Objective: Reciprocal relationships between positive affect (PA) and health are now subject of a heuristic debate in psychology and behavioral medicine. Two radically opposed approaches address the link between subjective well being (SWB) and physical health: top-down (i.e., psychosomatic hypothesis) and bottom-up (i.e., disability/ability hypothesis) approaches. The aim of the present study was to test these two approaches by investigating thirteen-year longitudinal relationships between PA, as an affective dimension of SWB, and functional health in older people.
Methods: The study included 3754 participants aged 62-101 years assessed 6 times over a thirteen-year period. PA was measured by the mean of the positive affect subscale of the CES-D scale. Functional health was assessed by four composite items: a single-item self-rating of hearing impairment, a single-item self-rating of vision impairment, the number of medically prescribed drugs, and a single-item self-rating of dyspnoea. We used cross-lagged modeling with latent variables, which is appropriate for testing specific theories. Mean arterial pressure, diabetes mellitus and hypercholesterolemia status, sequelae of stroke, gender, level of education, and age at baseline were use as control variables in the models.
Results: Results indicated that good health significantly predicted subsequent levels of PA (average β = -0.58, p < 0.001), but PA did not predict subsequent levels of good health (β = 0.01, ns).
Conclusion: This finding, obtained from a sample of older people, is in keeping with the bottom-up approach, and supports the popular adage "As long as you've got your health". Limitations of this finding are reviewed and discussed. Models including longitudinal mediators, such as biomarkers and life style patterns, are needed to clarify the nature of the link between these constructs.
Keywords: Cross-lagged analysis; Functional health; Later life; Longitudinal; PAQUID cohort; Positive affect.
Copyright © 2015 Elsevier Ltd. All rights reserved.