Focus on Cardiovascular Disease

Improving the Management of Risk Factors for CVD*

Although conducted as separate quality improvement activities (QIAs), one health plan’s efforts to manage two of the most important risk factors for CVD (dyslipidemia and hypertension) - as profiled on these two pages - are an initial step in the direction of addressing multiple risk factor management. The health plan began its efforts to manage these risk factors in 1997 and completed the QIAs in 2001. Patients in the commercial health maintenance organization (HMO) were included in both QIAs, and there were no exclusions for coronary artery disease (CAD) or hypertension in either initiative. Thus, it is possible that there was some overlap in the patient populations. Most importantly, the timing of and similarities between the two initiatives indicates that - as early as 1997 - the health plan recognized that CVD is a multifactorial disease.

Responsible for more than one out of every five deaths, CAD, also known as coronary heart disease (CHD), is the leading cause of mortality in the United States. Despite the availability of clinical guidelines and the substantial risk of another heart attack, stroke, or other serious complications, some patients with CAD are still not receiving optimal care.[11]

The health plan felt that CAD was a priority, because approximately 1,400 of its members had CAD, and it was the number one inpatient diagnosis for men >40 years of age and women >65 years of age. It chose CAD for a disease management program based on its belief that the program would impact a significant number of its members, and that its management would proactively prevent complications and improve member quality of life.

The health plan took a unique approach to its QIA by using pilot studies to confirm the opportunity for improvement, identify barriers to optimal care, and fine-tune its initiative. First, a pilot study involving two clinics with a total of 153 CAD patients was conducted in June-July 1997. Results confirmed the need for better management of low-density lipoprotein (LDL) levels, revealing that <30% of CAD patients at both clinics met the LDL goal of <100 mg/dL. A plan-wide analysis of LDL levels performed in January 1998 for CAD patients produced comparable findings.

Getting Clinicians On-Board This health plan found that its clinicians were much more likely to enthusiastically participate in a program if they knew that their efforts would make a difference. By using pilot studies to investigate root cause issues, best interventions, and needed processes, the health plan was able to assure the clinicians that it had “done its homework.” After getting buy-in from a clinic, the health plan could work with the clinic’s staff to implement the same or similar interventions. In this way, the health plan could avoid wasting the time and efforts of busy practices.

For its QIA, the health plan used claims data to identify members with CAD who had been continuously enrolled for the reporting period. No age criteria were set. Two or more different dates of service with a CAD-related International Classification of Diseases, Ninth Revision (ICD-9) diagnosis code and/or at least one CAD-related ICD-9 Diagnosis/Procedure code or Current Procedural Terminology (CPT) code was considered evidence of CAD in claims. Primary care physicians (PCPs) were then contacted to verify the CAD diagnosis.

To assess the impact of its interventions, the health plan used administrative data (claims, automated laboratory data, and medical records) to measure control of LDL levels to <100 mg/dL and testing for LDL levels (within the least year) on a monthly basis. The health plan developed a registry to capture information from all administrative sources (see Implementing the Initiative). The registry also allowed clinics to add laboratory results from LDL tests completed and paid for by other means. Its accuracy was verified by the health plan through a medical record audit of a randomly chosen sample of entries made by the clinics.

Based on control rates seen in 1997%, the health plan felt that a goal of 40% was realistic for LDL control, and if achieved at the first remeasurement, would represent a statistically significant result. The goal for LDL testing was set at 100% because the health plan felt that all patients with CAD should have LDL levels tested at least once yearly (Table 1).

Baseline data for 1,061 patients was obtained for the time period of January 1, 1998, to December 31, 1998. With only 32.5% of members having achieved LDL levels of <100 mg/dL (7.5 percentage points lower than the goal of 40%), these baseline data confirmed the need for improved LDL management.

To identify barriers to optimal lipid management, the health plan went back to the pilot study data and performed a root cause analysis. The following barriers were identified:

• When LDL levels were borderline, physicians felt that a change in treatment was not warranted
• LDL could not be calculated because there was no high-density lipoprotein (HDL) measured and/or the triglycerides level was too high to calculate the LDL
• The patient was not complying with diet/medication regimens
• Cholesterol profile was incomplete
• The patient was being followed by a cardiologist, so the PCP was either unaware of the patient’s LDL status/medication regimen and/or both physicians thought the other was treating
• The patient was unable to tolerate the medications

The health plan concluded that better tracking and coordination of care would be required. This conclusion was validated when the health plan’s quality staff attended conferences on disease management programs and analyzed other health plans’ experiences.

