tors among subpopulations for whom the index is low (e.g., the very old or minorities).
Behavioral research: (1) Study the psychological concomitants of illness and how these affect self-care and response to formal care; (2) study the comparative effectiveness of different modalities in the treatment of chronic mental disorders; and (3) examine the effect of behavioral and social intervention on the outcome of long-term illness.
The committee suggests that three methodological approaches be used to carry out the new research agenda in behavioral and social investigation. These are as follows: (1) refined measurement and analytic instruments such as tests, scales, and models for forecasting or analysis; (2) cooperative multidisciplinary large-scale investigations utilizing multicultural, diverse populations studied over time and cohort sequentially; and (3) studies of intervention techniques utilizing the findings of the foregoing research.
Examples of new and improved measurements include (1) measures of functioning that build on current activities of living scales; (2) psychological and other behavioral assessment techniques that are age fair and culturally fair, such as tests of skills, abilities, knowledge, and capacities appropriate for older people and people in different population subgroups; (3) measures of social contexts such as home, workplace, and community; (4) more accurate ways to measure behavioral and social components of illnesses such as Alzheimer 's disease; (5) improved techniques for longitudinal and cohort analysis; and (6) modeling techniques to facilitate trend prediction.
Although multidisciplinary research is difficult and costly, it is the only way to address many of the central issues of interrelationships and interactions among the variables and processes discussed above. These issues include demographic and epidemiological matters related to the postponement of morbidity, antecedents and consequences of health differences and sequencing among different groups, dimensions of caregiving (including effectiveness, duration,