Measuring the "Managedness" and Covered Benefits of Health Plans

Health Services Research, August, 2000 by David E. Grembowski, Paula Diehr, Louise C. Novak, Amy Elizabeth Roussel, Diane P. Martin, Donald L. Patrick, Barbara Williams, Cornelia M. Ulrich

Study Aims. (1) To develop indexes measuring the degree of managedness and the covered benefits of health insurance plans, (2) to describe the variation in these indexes among plans in one health insurance market, (3) to assess the validity of the health plan indexes, and (4) to examine the association between patient characteristics and the health plan indexes. Measures of the "managedness" and covered benefits of health plans are requisite for studying the effects of managed care on clinical practice and health system performance, and they may improve people's understanding of our complex health care system.

Data Sources/Study Setting. As part of our larger Physician Referral Study, we collected health insurance information for 189 insurance product lines and 755 products in the Seattle, Washington metropolitan area, which we linked with the study's data for 2,277 patients recruited in local primary care offices.

Study Design. Managed care and benefit variables were constructed through content analysis of health plan information. Principal component analysis of the variables produced a managedness index, an in-network benefits index, and an out-of-network benefits index. Bivariable analyses examined associations between patient characteristics and the three indexes.

Principal Findings. From the managed care variables, we constructed three provider-oriented indexes for the financial, utilization management, and network domains of health plans. From these, we constructed a single managedness index, which correlated as expected with the individual measures, with the domain indexes, with plan type (FFS, PPO, POS, HMO), with independent assessments of local experts, and with patients' attitudes about their health insurance. For benefits, we constructed an in-network benefits index and an out-of-network benefits index, which were correlated with the managedness index. The personal characteristics of study patients were associated with the managed care and benefit indexes. Study patients in more managed plans reported somewhat better health than patients in less managed plans.

Conclusions. Indexes of the managedness and benefits of health plans can be constructed from publicly available information. The managedness and benefit indexes are associated with the personal characteristics and health status of study patients. Potential uses of the managed care and benefits indexes are discussed.

Key Words. Health insurance, managed care, managed care organization, health maintenance organization, health outcomes, plan satisfaction, utilization, costs, health policy, measurement

"It looks and feels like a PPO and yet underneath the hood is our HMO management and cost containment."

from American Healthline [1]

In the era of managed care and market-driven health reform, fundamental changes are occurring in the U.S. health care system. Faced with higher costs that erode profits and competitiveness, employers, as well as government and nonprofit agencies, have replaced their traditional fee-for-service (FFS) health plans with one or more health plans offered by a variety of managed care organizations (MCOs), such as preferred provider organizations (PPOs), point-of-service (POS) plans, and health maintenance organizations (HMOs) (Miller and Luft 1994, 1997; Iglehart 1992; Miller 1988). To control the costs of their health plans, MCOs place a variety of constraints and incentives on physician and patient behavior. As more and more Americans obtain their health care through managed health plans (Winslow 1998), it becomes more important to understand the influence of their components on the delivery and outcomes of care.

Very little is known about the influence of managed health plans on the cost, utilization, and quality of care, partly because well-developed measures of managed care do not exist. Previous studies typically examined the association between the type of health plan (such as FFS, PPO, POS, or HMO) and the process and outcomes of care. However, these measures are problematic because MCOs manage care in different ways, and distinctions between types tend to blur (Johnson and Crystal 1997; Schoen and Davidson 1996; Isaacs 1996). As a consequence, associations between plan type and the delivery and outcomes of care are often inconsistent (Miller and Luft 1997). Even when consistent positive or negative associations are detected, it is often unclear which components produced them, undermining the ability of decision makers to pinpoint what does or does not contribute to quality care.

An alternative approach is to define managed care by the methods that health plans use to manage costs and improve quality of care (Grembowski et al 1998; Fraser 1997; Johnson and Crystal 1997; Schoen and Davidson 1996). Other things equal, as the number and strength of the methods increase, so does the intensity, or "managedness," of health plans. Because the methods vary in complex ways across MCOs (Schoen and Davidson 1996; Gabel et al. 1997), variation in the methods--and therefore, the intensity of managed care--may be associated with the delivery and outcomes of care (Hillman, Pauly, and Kerstein 1989). Developing measures of these managed care methods is requisite to studying ways in which they influence the delivery and outcomes of care, as well as patient selection and satisfaction with health plans.