Another Plan's Approach Identifying Barriers via a Multidimensional Model There are many ways to identify barriers when initiating a QIA. While the QIA profiled here used a root cause analysis of pilot study data, another health plan chose to conduct its barrier analysis using a multidimensional model of health improvement. The multidimensional model is based on the concept that, in managed care, clinical outcomes are dependent on the influences of several key stakeholders: members, providers, employers, the community, and the health plan itself.12 Thus, to increase the chance of success, representatives from each stakeholder category should be consulted with planning a QIA. This model is particularly useful since it is evidence based and relies heavily on the results of literature searches and rigorous data analysis. Because of this, the chances for success are greater than those using a unidimensional approach. Reprinted with permission. Tipton ML, Fleming M. Stroke prevention in managed care: a five-dimensional health improvement model. In: Gorelick PB, Alter M, eds. The Prevention of Stroke. New York, NY: The Parthenon Publishing Group; 2002;123-130.

To tailor interventions to specific patient needs, the health plan developed a registry, which was prepopulated with electronically downloaded laboratory testing dates and results, pharmacy data, and ICD-9 codes, then supplemented with manual entry by the health plan of claims data and chart audit data (eg, laboratory results from LDL tests completed and paid for by other means). The registry included CAD patients (regardless of age or continuous enrollment), and tracked demographic information, comorbidities, medications, risk-stratification level, smoking status, and laboratory dates and results.

Setting the Stage for Success One of the early lessons learned by this health plan was that it couldn’t assume that its clinics would be able to implement the disease management program without assistance. The health plan discovered that the best approach was to proactively work with clinic staff to help them implement the disease management program, optimize use of the registry, set up the necessary office processes, and problem-solve the day-to-day challenges. Although this extra effort required additional internal resources, the health plan found that success was much more likely - and clinician acceptance greater - when the program was made manageable for the clinic.

The registry used by the health plan was not only an important source of data on the quality of members’ cardiovascular care, it was also an invaluable component of the plan’s CAD Health Management Program. It facilitated the daily identification of patient needs and the planning of office visits. Office staff used it to implement interventions aimed at improving patient knowledge of, and compliance with, their lipid management regimen. The registry also could be used to help staff remind patients of needed services, monitor pharmaceutical data, and make adjustments based on laboratory data. The registry was, and continues to be, the working tool for physicians’ offices. Despite being labor intensive in its earlier versions, the health plan maintained the registry and considered it to be well worth the investment of both time and resources.

Beginning in the winter of 1998 through early 1999, small group presentations were given at the 14 highest-volume clinics. Lower-volume clinics received this information by mail. These presentations educated clinicians on:

• The CAD Health Management Program


• Updated hyperlipidemia clinical guidelines and patient education materials


• Measures and target LDL levels (inclusive of LDL levels and office visits)


• The need to treat to target levels even if the patient is only slightly above the target


• The registry and the reports they would be receiving

In addition to educating the PCPs and their staff, the health plan met with clinic staff to problem-solve and help them implement the program. Practitioners were provided with monthly risk-stratification lists of their CAD patients so that they could verify diagnoses and laboratory and office visit data. In addition, dedicated internal staff at the health plan provided offices with summary goals reports, assessed the progress of the offices, and followed up with offices regarding variations in practice and participation in the program on a monthly basis. Quarterly assessments of compliance were also conducted, with referrals to medical directors as needed. Every two months, the plan provided practitioners with unblinded data (all health plan clinics and practitioners included and identified) across all performance measures, including LDL control.

Feedback is Key How can you help motivate physicians? Feedback! Physicians respond to information on how they - and their colleagues - are performing. In an effort to give physicians solid information on which to act, this health plan progressed from providing physicians with plan-wide data on LDL control and testing rates to providing them with more through-provoking unblinded data for the health plan overall, the clinic, and each physician within the clinic. This data - combined with the unblinded reimbursement information - proved to be a powerful motivator.

