What is quality, and how is it measured?

Past competition between health providers has been based largely on comparative costs to the consumer. Proliferation in the types of health plans offering coverage has complicated decision making. Government, employers, and individuals, confronted by variations in premiums, benefits, co-insurance rates, consultant panels, and other aspects of coverage, are now looking at quality of care as the critical criterion. Third-party payers are demanding that quality of care be documented and given a priority as high as measures of cost-effectiveness.

There are two aspects to quality assessment:

* Subjective--How the patient perceives quality of care.

* Objective--How clinical care compares to established standards.

Performance measurements of quality in the past have looked at the structure, process, and outcomes of health care delivery systems. [1] However, so many intrinsic factors influence outcomes (e.g., genetics, patient compliance, socioeconomic status, comorbid states, etc.) that relationships have been difficult to verify.

The subjective, or patient, perception of quality is readily obtainable through patient interview surveys. The objective aspects of medical care still lack national standards, as practice patterns vary considerably from area to area. [2,3] Brook [4] and others are seeking acceptable outcome measures to assess quality of care. Rising health care costs have created concern about cost-effectiveness in the global spectrum of quality evaluation when different processes with varying costs produce similar outcomes. [5,6]

Medical care quality is a complex and abstract concept that eludes precise quantification. However, it can be evaluated if we define quality of care in terms of desired patient outcomes in their contacts with health care providers. A number of quality assessment monitors are currently operational at AV MED.

The AV-MED Health Plan is an IPA-model HMO with Plan programs in five metropolitan areas--South Florida, Tampa, Orlando, Jacksonville, and Gainesville. In 1989, there were 209,900 members, 4,000 physician providers, and 82 contract hospitals, plus ancillary diagnostic and resource services under contracts. The central office in Miami processes most of the medical, fiscal, and claims data through a sophisticated MIS Department. Products offered are HMO, PPO, and Medicare Plans. The Corporate Medical Office has a well-developed Medical Care Research Unit and a Quality Assurance Program for coordinating activities in all plan areas. A primary care physician (PCP) is selected by each member for case management. The PCPs are family practitioners, internists, and pediatricians, most of whom are paid by capitation. Specialist consultants are generally fee-for-service practitioners.

Methodology

A series of ongoing monitors provide data for analysis of utilization rates, volume of potential adverse outcome diagnoses (e.g., diabetic keto-acidosis, stroke, adverse drug reactions, etc.), patient complaints, and other information that identifies potential problems requiring investigation. Monitors related to quality assessment are described below.

An annual Telephonic Patient Satisfaction Survey of all primary care physicians measures satisfaction of members with their total health care. Responses involve access, staff cooperation, physician attitude, waiting time, outcome of care, and perceived quality of care. A final "satisfaction score" is obtained from the interviewee using a score of 1 to 10, 10 being very satisfied and 1 dissatisfied. Twenty randomly selected members of each physician's panel are contacted to complete this survey. The survey tool is shown in figure 1, page 23.

In 1987, the average of all scores obtained was 8.6, indicating that 86 percent of the members were satisfied. In 1988, the survey indicated an 87 percent satisfaction rate. The goal of continuous improvement looks for higher degrees of satisfied members in future surveys. Each physician receives a report of his members' positive and negative comments, and potential behavior changes are monitored in future surveys.

A Medical Expense Report, profiling each individual Primary Care Physician's cost per member per month (PMPM), is provided to the PCPs on a monthly basis, using cumulative accounting for the year. Costs include capitation payments, copayments, fee-for-service payments, physician consultant costs, hospital inpatient costs, hospital outpatient costs, and diagnostic and prescription costs. Total panel costs to AV-MED are totaled and reduced to a cost PMPM.

An adjustment to this cost is made to reflect the percentage of high-cost patients in the physician's panel, to the extent that this is different from the percentage of such patients for all physicians. High-cost patients are those that create costs of $7,500 or more a year. Costs over that level are not charged against the physician's account. A greater than average number of high-cost members lowers the defined PMPM figure, and less than average high-cost members raises the PMPM by a calculated factor related to the extent of the variance.

Comparison is made with the average PMPM cost for all capitated Primary Care Physicians. This is determined independently for each patient age group and the comparison PMPM cost represents the weighted average of all age groups in the physician's panel. This cost-effectiveness comparison is based on a peer group average for a hypothetical patient panel that matches the physician's panel in terms of age group characteristics and percentage of high-cost patients. A typical Medical Expense report is shown in figure 2, page 24.

