THE MERCK MANUAL OF GERIATRICS, Ch. 16, Managed Care

Section 1. Basics of Geriatric Care
Chapter 16. Managed Care
Topics: Introduction \| Medicare Managed Care Organizations \| Health Risk Appraisal \| Geriatric Care Management \| Disease Management

Health Risk Appraisal

A form of screening to identify persons who are likely to need complex health care, who are at risk of adverse health outcomes, and who will benefit from care management programs.

Health risk appraisals are used in managed care as a preventive population-based approach to health screening (see Figure 16-1). Traditional fee-for-service programs rarely use these appraisals because of the lack of incentives and reimbursement. Using health risk appraisals with targeted geriatric care management can reduce costs and improve the quality of care. Targeting patients who are too healthy to benefit from the programs results in inefficient utilization of geriatric care management. It is illegal for an MCO to use risk appraisals on persons before they enroll in the MCO. This policy prevents MCOs from "cherry picking," the practice of enrolling only healthy persons.

Health risk appraisals may be based on self-reported health risk data and/or administrative data. Self-reported health risk data can be obtained via surveys that assess health status based on various health elements or on screening instruments, which are brief questionnaires used to identify persons at risk of adverse health outcomes (see Figure 16-2). Surveys may be conducted by personal interview, by telephone, or by mail. Interviews and telephone surveys are generally more expensive and time consuming than are mailed surveys, although they obtain more reliable and more detailed clinical information. Mailed surveys are less expensive, are more efficient, and have relatively high response rates (50 to 60%) in the elderly.

Administrative data include data derived from insurance claims, including medical and pharmacy claims, and data derived from primary care visits. Administrative data are relatively inexpensive to obtain in most Medicare MCOs with sophisticated information systems. However, the quality, timeliness, and availability of such data can be suboptimal.

In general, self-reported health risk data and administrative data equally predict health care use. However, because response rates to self-reported health surveys are generally 50 to 60%, with nonresponders usually having higher health risk, administrative data have advantages because they can be obtained for almost all members and at a much lower cost. For persons newly enrolled in an MCO, self-reported health surveys are more useful for identifying those at high risk because administrative data are generally unavailable. Persons with certain problems (eg, smoking, alcohol use, cognitive impairment, falls, nutritional problems) also cannot generally be identified by administrative data. MCOs often use self-reported health risk data to identify such persons, so that they can be directed to disease management programs. Self-reported health risk data are also useful for promoting wellness (eg, exercise, stress reduction, smoking cessation) in the healthy elderly and for eliciting patient preferences, self-perceived health status, values, and the need for ancillary service (eg, homemaker services, transportation).

Health risk appraisals are the first step in the population-based medical management of health care. Health risk appraisals must be linked to care management programs so that high-risk members can be proactively managed. The proportion of high-risk members (those at risk of adverse health outcomes who are likely to need complex care) in a Medicare MCO profoundly affects the MCO's earnings; typically, 5% of members at the highest risk account for 62% of hospital costs. At present, Medicare pays a fixed capitation (called the average adjusted per capita cost, or AAPCC) for each member, which reflects 95% of the average cost to Medicare under fee-for-service programs for a patient of the same age, sex, and social situation (eg, residence, employment status, financial status). Thus, MCOs with many high-risk members (adverse selection) may lose money, whereas MCOs with few high-risk members (favorable selection) may profit.

Responding to this situation, the Balanced Budget Act of 1997 required that the Health Care Financing Administration (HCFA) institute a health risk-adjusted payment system by January 1, 2000. This system decreases payments to MCOs with favorable selection and increases payments to MCOs with adverse selection, thus potentially increasing the financial incentive to enroll and retain high-risk elderly patients. Consequently, health risk data, primarily from administrative data, will become increasingly important and will be required.