Overview: Surveillance for Selected Public Health Indicators Affecting Older Adults -- United States
Donald K. Blackman, Ph.D.1
1Division of Cancer Prevention and Control
The United Nations has proclaimed October 1, 1998, through December 31, 1999, as the International Year of Older Persons (IYOP). Federal agencies are working together to sponsor IYOP activities in the United States. To commemorate the goals of IYOP, CDC has published these surveillance summaries to describe important health issues and to highlight the role of public health surveillance for older adults aged greater than or equal to 65 years in the United States. Although older adults are the focus of these surveillance summaries, persons aged 55-64 years have also been included, when data were available, as a comparison group.
The concepts and methods of public health surveillance are useful for meeting information needs for and about older adults in the United States. Surveillance, a core public health activity, is defined by CDC as the "ongoing, systematic collection, analysis, and interpretation of health data essential to the planning, implementation, and evaluation of public health practice, closely integrated with the timely dissemination of these data to those who need to know. The final link of the surveillance chain is the application of these data to prevention and control practices" (1). Public health interventions have been the major factor underlying the greater than 30-year increase in life expectancy since 1900 (2). Surveillance provides data to support the U.S. public health mission of improving health and quality of life as well as extending lives.
Public health surveillance has evolved and expanded during the preceding 40 years from activities that focused primarily on the prevention and control of acute infectious diseases to include chronic diseases, injuries, risk factors, and health practices (3). As a part of public health practice, surveillance is used to identify needs, inform policy, and guide action. Public health surveillance for older adults should include a spectrum of activities, including estimating disease, disability, and service-use rates; identifying older adults who are at high-risk and underserved; and studying risk factors for illness and disability (4). As with chronic and infectious disease surveillance in general, each level of government has various needs and priorities for surveillance data and concentrates on specific aspects of surveillance (5).
Public health surveillance is likely to become increasingly important to older adults in the United States. Although infectious diseases (e.g., pneumonia and influenza) continue to be a major cause of morbidity and mortality among older adults, the prevalence of chronic disease increases with age, and chronic conditions are associated with disability and mortality, either alone or in combination with age-related physiologic changes (6). Although an important goal is that each generation have better health among its older adults than preceding generations, age-related increases in the prevalence of chronic diseases and injuries or their sequelae are not expected to disappear (7,8). Older adults will continue to experience more chronic conditions than younger persons, experience more activity limitations and disability related to chronic disease, use more health-care resources because of chronic diseases, and have multiple chronic conditions (comorbidities) among the oldest of the elderly (6). Because of growth in the number and proportion of persons in the United States who will be aged greater than or equal to 65 years in the coming decades and because of substantial increases in life expectancy, the number of older adults with chronic diseases will remain high. However, public health interventions can decrease their risk of chronic diseases and injuries. Research indicates that changes in lifestyle (e.g., increasing physical activity and eating a balanced diet rich in fruits and vegetables) and using preventive services (e.g., cancer screening and vaccination against disease) help prolong the health and preserve the quality of life of older adults (7).
The surveillance summaries in this publication explore several major issues related to the health and abilities of older adults and demonstrate the potential of surveillance using available databases. This publication focuses on selected public health issues, although all issues related to public health surveillance apply to older adults.
This report includes an overview regarding surveillance using existing public health databases. This overview describes some problems and potential solutions that are important for disease surveillance among older adults. A more detailed discussion is beyond the scope of this publication and has been addressed elsewhere (9). In the second report, topics regarding older adults that are explored include a) leading causes of both hospitalization and death, b) prevalence estimates for leading chronic diseases, and c) economic burden of morbidity. The third report describes injury and violence among older adults. In the fourth report, the prevalence of several types of barriers that can block access to health care, including financial and structural (i.e., physical) barriers, are highlighted. The fifth report presents prevalence estimates for five health risks: a) overweight, b) drinking and driving, c) inadequate fruit and vegetable consumption, d) physical inactivity, and e) smoking. Three dimensions of health status are examined in the sixth report: a) sensory impairment, b) activity limitation, and c) health-related quality of life.
Representativeness and Sample Size
Much of the research on older adults in the United States comes from the secondary analysis of data that have been developed for other purposes (3). In the surveillance summaries in this publication, surveys sampled from the general population (e.g., the National Health Interview Survey [NHIS] and the Behavioral Risk Factor Surveillance System [BRFSS]) and data from administrative records (e.g., death records, hospital discharge records, or insurance claims records) were used extensively. When these kinds of data are used, two methodological questions must be addressed: "Is the dataset representative of all older persons in the United States?" and "Are sample sizes large enough for meaningful analyses?"
Defining the appropriate population and a sampling frame that accurately represents the study population is a critical element of public health surveillance (10). The sampling frames of household surveys (e.g., NHIS and BRFSS) consist of noninstitutionalized persons and exclude institutionalized persons. Approximately 5% of persons aged greater than or equal to 65 years are nursing home residents and therefore are not included in the sampling frames of household surveys; among persons aged greater than or equal to 85 years, more than 20% are nursing home residents. Functional impairment and comorbidity are major risk factors for institutionalization and are directly related to advanced age. Therefore, estimates of the occurrence of chronic conditions and activity limitations in the noninstitutionalized population of older adults might underestimate the occurrence in the entire elderly population. Several of the analyses in this publication are limited to the adult noninstitutionalized population.
