Persistence of personal health practices over a 1-year period.

  • 1 September 1987
    • journal article
    • research article
    • Vol. 102  (5) , 483-93
Abstract
A large amount of research has been devoted to identifying the psychosocial and demographic correlates of personal preventive health practices. An additional factor to consider, however, is the stability of personal health practices over time. At least over short periods, the prediction of current behavior may be substantially improved by having information about an individual's previous performance of the practice being studied. To address this question, data from Wave 1 (1979) and Wave 2 (1980) of the National Survey of Personal Health Practices were examined. Using nine health practices as indices, performance reported at Wave 1 was used to predict performance of that same practice as reported at Wave 2, 1 year later. A two-step analysis strategy was followed to estimate how much more variance could be explained when the behavioral reports were added to a list of psychosocial and demographic predictors. Results showed that over the 1-year interval, the Wave 1 behavioral reports were by far the strongest predictors of their corresponding measures obtained at Wave 2. The explained variance was increased substantially for most of the nine health practices, suggesting a strong tendency for persistence of the practices. Psychosocial and demographic variables tended to account for much smaller amounts of variance and often dropped out of the prediction equation when the Wave 1 behavior report was entered in the second step of analysis. Health practices other than the corresponding Wave 1 index did not improve prediction of the Wave 2 index greatly. The tendency for previous behavior to persist may overshadow the influence of other factors and account for the modest amounts of explained variance that are usually found for psychosocial and demographic indices in cross-sectional surveys. Although stability does not imply rigidity or impossibility of change, the strength of prediction found in these data attest to the "force of habit" that community interventions can encounter.