A.3
Prototype Indicator Set: Elder Health
BACKGROUND
The age structure of the United States, indeed that of North America, is "graying." The most rapidly growing segment of the general population is the group age 85 and over. If present trends continue, approximately 20 percent of the population will be 65 years of age or older by the time this dynamic change has peaked in the year 2050 (Bureau of the Census, 1995). Providing appropriate health supports for this growing population, especially the frail elderly, will challenge many sectors within society.
The demographic shift has several important implications for health and health care. First, health care expenditures will be driven upward by the demographic shift since the elderly are provided health care resources out of proportion to their numbers. Next, the need for a full continuum of care for the frail elderly will become fully apparent in the next several decades, greatly expanding the demand for nursing home capacity, congregate care facilities, adult day care programs, and respite as well as other care giver support programs. The growth of capacity across this full continuum will be accelerated by the further penetration of managed care, since integrated systems of managed care will continue to drive down hospital use. Third, the health care workforce is not adequately prepared to meet the need for geriatric care, including assessment and care management services. Consider
able investment will have to be made in undergraduate and graduate programs if appropriate services and care are to be available. Finally, the scope of "health care services" for the elderly is broad. Curative services remain important, but as the burden of chronic diseases increases with age, maintenance of function and satisfaction with care loom large as the primary outcomes in assessing the appropriateness and effectiveness of health care services. Critical services will be diverse, including clinical and personal care, functional assessment, education and social services, transportation, housing, social support, income supplementation, and others. Further, it is possible that social investments outside clinical geriatrics (e.g., in lay care giver training or in housing or social environments) will, in the end, be more effective in improving elder health than those in the medical care sector. Even within the domain of clinical activities, some critical services for sustaining elder health (e.g., annual influenza vaccination) may be the "product" of multiple providers such as the public health department or senior citizen centers, as well as the responsibility of clinical organizations.
"FIELD" SET OF PERFORMANCE INDICATORS
Given the size of the elder population, its political salience, and importance to medical care expenditures, many stakeholders would see the health status and effectiveness of services provided to this population as important, including providers, insurers, and state and local public health authorities. The frail elderly are an at-risk population in head-to-head competition with education of youth in state budget plans. Even employers and local industry, to the extent that they are committed to medical care for their retired employees and are concerned about current employees' obligations to care for elderly relatives, have a stake in the health of the elderly.
By using the domains of the field model, it is possible to identify a variety of measures that might serve as performance indicators for a community's efforts to improve the health of the elderly. Because of the nature of aging, both the medical and the social needs of elders must be addressed.
Health Care and Disease
An extraordinarily diverse set of performance indicators could be considered for use as part of a monitoring system that ad-
dresses health care and disease among the elderly. At a state level, both U.S. Medicare and Medicaid data sets have been used to characterize small-area variations in care, utilization of specific health care services by the elderly, mortality rates for specific procedures, hospital mortality rates, and even individual provider performance. Medicare data are specifically applicable to elders. Medicaid data are a source of information on the use of long-term care resources. Other potential sources of information on available resources at a community level would include health department information and "community resource" directories available in Area Agency on Aging offices in the United States.
In these data sets or others, potential indicators include a large universe of utilization and outcomes measures expressed as annual rates per thousand Medicare- or Medicaid-eligible persons living in a geographically defined residence community, ideally divided into two age strata, 65–84 years of age and 85 and over. These might include annual rates for mortality, hospitalization, nursing home days, physician visits, myocardial infarction, stroke, hip fracture, transurethral resection of the prostate, cholecystectomy, and coronary artery bypass surgery.
Kaiser Permanente's Northern California Region performance area "bundles" (see Table A.3-1), including those for cardiovascular disease, cancer, and common surgical procedures, suggest indicators that could be appropriate for elderly age groups (Center for the Advancement of Health and Western Consortium for Public Health, 1995). HEDIS-like indicators such as cholesterol screening, mammography, colorectal cancer screening, and provision of influenza vaccination annually would be highly appropriate. If cancer registry information is available, the incidence per thousand of late-stage breast cancer and advanced-stage colorectal cancer would be appropriate indicators. Health care providers and health care plans alike would be interested in community-level measures of self-rated health status and disability days, should these be available. With the exception of functional status measures at a community level, virtually all the potential indicators cited above make use of available data of reasonably good quality.
