WMIP integrates services for aged, blind, and disabled

The pilot Medicaid Value Program that drew the most attention from the Center for Health Care Strategies and Mathematica was the Washington Medicaid Integration Partnership (WMIP), which brings together primary care, mental health, substance abuse, and long-term care services for categorically needy aged, blind, and disabled (ABD) Medicaid beneficiaries in Snohomish County, north of Seattle.

The program is operated by Molina Healthcare of Washington, a for-profit HMO focused on Medicaid and other vulnerable populations, under contract to the state Department of Social and Health Services (DSHS). State officials say the primary motivating factor underlying WMIP implementation was the disproportionate use of health care by these beneficiaries, who tend to have complex health profiles and are the fastest growing segment of the state agency's client base. Officials say ABD Medicaid clients in Washington are 15% of the total Medicaid caseload but account for 35%-40% of total fee-for-service expenditures.

According to state officials, before WMIP, ABD clients received substantial amounts of inappropriate care in emergency departments and hospitals due to lack of care management by physician and nursing facilities and because patients were not aware of or did not know how to access the care available for them.

Because the state agency was not sure it could integrate all the services at once, it chose to phase them in. Thus, WMIP clients were able to receive both primary and substance abuse care beginning in January 2005. Mental health care was added in October of that year and long-term care was added a year later in October 2006. Under WMIP, enrollees are eligible to receive all the same medical services that they would have received under fee-for-service Medicaid, except that Molina provides a central point for care coordination and management.

Molina's care coordination program includes health risk assessment, monitoring of patient symptoms, and education. Its care coordination teams coordinate home care, inpatient care, skilled nursing facility placement, long-term care, disease management, mental health care, substance abuse care, durable medical equipment, transportation, and day healthcare for patients in WMIP. The degree of patient contact varies from patient to patient depending on the patients' conditions. At a minimum, Molina staff contact the most stable patients once per quarter. Patients whose conditions are more fragile or require closer monitoring (some 30% of WMIP patients) are contacted at least twice per month and more often if needed.

Patient survey responses

The state agency measured claims-based outcomes and self-reported outcomes from surveys of enrollees and disenrollees, and compared results with a group of similar patients in other counties. Patient surveys identified reasons for WMIP enrollment or disenrollment and assessed patient satisfaction. Claims-based outcome measures included physician visits, inpatient admissions, emergency department use, and prescriptions filled. The agency also reported on the proportion of patients with mental health or substance abuse problems who used mental health and chemical dependency treatment and mental health hospital admissions.

Outcomes measures reported to the Center for Health Care Strategies in April 2007 suggested that WMIP appeared to have slowed the rate of inpatient admissions and mental health hospital days. But other measures were either flat or counter to expectation.

Results for mental health/substance abuse service utilization outcomes also were mixed. Mental health prescriptions filled rose slightly more in the intervention group than in the comparison group, suggesting to the evaluators that intervention group members were receiving prescriptions required to manage their behavioral issues, but at only a slightly better rate than the comparison group. Less encouraging to the evaluators, the proportion of patients with identified needs for alcohol or other drug treatment services who received these services rose at a slower rate in the intervention group compared with the comparison group. Molina staff reported that WMIP enrollees likely underreported substance abuse/chemical dependency issues, making it challenging to provide services to patients who failed to report a need for them.

Survey results indicated that WMIP improved client satisfaction with some aspects of care delivery (and reduced it for others) compared with a comparison group, and improved care coordination for many intervention group members. WMIP enrollees reported improved satisfaction with some aspects of care delivery, including wait times for routine care appointments, delays while waiting for health care approval, and problems with customer service or paperwork. However, enrollees were less satisfied with other aspects of care than their fee-for-service counterparts, including help when calling health care providers during regular office hours, help for urgent care right away, needed treatment or counseling for a personal or family problem, and prescription drugs. On average, WMIP enrollees who responded to the survey rated their health care and health plan lower than fee-for-service clients.

Significant disenrollment

Of the more than 5,000 members enrolled in WMIP in December 2004, nearly 2,000 chose to disenroll within the first month. Enrollment steadily fell to 2,180 active participants by June 2005 and 1,700 by March 2006, as patients lost Medicaid eligibility or moved out of the service area.

After the agency identified additional eligible patients in early 2006, enrollment rose to nearly 2,700 in June 2006 and remained steady through April 2007. Agency staff said the primary reason for enrollment stabilization was addition of a staff member to manually search for new or reconnected clients to be auto-enrolled into WMIP, a task the data system is unable to perform automatically.

WMIP conducted a disenrollee survey in Spring 2006 and found that more than half of those who left either lost Medicaid eligibility or moved away, while 37% opted out voluntarily. The primary reasons people opted out of WMIP included problems with access to providers and prescription drugs. Thus, 36% of those who opted out said their regular doctor was not affiliated with Molina Healthcare, 24% said they had to travel further to visit their Molina Healthcare physician, and 18% reported issues with the language spoken by their Molina physician. Some 30% said a family member or case worker influenced their decision to leave.

DSHS care coordinator Alice Lind tells State Health Watch preliminary cost-benefit data show some positive differences on measures such as avoiding emergency department use and inpatient admissions. While trends in both the intervention and comparison populations went up, she said, the increase was not as much for WMIP patients.

Ms. Lind says DSHS and Molina identified two general categories of implementation barriers they confronted as they put the program in place. First were internal barriers created at the state agency level such as claims being paid from different systems, eligibility information being kept in different systems, and systems not talking to each other and requiring manual work-arounds.

At the community level, concerns were raised because Molina is a for-profit health plan and there were questions about taking money from a not-for-profit system to go to a for-profit plan. There also were many stakeholders involved who wanted to be sure the changes would benefit clients.

Looking at program evaluation, Ms. Lind says it was challenging to find meaningful comparison groups. In addition, she says, while there is a lot of information available on quality measures, there isn't much that has been tried and tested. It's particularly difficult, she says, to develop quality measures for an integrated system rather than for some of the system's component parts.

She says she believes the program is worth replicating but cautions it needs committed executive-level leadership and legislative political will to bring it about. The Washington legislature has given its approval to doubling potential enrollment in WMIP and the state plans to expand it while trying to fix the problems that have cropped up.

Mathematica's evaluation is available online at www.chcs.org. Contact Ms. Lind at (360) 725-1629.