DIABETIC RETINAL EXAMS

Upfront Barrier Analysis Leads to Strong Interventions

Diabetes is the leading cause of blindness in adults. Studies have shown that diabetes-related blindness can be decreased with early detection and treatment. As a result, the American Diabetes Association (ADA) recommends annual dilated retinal exams for diabetic patients. Studies suggesting that only half of diabetic patients receive the recommended annual eye exams prompted the development of a HEDIS 2.0 measure, which looked at the percentage of adult diabetic patients who receive an annual dilated retinal exam.

This plan was one of 21 participants in the 1994 NCQA Report Card Pilot Project. When the results of the collaborative effort were published, the plan discovered that the percentage of its diabetic patients receiving annual retinal exams was below average. With this data in hand, the plan initiated a quality improvement (QI) activity to better manage its 15,000 members who had been diagnosed with diabetes.

HEDIS 2.5 (and 3.0) methodology was used throughout this activity. Pharmacy and encounter data were used to target the population of all diabetic patients over age 30. The hybrid method (a combination of chart review and administrative data) was used to calculate the retinal exam rate of a representative sample of this population.

Baseline data from 1995 revealed that only 42.8% of the plan's diabetic patients were receiving the recommended annual exam. The plan selected 47% as it benchmark, the grand mean from the NCQA Report Card Pilot Project. Attempting to surpass the benchmark, the plan established a performance goal of 50%.

Before implementing any intervention, the plan sought to identify potential barriers that might discourage its members with diabetes from receiving the recommended annual retinal exams. The plan used surveys and focus groups to get input from members and providers.

BARRIER ANALYSIS This plan conducted a thorough analysis through the use of focus groups and surveys, and conducted its analysis before taking action. One of the secrets of a successful QI activity is this type of effective analysis of the barriers to improvement.

In March 1996, the plan completed a simple telephone survey of a random sample of members with diabetes. (The survey tool used by the interviewer is in QP Tool .) About a third of members surveyed were unaware of the need for an annual eye exam. Others said they were too busy, or simply hadn't yet scheduled an exam. Many members said that they hadn't gotten an annual exam because their benefits didn't cover one. This misunderstanding of coverage turned out to be a common, but previously unsuspected, reason for missing the exam.

At about the same time, the plan convened a focus group of participating physicians from the specialties of family medicine, internal medicine, and ophthalmology. (The discussion guide used by this focus group is in QP Tool .) This focus group revealed that physicians were aware of the recommendation for annual eye examinations in patients with diabetes and knew that is was a covered benefit, but they did not believe that members were knowledgeable about these issues. This belief was reinforced by the results of the member survey.

The primary care physicians (PCPs) in the group suggested that education about the need for annual eye examinations might be omitted or lost in the larger context of the overall diabetic education and care that they typically provided. Ophthalmologists acknowledged that they often did not use recall systems to encourage patients to return yearly. The physicians mentioned a number of ways the plan could help increase the rate of annual retinal exams:

• provide lists of noncompliant members with diabetes to each PCP
• provide members with diabetes direct education regarding both the need for annual examinations and the available benefit coverage, especially since self-referral to eye care was permitted
• provide education to physicians via newsletters: the group felt that PCP education should focus on ensuring initial referral to an eye care specialist, while eye care specialist education should focus on encouraging patients to return annually and using a recall system such as a postcard.

Looking for missed opportunities, the plan reviewed claims data to identify if members with diabetes were visiting eye care specialists, but not receiving comprehensive care (including a dilated retinal exam). The results revealed that 21% of the diabetes patients seen by ophthalmologists and 5.5% of those seen by optometrists did not receive comprehensive services. It was unclear whether these findings represented incomplete coding or inadequate care. Subsequent analysis indicated that improper coding was responsible for the majority of these "missed opportunities" to render comprehensive care.

Finally, the plan identified two important internal factors that could discourage compliance with annual screening requirements:

• member services representatives did not understand the benefit coverage themselves, and provided members with inaccurate information
• an improperly configured claims system was reimbursing eye care services inaccurately

The plan began to address the findings of the barrier analysis with a mailing to all PCPs. It informed the PCPs of the baseline measurement and the initiative at hand. All PCPs received a sample copy of a proposed educational brochure for their diabetic patients. This mailing included lists of the PCP's patients identified as having diabetes by the claims system, and information on each patient's most recent eye examination. Physicians were asked to review the lists to verify the diagnosis of diabetes, and to exclude members that would be inappropriate for intervention (such as those who were already blind). The majority of the physicians took time to review the lists; 79% of patient names were checked through this process.

This plan used newsletters, bulletins and subcommittee meetings to communicate the findings from the focus groups, as well as the appropriate coding methods for eye exams.

The plan then mailed a letter and brochure to all physician-confirmed diabetic patients. (The brochure is pictured in QP Tool .) This effort was designed to increase awareness of the need for annual eye examinations and to clearly communicate available benefits.

A reply postcard was included in the mailing. Of the more than 1,300 members who returned cards, approximately 500 indicated that the reminder had resulted in an eye care appointment. Many indicated that they had not previously known that diabetic eye exams were covered on an annual basis.

The plan also implemented internal quality improvements. Member services representatives were trained to communicate that annual eye exams for diabetic patients were a covered benefit, and did not depend on a separate vision rider.

The first remeasurement, based on 1996 data, documented that the plan had exceeded its goal with a rate of 54.2%. The positive results of the interventions led the plan to implement a comprehensive disease management program for diabetes. Working together with the disease management arm of a major pharmaceutical company, the plan expanded its member and provider education program. As part of this program, the plan collaborated with a physician advisory group to develop a clinical practice guideline and a diabetic patient profile. The profiles provided each PCP with a list of members with diabetes, and the most recent date for key preventive services, including dilated retinal examination. These profiles were sent semi-annually, starting in March 1997.

CLINICIAN INVOLVEMENT The most successful QI efforts endeavor to involve clinicians early and often. This plan not only involved clinicians in the barrier analysis, but also enlisted their support in identifying the appropriate population for intervention, and in designing reports that were useful in improving performance.

Physician education was accomplished with both seminars and newsletter articles. Member education included diabetes education classes, newsletter articles, and a mailing to 15,000 diabetic members announcing the diabetes disease management program. The mailing once again highlighted the importance of annual eye examinations.

Nurse case managers personally contacted more than 500 diabetes patients in November 1997. These members were encouraged to undergo annual eye exams. Noncompliant members were reminded every six months.

The second remeasurement, based on 1997 data, showed a 58% rate of annual retinal exams for diabetic patients. This represented a statistically significant improvement over the prior year's performance. The plan had surpassed its goal and sustained its improvement.

The health plan continued its disease management program throughout 1998. Recognizing that repetition is essential to member education, it has continued to send regular reminders to its diabetic members. The 1998 results indicated further improvement, with the compliance rate increasing to 61.8%.

Seemingly simple problems can sometimes by the most difficult to resolve. Despite efforts at focused education, some plan members have remained confused about the coverage for annual diabetic eye exams.

The program has received a uniformly positive response from plan members. Thousands of members provided feedback. Many specifically commented on the value of the annual reminders. No negative comments were received.

The plan intends to continue this effective and popular initiative as one part of its evolving comprehensive diabetic disease management program.