Social Environment and Prosperity
Indicator sets in general use have emphasized medical care utilization measures, relatively underrepresenting key social and economic determinants of the functional health of the elderly.
TABLE A.3-1 Kaiser Permanente Northern California Region Areas of Performance Measurement
Member Satisfaction |
|
Confidence in medical care |
|
Access to care |
|
Service |
|
Overall satisfaction |
|
Childhood Health |
|
Rates of preventable diseases |
|
Immunization rates |
|
Disease outbreaks per 100,000 members |
|
Pediatric asthma |
|
Accidental poisoning |
|
Perceptions of experience of pediatric care |
|
Maternal Care |
|
Rates of prenatal care |
|
Prenatal screening rates |
|
Birth outcomes (LBW, VLBW, neural tube defects, complex newborn rates, percentage of births to ICU, in-hospital mortality rate, perinatal mortality rate) |
|
Cesarian section rates |
|
Vaginal birth after cesarian section rate |
|
Perceptions of experience of obstetric inpatient care |
|
Cardiovascular Disease |
|
Cholesterol screening rate |
|
AMI inpatient discharge rate |
|
AMI in-hospital mortality rate |
|
AMI mortality rate within 30 days of admission |
|
CABG inpatient discharge rate |
|
CABG mortality rate within 30 days of admission |
|
Heart disease mortality rate |
|
Hypertension screening rate |
|
Hypertension screening follow-up rate |
|
Hypertension treatment effectiveness (normal blood pressure after one year) |
|
Cerebrovascular accident inpatient discharge rate |
|
TIA inpatient discharge rate |
|
Cerebrovascular disease mortality rate |
|
Cancer |
|
Mammography screening rate |
|
Breast cancer stage at diagnosis (local, regional, distant stages) |
|
Breast cancer five-year survival rate (by local, regional, distant stages at diagnosis) |
|
Breast cancer mortality rate |
Acute general hospital event-based measures, moreover, have been used more frequently than measures that relate to health maintenance, rehabilitation, and long-term care.
As people enter elderly age groups, it is likely that they will face chronic diseases and conditions such as arthritis, diabetes, osteoporosis, and senile dementia (Bureau of the Census, 1995). Elderly persons with such conditions are likely to need assistance in performing the activities of daily living. Among noninstitutionalized elderly, 9 percent of those aged 65 to 69 and 50 percent of those 85 years or older needed assistance with daily activities such as bathing, preparing meals, and doing chores around the house (Bureau of the Census, 1995). This information raises concerns about the availability, accessibility, and quality of services for the elderly who need assistance with daily activities but are able to live in noninstitutional settings. Although only 1 percent of people aged 65 to 74 lived in nursing homes in 1990, nearly 25 percent who were age 85 or older did. These data point to the growing need for a range of social and health services for people between the ages of 65 and 84, as well as the growing need for institutional services for the "oldest old."
As the elderly grow in number, it will also be necessary and appropriate to monitor the development of an expanded capacity in the full continuum of care—that is, nonmedical services to assist with daily activities (e.g., bathing, meals, chores, transportation), low-intensity periodic nursing or medical services (e.g., checking vital signs, blood sugar, or medication compliance; changing dressings), and institutional services (e.g., adult day care, congregate care, senior housing with modified physical and social environments, nursing home care, respite care). In addition, coordination with the long-term care community is needed to ensure that adequate, appropriate-level, high-quality care for the elderly is available in the community.
Cooperation among federal, state, and local health agencies; social and housing agencies; community residents; and the medical care community is essential for achieving improved elder health. Communities could compile information from special studies that link health and other sectors; indicator measures might include various rates (expressed as an annual incidence per thousand persons aged 65 and older), including crimes against persons, residential burglaries, senior citizens bus ridership, participation in library services, access to sporting and cultural events (counting "senior citizen discount" tickets issued), voting in public elections, and property ownership.
