Increasing physical exercise among older adults: the effects of information, peer-modelling, personalized planning, and commitment-making

Abstract

Increasing physical activity among individuals in the U.S.A. is a primary health concern. Sedentary life-style is a risk factor for a number of diseases. Life expectancy in the U.S. has increased dramatically over the past several decades, yet frequently, few of these additional years remain disease free. It is now known that life-style factors, including exercise,. are important determinants of the number of healthy, functionally independent years remaining for older adults. Unfortunately, few older adults exercise regularly. Interventions aimed at ameliorating this problem are clearly needed.

The Exercise and Older Adults study was designed to assess the effectiveness of an intervention to increase exercise among sedentary and lightly exercising adults ages 55 and older. Unlike other studies which typically involve a supervised aerobic program, this intervention involved the creation of individually tailored exercise programs which participants could maintain without the aid of the researchers. Social learning theory and behavioral principles led to the creation of an intervention combining information, peer modelling, personalized planning, and commitment-making. Subjects came to a single meeting following a three week period of exercise self-monitoring. During the meeting, experimental group subjects were given an exercise information packet, viewed two videotapes about older adults and exercise, created personal exercise plans involving several participation and relapse prevention strategies, and signed a commitment sheet stating that they would try to follow their new exercise plan. Control group subjects received the information packets only. Subjects continued to self-monitor their exercise for six weeks.

Repeated Measures Analyses of Covariance did not support the hypothesis that the intervention received by the experimental group was more effective in increasing exercise than the minimal intervention received by the control group. Hierarchical Regression Analyses did not support the hypothesis that self -efficacy and outcome expectancy predict exercise changes. However, both groups in this study increased their exercise significantly and mean differences between groups at the three week follow-up point were significant for several exercise outcomes. Possible reasons for the failure to find statistical significance across time between groups is discussed and future research directions are outlined.