Quality of care evaluations for each Primary Care Physician are based on composite scores of three measures:

* Office medical record reviews for PCPs scored by standards set by a peer group (see figure 3, page 24).

* Verified data from patient complaint logs, as well as serious complaints from patient satisfaction surveys per 100 members.

* The number of patient transfers to another PCP based on patient dissatisfaction per 100 members.

Other factors that affect a quality of care score are poor results from our frequent Quality Assessment Surveys of office records, such as hypertension screening, rate of completed immunizations, etc. A computer program is nearing completion to flag adverse outcomes that involve substandard care for a physician's panel members.

The AV-MED Physician Incentive Bonus Plan Based on Quality of Care creates a composite score for each primary care physician. The composite score is based on the Patient Satisfaction Score, quality of care measurement, and cost-effectiveness compared to peers, i.e., pediatricians compared to pediatricians, internists to internists, and family practitioners to family practitioners. Each of these measures receives equal weight in the final tabulated score. A sorting of highest to lowest scores for each primary care group is constructed using a standardized distribution for the composite variables (Fisher's Z-distribution) and then transforming them to a T-distribution with a mean of 70 and a standard deviation of 10.

The highest scorers receive a bonus at the end of the year amounting to 15 percent of their income for that year. The bonus decreases with decreasing scores and is zero for the lowest 25 percent of the list. The lowest 10 percent receive a consultation visit from their medical directors, and if improvement does not occur, they are considered for termination.

Discussion

Physician's reactions to this program have been very positive. In 1989, about $400,000 was distributed in bonuses representing care that was patient satisfying and of good quality. A correlation analysis, using a scattergram presentation, revealed a definite relationship between high patient satisfaction and cost effectiveness. This justified the use of the quantifiable cost-effectiveness program as a component of the quality assessment factors in the bonus plan.

Feedback of comments from patient interviews was requested by the involved physicians and has been well received. The monthly Medical Expense Reports have startled some of the physicians, and, with discussions from the medical directors, practice patterns have shown change. The Quality Assessment Surveys of office care have been followed by repeat surveys in six months for under-achievers. These repeat surveys invariably demonstrate an increase in performance to the set standard. Continued monitoring is done to evaluate possible backsliding to old behavior.

One area that we would like to strengthen is specific measurements of quality of care of our physicians by more quantifiable means. The evaluation of quality of care for the primary care physicians is considered of highest priority, as basic health maintenance, preventive care, and most medical decisions start there. Evaluating the quality of the specialist panels is also important and is an ongoing process. The Medical Care Research Unit is collecting data on variations in practice patterns in the five Health Plans of AV-MED. Cesarean section rates, hysterectomy rates per 1,000 members between 25 and 59 years, appropriateness of coronary angiography procedures, and other studies identify outlier performance that leads to quality evaluations by peer groups of the involved specialty. This is an important aspect of quality of care evaluation that deserves ongoing monitoring.

References

[1] Donabedian, A. The Criterion and Standards of Quality. Explorations in Quality Assessment and Monitoring, Vol. II. Ann Arbor, Mich.: Health Administration Press, 1982.

[2] Wennberg, J., et al. "Use of Claims Data Systems to Evaluate Health Care Outcomes. Mortality and Reoperation Following Prostatectomy." JAMA 257(7):933-6, Feb. 20, 1987.

[3] Chassin, M., et al. "Does Inappropriate Use Explain Geographic Variations in the Use of Health Care Services? A Study of the Procedures." JAMA 258 (18):2533-7, Nov. 13, 1987.

[4] Brook, R., et al. "Assessing the Quality of Medical Care Using Outcome Measures: An Overview of the Method." Medical Care (Supp.) 15(9, Suppl): 1-165, Sept. 1977.

[5] Luft, H. "How Do Health Maintenance Organization Achieve Their Savings?" New England Journal of Medicine 298(24):1336-43, June 15, 1978.

[6] Kemeny, M., et al. "Measuring Adequacy of Physician Performance: A Preliminary Comparison of Four Methods in Ambulatory Care of Chronic Obstructive Pulmonary Disease." Medical Care 22(7):620-31, July 1984.

Jerome Beloff, MD, is Consultant and Member, Board of Directors, AV-MED Health Plan, Miami, Fla. He is a member of the College's Society on Managed Health Care.