Physical Environment
People negotiate the challenges of the physical environment every day as they drive automobiles and heavy equipment, walk on busy streets, do chores around the house or farm, and work in high-crime metropolitan areas. As individuals age, the challenges presented by the physical environment can become overwhelming due to changes in vision, hearing, bone density, muscle tone, and response time. Modifications in the physical environment and in individual behavior may be necessary to ensure that the elderly are able to maintain their independence and quality of life (IOM, 1990; Nuffield Institute for Health, 1996).
SAMPLE INDICATOR SET
-
Self-rated health status.
This indicator was chosen for its direct salience to the aims of health and social services for the elder population. Under current circumstances, measuring self-rated health status would require a special community-wide survey, although attempts could be made to "model" the community-based result from data routinely available at health departments (from the Behavioral Risk Factor Surveillance System), Area Agencies on Aging, senior citizen centers, or medical care plans operating in the community. Establishing accountability for maintaining health status would certainly be difficult, but it could be achieved if health status information were available for all and if individual organizational performance could be characterized on the basis of health status maintained in served populations (adjusted for age and gender). In a less precise approach, the distribution of poor health status in the community could be compared with catchment areas for clinical organizations, to determine whether community health organizations were serving vulnerable populations.
-
Physician visits per annum.
These data should be available from Medicare. This indicator is a utilization-based measure of access to health services.
-
Area-adjusted average per capita medical care expenditures for the elderly
This is a direct measure of medical care expenditure intensity available from Medicare. It can be interpreted as a measure of
-
population need, the resource intensity of conventional medical practice, or both.
-
Influenza vaccination.
This is an efficacious service with major implications for morbidity and mortality from respiratory disease (influenza or pneumonia). Medicare claims data should be able to provide an approximation of the influenza vaccination rate in the community, but these data will not reflect vaccinations provided by hospitals, health departments, or managed care organizations. Accountability for achieving high vaccination rates among the elderly is shared. More specific measurements could be made within health care delivery systems in order to attach accountability more directly to organizational performance, but this approach would not provide community-wide information about coverage achieved.
-
Advanced-stage cancers of the breast.
A cancer registry (state health department) should be a reliable source of information about breast cancer in a community; other sources might include Medicare. This is a ''sentinel event" measure of relevance to older women's health care. Accountability within systems of care for improving the performance of mammography and clinical examination screening could be established if individual-level data can be linked to a source of care.
-
Percentage of elderly residing in a nursing home on a given date.
For at least a subset of the population, these data should be available from Medicaid. State health department surveys of nursing homes and the decennial census are alternate sources of data. The indicator is a measure of population frailty, the lack of available alternative services for the frail elderly, or both.
-
Presence of the full continuum of care.
As the elderly increase in number, it will also be necessary and appropriate to monitor the development of an expanded capacity in the full continuum of care—that is, nonmedical services to assist with daily activities (e.g., bathing, meals, chores, transportation), low-intensity periodic nursing or medical services (e.g., checking vital signs, blood sugar, or medication compliance; changing dressings), and community long-term care services (e.g., adult day care, congregate care, senior housing with modified physical and social environments, nursing home care, respite
-
care). In addition, coordination with the long-term care community is needed to ensure that adequate, appropriate-level, high-quality care for the elderly is available in the community.
This is a measure of "care capacity," one that could presumably be based on information from the Area Agency on Aging. Accountability for developing and maintaining the full continuum of care can best be described as a shared responsibility, since government, voluntary organizations, and the health care industry may all be needed to create the capacity.
-
Library readership, voting.
Social participation is a health-enhancing feature of an elder's lived experience and a manifestation of expanded function. This indicator cluster is a direct attempt to characterize the participation of elders in the social life of the community. Other possible measures might include senior theater, cinema, or other special event tickets per population base; bus ridership; church membership; and membership in the American Association of Retired Persons. Final selection of the measures included in this indicator should be appropriate to the community's social structure (e.g., bus ridership may not be appropriate in rural areas with no regular bus routes).
-
Senior citizen income and property ownership.
It is very difficult to measure the "prosperity" of the elderly. Income data are available but may underrepresent the wealth and savings of the elderly. Property ownership is an indirect measure of economic well-being but is a major component of the personal estate of older people in our society. Other possible approaches to assessing economic well-being might be to use Internal Revenue Service data to characterize income or bank data to measure savings, but both of these approaches seem less feasible. Accountability for economic well-being at a community level is difficult to establish, but the identification of populations in special need or at special risk would seem to be feasible if economic well-being can be measured.