Older adults who are substantially impaired, regardless of residence, are more likely to make errors in survey responses, less likely to respond to specific items, and might be generally less likely to participate in surveys, compared to nonimpaired older adults (11,12). Several reasons have been offered, including low levels of comprehension and concentration; fear of interacting with strangers; sensory or cognitive impairment; other health problems that might make some older adults reluctant to participate; or gatekeeping activities by household members, which might limit interviewers' access to older adults (13,14,15). Regardless of the reasons, nonresponse and incomplete response are problems in research focused on health status (16). Failure to retain the oldest old, the physically impaired, and the cognitively impaired in surveillance samples reduces the accuracy of population estimates and reduces the ability to study these important target groups. Techniques such as mixed modes of data collection and reliance on proxy reporters (as was done in the National Health Interview Survey and Longitudinal Survey on Aging) have been suggested as techniques to maximize sample coverage (16).
A related issue is the geographic representativeness of data. Most available datasets are designed to be nationally or regionally representative but are not locally representative. Adequate state-specific or local data on the health status and service needs of older persons are often not available. Health planners use national findings as proxies for the needs of older persons within local areas. Of the datasets used in these surveillance summaries, only the BRFSS was designed to provide state-specific population estimates. The analyses demonstrate the usefulness of BRFSS for aging surveillance and programmatic decision-making at the state level.
All the analyses reported in these surveillance summaries examined both findings for all older persons and findings for important subgroups of elders. A limitation of datasets designed to be representative of the general population is that subgroups of elders of particular interest might be sampled with few persons (i.e., have small numbers) (13). When numbers are small, estimates can be made for subgroups of particular interest, but the standard errors for those estimates are so large that they are unreliable, difficult to interpret, and of limited use for understanding the characteristics of the subgroup in the underlying population. For example, even in a large random sample of the U.S. population in which no special provisions were made to oversample the elderly, approximately 13% of the respondents will be aged greater than or equal to 65 years and only 5% will be aged greater than or equal to 75 years (14). Although the actual numbers will depend on these proportions, the size of the total sample, and the sampling scheme, small numbers of persons are likely to fall into specific subgroups of older persons (e.g., persons aged greater than or equal to 75 years, by race; male smokers aged greater than or equal to 65 years; or all persons aged greater than or equal to 85 years). When the sample is limited to a geographic area or when an event of interest is rare (e.g., drinking and driving among persons aged greater than or equal to 85 years), the problem is magnified.
Several techniques for managing these difficulties were used in this publication. Multiple years of data were combined in some studies to yield larger sample sizes. In studies using BRFSS, data from multiple states were combined to provide "large enough" samples of particular subgroups (e.g., racial groups other than black and white, and persons of Hispanic origin). In other analyses, examination of racial subgroups was limited to black and white, so that there would be adequate sample size. Finally, in each of these analyses, findings for any subgroup that generated estimates with a relative standard error of greater than or equal to 30% were not reported.
As the need for more and better information about the aging population grows, an increase in special studies (e.g., the Medicare Current Beneficiary Survey) in which the sampling frames are composed of a substantially high proportion of older persons is likely. Alternatively, more studies might include oversamples of older respondents. Although a complete discussion of oversampling is beyond the scope of this paper, oversampling is not without problems, including increased costs and the need for special analytic techniques.
Aging Issues in Public Health
Although public health programs have always served older adults, the elderly have not been identified traditionally as a primary target for public health (22). Nonetheless, many public health professionals have had substantial roles in gerontologic and geriatric research and practice during the preceding 40 years. The emerging emphasis on aging issues in public health is the result of the convergence of several forces including
Since 1965 and the passage of the Older Americans Act (OAA), OAA-funded state and local agencies (the Aging Network) have provided a variety of services for older adults. In 1992, OAA began funding organizations that promote healthy aging and disease prevention activities related to health screening, physical activity, chronic disabling conditions, and nutritional screening (22). Public health agencies are joining an existing network of organizations that are committed to promoting healthy aging. Both public health agencies and the Aging Network have a growing need for information on the health status of older adults, including measures of activity limitations, sensory impairments, quality of life, and the availability and use of preventive health services. Both groups have data that might be valuable to each other but have not been generally shared or coordinated (23).
Surveillance is an important contribution of public health. Current surveillance activities can be modified to ensure adequate coverage for special target groups (e.g., the very old, ethnic minorities, groups for whom particular preventive services might be especially important [e.g., women aged greater than or equal to 65 years who have not had a recent mammogram], and impaired older adults who live in the community). These surveillance activities should provide guidance and feedback for programs at the Federal, state, and local levels. Health surveillance should be expanded so that all available data can be used in an integrated, coordinated manner to more effectively guide disease prevention and health promotion for older adults in the United States.
Disclaimer All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.**Questions or messages regarding errors in formatting should be addressed to firstname.lastname@example.org.
Page converted: 12/14/1999
This page last reviewed 5/2/01