HIV/AIDS TREATMENT AND QUALITY OF CARE

Measuring and Improving Compliance With Practice Guidelines

This initiative began in response to an appeal from state public health officials. In 1992, they predicted a 50% increase in the numbers of HIV/AIDS patients over the next three years. They asked health plans to develop programs to address the needs of this growing population. At the same time, the plan received a number of requests from physicians for information on the optimal care of these patients. Although the absolute numbers of plan members identified as having HIV were small (20 to 30), the plan recognized that HIV infection was an important health concern. It took the opportunity to improve the consistency and quality of treatment for the high-risk and high-cost HIV/AIDS population.

The targeted population consisted of all known HIV/AIDS members. Members were identified through clinical laboratory reports, pharmacy data, claims, and encounter data. Many of these patients were well known to the plan's case management staff. The plan chose a performance measure that looked at the percentage of compliance with the plan's HIV/AIDS clinical practice guidelines. Medical records audits were conducted annually, with reaudits every six months for poor performers.

Infectious disease specialists affiliated with the plan produced these clinical guidelines based on the medical literature. The initiative's steering committee developed an audit tool that was used for the chart review. To maintain confidentiality, qualified clinical staff performed the medical chart review.

Compliance with HIV/AIDS clinical guidelines was measured in six different areas:

• nature, frequency, results, and follow-up of various lab tests
• immunizations
• accurate disease reporting to the CDC, with appropriate patient confidentiality
• referrals to infectious disease, eye care, and dietary practitioners
• education and counseling
• discussion of advance directives

The measure used was the mean compliance with all of the clinical guidelines for all providers.

Baseline measurement, using 1992 data, revealed a baseline performance measurement of 42.7% for the mean compliance to the guidelines. No benchmark data were available for this measure. The objective of this initiative was to improve the quality of care provided to the high-risk HIV/AIDS population. The plan established a performance goal of 80% compliance with its HIV/AIDS clinical guidelines by 1997, and 90% compliance by the year 2000. The year 2000 goal was established to be consistent with a statewide effort.

The plan discovered a number of barriers that made guideline compliance difficult for physicians:

• frequent changes in treatment guidelines
• the lack of a dedicated case manager to help make needed resources available in the community
• the lack of uniform and consistent medical record tools to track compliance

The initiative was steered by a committee that consisted of plan quality improvement staff, a biostatistician, pharmacy staff, primary care physicians, and specialty care physicians. The interventions were rolled out over a two-year period, first to staff model centers, then to IPA physicians.

First, the plan developed an HIV/AIDS treatment protocol manual. This manual contained clinical guidelines, audit forms, listings of plan and community resources, and member-education materials/literature. It disseminated the manual to practitioners, and followed up with an extensive educational effort.

REVISION CONTROL Practice guidelines for diseases such as AIDS represent a moving target. To cope with the rapidly changing picture of management of HIV positive patients, the plan implemented a formal process of revision control. This ensured that guidelines were reviewed periodically and formally updated on an annual basis.

During 1993, the plan conducted annual audits of provider compliance with HIV/AIDS clinical guidelines and notified providers of the results (a sample letter and scoring sheet are in QP Tool ). Corrective action plans were requested of those scoring in the bottom quartile. These low-scorers were reaudited six months later.

Remeasurement, using 1993 data, demonstrated a compliance rate of 44.9%. This increase was not statistically significant.

Barriers and areas still in need of improvement were identified:

• laboratory workup still showed a low rate of compliance
• discussion of advance directives was still frequently not occurring
• coordination of care was often poor
• time frames for specialty referrals were unclear in the guidelines

The plan addressed these issues with a number of interventions:

• institution of case management protocols
• clarification of referral guidelines
• continuation of guideline dissemination with reinforcement

The plan added a provider education seminar. The session was taught by specialists in early intervention for HIV positive patients.

The second remeasurement, using 1994 data, resulted in a compliance rate of 50.6%. Once again, this improvement was not statistically significant.

A barrier analysis was repeated. Issues identified included:

• low referral rate to case management services
• frequent poor coordination of care
• a voluminous medical record with widely dispersed information, making case management and coordination more difficult

The plan continued its interventions. It published articles in newsletters informing both provider and members of the availability of case management services for HIV/AIDS patients (sample newsletters are included in QP Tool ). The plan also created and implemented a one-page "quick check" reference guide. This helped track information in one easily accessible location in the medical record (sample in QP Tool ). A phone consultation service was set for providers to enable easy access to HIV specialists.

Recognizing an unmet need for a significant portion of its membership, the plan created dietary brochures in Spanish.

CURBSIDE CONSULTS In addition to establishing a telephone consultation service, the plan encouraged providers to bring cases to the steering committee for peer review, with suggestions about care options. Cases were discussed and a consensus reached on the best treatment approach. Plan physicians found this to be very helpful.

The third remeasurement, using 1995 data, showed a statistically significant increase in compliance rates, to 64.1%. Seeing that the interventions were providing positive results, the plan continued its educational interventions for providers and patients. It updated its protocol manual, and included a more user-friendly listing of community resources.

The plan also recruited an in-house clinical "champion" - a local HIV/AIDS specialist who gave periodic input about the initiative and its procedures.

CONTINUOUS IMPROVEMENT Each year, the plan analyzed its results, identified barriers, and implemented new interventions. Despite the fact that no statistically significant improvement occurred initially, the plan was undeterred in its efforts. Its persistence and efforts at continually refining its interventions made this activity successful.

Remeasurement, using 1996 data, showed that the plan achieved a compliance rate of 70.9%. This did not represent a statistically significant improvement over the prior year, but did represent a significant improvement over results from 1992 to 1994.

The plan maintained its previous interventions. It updated its protocols, and sent information about compliance with specific immunization guidelines to all practitioners. Individualized letters were sent from the now in-house clinical champion to providers with low rates of compliance.

The plan also added an HIV/AIDS focus to a Health Fair. This consisted of a booth with member-education materials and program literature, provided in conjunction with community AIDS awareness activities.

The plan's final remeasurement, using 1997 data, showed a compliance rate of 76.3%. Although this was not a significant improvement over the 1996 results, it was a significant improvement from 1995.

The plan continues to pursue its 2000 goal of 90%. It has added to its list of interventions by targeting non-HIV/AIDS members with preventive education.

The plan also reported that, as a result of its efforts to understand why the rate of compliance was so low at the outset, barriers to care were noticed that had not been expected. For example, it was found that HIV/AIDS patients often used multiple providers. Communication between providers tended to be poor, resulting in fragmented care.

This activity is a fine example of a low volume, high risk clinical issue addressed comprehensively, with good analysis and a broad range of strong interventions.