Member interventions included case management of CAD patients who had experienced an acute event, telephonic contacts to encourage compliance with preventive care measures and provide assistance, and educational mailings.

Another Plan's Approach Reaching Out to Patients Patients can be an invaluable source of data for barrier analyses. While the health plan profiled here used a pilot study to identify barriers to optimal disease management, another plan went directly to patients for information on barriers. Using a three-page survey developed internally, the health plan found out that over half of the patients who - about a year earlier - had been diagnosed with hypercholesterolemia while in the hospital or shortly afterwards had not had their cholesterol rechecked during their latest doctor’s visit. Over half (53%) of patients who had high cholesterol in the hospital or afterwards reported they didn’t have high cholesterol at their most recent visit. Among those who had high cholesterol, 34% were having trouble following their diet and 31% couldn’t follow their exercise programs. This health plan focused on adhering to diet and exercise recommendations as the biggest modifiable obstacles to managing cholesterol (see appendix 11). For examples of the patient education components that complemented the QIA, please see appendices 12-13.

With both remeasurements obtained from January 1999 through December 1999 showing statistically significant increases over baseline, the implementation of the disease management program and the registry were proven to be instrumental in improving lipid management (Table 2). The new goal for LDL control in 2000 was raised to 55%, because the program had been so successful that the previous 40% goal had been surpassed by over seven percentage points. The goal for LDL testing remained at 100%.

To add a financial incentive for physicians to improve their clinical performance rates - and to help them offset the added cost of hiring staff to implement the initiative - the program was augmented with a reimbursement plan beginning in January 2000. Physician management bonus fees (average $0.10 per member per month for all members in practice or clinic) were attached to the LDL control measure (LDL level of <100 mg/dL). Unblinded reimbursement data, including the performance rates, were mailed to care groups on a quarterly basis

Another step taken by the health plan to address members’ lack of knowledge about CAD and how to manage lipid levels was the Everyone Teaching Compliance (ETC) sheets, which were made available for education at clinics and pharmacies. These tools could be quickly reviewed with and given to patients to educate them about their disease (CAD) and proper management (lipid control) (see appendices 14-16).

Given the continued, statistically significant improvement in both measures from January 2000 to December 2000, the positive impact of the CAD Health Management Program and associated interventions was clear to the health plan (Table 3). The 2001 goal for LDL control was raised to 60%, because the goal of 55% had been surpassed, while the goal for LDL testing remained at 100%.

Total health care costs for the health plan’s population with CAD were also beginning to decrease. In fact, since the implementation of the CAD Health Management Program, the per member per month costs for the CAD population had decreased, while the health plan’s overall per member per month costs continued to rise.

The health plan also contacted other health plans from various regions in the United States that had achieved a benchmark level of performance on the relevant Health Plan Employer Data and Information Set (HEDIS®) measures (per NCQA’s Quality Compass®) to investigate what additional interventions might be considered. Interestingly, the health plan found that interventions used by the other health plans were the same or similar to its own. Conversations with other health plans did reveal that improved LDL control and testing could be achieved through greater coordination of care from inpatient hospitalization for an acute cardiovascular event to the outpatient setting. As case management was already in place, the health plan decided not to implement any new interventions in 2001, but to continue by intensifying current interventions.

For example, in May and November 2001, an updated version of the registry with a data warehouse was created. It featured daily downloads of demographic, coverage, and clinical data from various sources; daily updates (rather than monthly); and a “drill-down” report that identified patients in need of services at the clinic level. Implementation of the new registry included comprehensive training of all users.

Patient education was also intensified during 2001. As opportunities became available, the health plan’s quality staff gave presentations on good CAD care and the CAD Health Management Program at employer group staff meetings and at meetings with employer representatives.

The positive impact of the CAD Health Management Program and all associated interventions continued from January 2001 through December 2001, as demonstrated by the statistically significant improvements in both measures (Table 4). Since the 2001 goal for LDL control had been achieved, a new goal of 65% was set for 2002. The goal for LDL testing remained at 100%.

