Behavioral Conformity to Patterns of Adapting to Health and Medical Expenditures Effects
Abstract
Sex and age are valid predictors of medical expenditures. Recent research has identified the Patterns of Adapting to Health (PATH) as valid predictors of medical expenditures after controlling for sex and age. However, indirect effects of the PATH on medical expenditures remained unexplored. Attributing the PATH to multiple attractors within a phase space defined by multidimensional health-related behavior, this study explored the direct and indirect effects on population medical expenditures as participant health-related behaviors increasingly conform to these patterns. Sample: A sample of 3,955 adult health plan members aged 18 to 64 was used. Methods: Self-reported health-related behaviors were captured with a short inventory obtained from a mail survey. One year’s total allowed medical claims, age, and sex where drawn from member records. Inventory data were used to assign participants to one of the nine PATH. Computed posterior probabilities of pattern assignment assessed participant’s degree of conformity to all nine of the patterns. Multiple regression analysis was used to decompose pattern conformity direct and indirect effects on total medical claims. Results: Six of nine patterns had positive direct effects on medical claims with increasing conformity to the patterns after controlling for sex and age; other patterns had only indirect effects via sex, via age, or via sex and age. Largest total effects on medical claims were associated with increasing conformity to one pattern with mean per person expenditures comparable to 82-year-old obese females. Conclusions: the PATH effects on population medical expenditures extend beyond direct effects of specific patterns of health-related behavior and include indirect through sex and age.
Sex and age are valid predictors of medical expenditures. Recent research has identified the Patterns of Adapting to Health (PATH) as valid predictors of medical expenditures after controlling for sex and age. However, indirect effects of the PATH on medical expenditures remained unexplored. Attributing the PATH to multiple attractors within a phase space defined by multidimensional health-related behavior, this study explored the direct and indirect effects on population medical expenditures as participant health-related behaviors increasingly conform to these patterns. Sample: A sample of 3,955 adult health plan members aged 18 to 64 was used. Methods: Self-reported health-related behaviors were captured with a short inventory obtained from a mail survey. One year’s total allowed medical claims, age, and sex where drawn from member records. Inventory data were used to assign participants to one of the nine PATH. Computed posterior probabilities of pattern assignment assessed participant’s degree of conformity to all nine of the patterns. Multiple regression analysis was used to decompose pattern conformity direct and indirect effects on total medical claims. Results: Six of nine patterns had positive direct effects on medical claims with increasing conformity to the patterns after controlling for sex and age; other patterns had only indirect effects via sex, via age, or via sex and age. Largest total effects on medical claims were associated with increasing conformity to one pattern with mean per person expenditures comparable to 82-year-old obese females. Conclusions: the PATH effects on population medical expenditures extend beyond direct effects of specific patterns of health-related behavior and include indirect through sex and age.