No abstract

No abstract

No abstract

In an aging population, potentially modifiable factors impacting mortality such as diet quality, body mass index (BMI), and health-related quality of life (HRQOL) are of interest. Surviving members of the Geisinger Rural Aging Study (GRAS) (n = 5,993; aged ?74 years) were contacted in the fall of 2009. Participants in the present study were the 2,995 (1,267 male, 1,728 female; mean age 81.4 ± 4.4 years) who completed dietary and demographic questionnaires and were enrolled in the Geisinger Health Plan over follow-up (mean = 3.1 years). Cox proportional hazards multivariate regression models were used to examine the associations between all-cause mortality and BMI, diet quality, and HRQOL. Compared to GRAS participants with BMIs in the normal range, a BMI < 18.5 was associated with increased mortality (HR 1.85 95%CI 1.09, 3.14, P = 0.02), while a BMI of 25-29.9 was associated with decreased risk of mortality (HR 0.71 95%CI 0.55, 0.91, P =0.007). Poor diet quality increased risk for mortality (HR 1.53 95%CI 1.06, 2.22, P = 0.02). Finally, favorable health-related quality of life was inversely associated with mortality (HR 0.09 95%CI 0.06, 0.13, P < 0.0001). Higher diet quality and HALex scores, and overweight status, were associated with reduced all-cause mortality in a cohort of advanced age. While underweight (BMI < 18.5) increased risk of all-cause mortality, no association was found between obesity and all-cause mortality in this aged cohort.

The complexity of human behavior demands that research methods be capable of dealing with multivariate, multioccasion, multisubject data if successful explanatory accounts of behavior are to be constructed. When the research focus is on developmental phenomena such as aging, the complexity of the task is even greater because of the difficulties of modeling and accounting for systematic changes in behavior. Proper decisions about which research methods to use rest on four principal concerns: (1) general orientation of the research; (2) theoretical assumptions concerning the nature of the phenomenon being studied; (3) data collection strategy; and (4) data analysis tactics. Replicated, multivariate, single-subject research designs such as P-technique involve assessment with multiple variables at each of many times of measurement. The resulting data, which are analyzed to determine the nature of occasion-to-occasion changes in the variables, can inform about both covariation patterns and level. Such designs have not been exploited by researchers in aging even though concern with a variety of intraindividual changes in older adulthood is evident in the literature. The rationale of such designs and their potential utility for the study of adult development and aging are examined and discussed.