At the completion of the study, the health plan had achieved its goal for LDL control, which had risen 27.9 percentage points from baseline to remeasurement 3 (Table 5). The rate for LDL testing rose 26.2 percentage points, from 69.5% at baseline to 95.7% at remeasurement 3, just 4.3 percentage points short of its goal. Statistically significant improvements were seen for both measures, including LDL testing, at all remeasurements.

The health plan reported that the rates for both measures were maintained, but did not increase, in 2002: the LDL control rate was 60% and the LDL testing rate was 94.4%. However, neither result met goal (65% for LDL control and 100% for LDL testing). After analyzing the data, the health plan found that there was a downward shift from higher LDL levels to levels in the 100-130 mg/dL range. Thus, patients were achieving better control, but were having trouble lowering their LDL levels all the way to 100 mg/dL.

• Practitioners have been educated on and are following the recommendations of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III or ATP III)[13]


• All members with CAD receive an educational mailing from the National Institutes of Health (NIH) on the ATP III recommendations


• In keeping with the ATP III guidelines, the LDL control measure was changed in 2003 to the percentage of patients who had achieved an LDL level of <100 mg/dL; previously; the criteria was <100mg/dL


• The health plan is monitoring members who had had a recent acute cardiac event to ensure that LDL control and testing were achieved within on year of the event


• Letters are now being mailed annually to members with CAD, reminding them to have their LDL checked each year

The health plan is also working with its cardiology offices to pilot a program that uses the registry to manage non-CAD patients with hyperlipidemia, moving the care paradigm toward prevention.

[1] - Data on file, Pfizer Inc, New York, NY.

[2] - National Diabetes Education Program. The link between diabetes and cardiovascular disease. Available at: www.ndep.nih.gov/control/CVD.htm Accessed June 6, 2003.

[3] - Keevil JG, Stein JH, McBride PE. Cardiovascular disease prevention. Prim Care Clin Office Pract. 2002;29:667-696.

[4] - Pearson TA, McBride PE, Miller NH, Smith Stephen Covey Jr. 27th Bethesda Conference: matching the intensity of risk factor management with the hazard for coronary disease events. Task Force 8. Organization of preventive cardiology service. J Am Coll Cardiol. 1996;27:1039-1047.

[5] - American Heart Association. Heart Disease and Stroke Statistics - 2003 Update. Dallas, Tex: American Heart Association; 2002.

[6] - World Health Organization. Cardiovascular death and disability can be reduced more than 50 percent [press release]. October 17, 2002. Available at: www.who.int/mediacentre/releases/pr83/en/print.html. Accessed June 19, 2003. . Accessed July 24, 2003.

[7] - Gaede P, Vedel P, Larsen N, et al. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med. 2003;348:383-393.

[8] - Goldman L, Garber AM, Grover SA, Hlatky MA. 27th Bethesda Conference: matching the intensity of risk factor management with the hazard for coronary disease events. Task force 6. Cost effectiveness of assessment and management of risk factors. J Am Coll Cardiol. 1996;27:964-1047.

[9] - Elliott WJ, Weir DR. Comparative cost-effectiveness of HMG-CoA reductase inhibitors in secondary prevention of acute myocardial infarction. Am J Health Syst Pharm. 1999;56:1726-1732.

[10] - Grover SA, Coupal L, Zowell H, Dorais M. Cost-effectiveness of treating hyperlipidemia in the presence of diabetes: who should be treated? Circulation. 2000;102:722-727.

[11] - Physician Consortium for Performance Improvement. Clinical performance measures: chronic stable coronary artery disease. American Medical Association. Available at: www.americanheart.org/downloadable/heart/1055798504173CADMiniSetR030158_final.pdf. Accessed August 23, 2003.

[12] - Tipton ML, Fleming M. Stroke prevention in managed care: a five-dimensional health improvement model. In: Gorelick PB, After M, eds. The Prevention of Stroke. New York, NY: The Parthenon Publishing Group; 2002:123-130.

[13] - National Cholesterol Education Program. Third Report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Bethesda, MD: National Heart, Lung, and Blood Institute; 2002. NIH publication 02-5215.

[14] - Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 Report. JAMA. 2003;289:2560-2572.