project for its continued good results.