-
Crimes against elderly persons or residential burglaries.
This measure of well-being is focused on personal safety, a major quality-of-life issue among the elderly, who may be viewed as "easy targets" by criminals. This information should be available from the public safety databases within a community. Accountability, at the first level of analysis, may reside with the
-
police. At deeper levels of "ecological analysis," features of housing, transportation, and economic development within neighborhoods or urban subareas may also be relevant.
-
Falls among the elderly that result in hospitalization.
Falls among the elderly are a major cause of morbidity, disability, and mortality (IOM, 1990). Although hip fracture is the most devastating consequence of nonfatal falls, other consequences such as soft tissue injury, loss of mobility, and fear of falling can have a serious impact on quality of life. The prevention of falls provides an opportunity for multisectorial collaboration and cooperative efforts (Nuffield Institute for Health, 1996). Falls may have causes that are health related, pharmacologic, environmental, behavioral, or activity related (IOM, 1990).
This set of indicators has been composed to represent several underlying constructs, including health status (self-rated health, nursing home days), access to medical care (physician visits), resource use (per capita medical care expenditures), health care system capacity (continuum of care availability), states of well-being (economic, personal safety, social participation), critical health care services (influenza vaccination), and sentinel events (advanced-stage breast cancer). The indicator set substantially underrepresents the universe of potential indicators drawn directly from a medical care sector. This latter choice was a conscious one and could be controversial.
Taken together, the complex of indicators creates a profile of elder health production in a community, at least as one can characterize this process as cross-sectional. In the aggregate, it is an attempt to characterize the performance of health-relevant systems from a determinants-of-health perspective. In the aggregate, these indicators provide a relatively rich characterization of the health, health care utilization, social participation, and social welfare of the elderly population residing in a community. Health improvement initiatives might easily focus on any of these indicator sectors—from long-term care capacity to improving the quality of life for frail elderly residing in nursing homes. Any of these measures take on additional meaning if they can be compared (from one community to another or to a statewide average measure).
Specific organizational accountability for elder health is hard to assign on the basis of these indicators. Very few of the measures are likely to be precise enough to characterize the perfor-
mance of single-provider organizations or even health care plans of modest size. Establishing accountability for performance by such a disparate set of "providers" as nursing homes, hospitals, physicians, metropolitan transportation systems, libraries, and the police will require forging an accountable community coalition within which collaborative action can arise and joint accountability be felt.
REFERENCES
Bureau of the Census. 1995. Sixty-Five Plus in the United States. Statistical Brief. SB/95-8. Washington, D.C.: U.S. Department of Commerce.
Center for the Advancement of Health and Western Consortium for Public Health. 1995. Performance Indicators: An Overview of Private Sector, State and Federal Efforts to Assess and Document the Characteristics, Performance and Value of Health Care Delivery: Report on Field Research . Washington, D.C.: Center for the Advancement of Health.
IOM (Institute of Medicine). 1990. The Second Fifty Years: Promoting Health and Preventing Disability. R.L. Berg and J.S. Cassells, eds. Washington, D.C.: National Academy Press.
Nuffield Institute for Health and NHS Centre for Reviews and Dissemination. 1996. Preventing Falls and Subsequent Injury in Older People. Effective Health Care 2(4):1–16.
TABLE A.3-2 Field Model Mapping for Sample Indicator Set: Elder Health
Field Model Domain |
Construct |
Sample Indicators |
Data Sources |
Stakeholders |
Health Care, Disease |
Reduce impact of disease among the elderly |
Self-rated health status |
Community survey (BRFSS data from health department), Area Agencies on Aging, senior citizen centers, medical care plans |
Health care providers Health care plans State health agencies Local health agencies Medicare Agency on Aging Community organizations |
|
|
Physician visits per annum |
Medicare |
Special health risk groups General public |
|
|
Area-adjusted average per capita medical care expenditures for the elderly |
Medicare |
|
|
|
Influenza vaccination |
Medicare |
|
|
|
Advanced-stage cancers of the breast |
State cancer registry, Medicare |
|
|
|
Percentage of elderly residing in a nursing home on a given date |
State health department surveys, census, Medicare |
|