Researchers
Dena M. Bravata (Principal Investigator) - Stanford University
Vandana Sundaram - Stanford University
Robyn Lewis
John Sirard - Stanford Prevention Research Center
Nancy D. Lin
Crystal Smith-Spangler - Department of Internal Medicine, Stanford School of Medicine
Ingram Olkin - Project Mentor

Summary: Pedometers have become increasingly popular as a tool for motivating and monitoring physical activity, but there is little detailed evidence of their effectiveness. This study will be the first to quantitatively synthesize the literature on the effectiveness of pedometers to change physical activity and health outcomes among the elderly. The results could influence the activity recommendations that healthcare professionals make to their patients.

Objective: The primary purpose of this study is to evaluate the effect of pedometer use on physical activity among elderly adults. We also seek to evaluate the effects of pedometer use on health outcomes (including weight, serum lipids, fasting serum glucose and insulin, and blood pressure) and to evaluate the relationship between number of steps per day and improvements in health outcomes.

Data sources: We will search seven relevant databases with search terms such as pedometer, activity monitor, and step counter. We will review bibliographies of retrieved articles and relevant conference proceedings to obtain additional citations.

Study selection: We will consider studies eligible for this meta-analysis if they include an assessment of pedometer use among adult outpatients of at least 3 days duration and report a change in number of steps per day.

Data extraction: Two investigators will review all titles and abstracts to identify potentially relevant articles and will independently abstract study data. There are four types of variables of interest: information about the intervention (e.g., intervention duration, goal number of steps per day), the subjects studied (e.g., age, gender, baseline activity), the number of steps per day, and the health outcomes (e.g., presence or absence of obesity, diabetes, hypertension, or hyperlipidemia). We will use generalized-least-squares regression to calculate the summary effect of the activity intervention variables on the outcome variables.

Data synthesis: From our preliminary searches, we anticipate that 40 to 50 studies will meet our inclusion criteria and that approximately 75% of the included studies will have a mean age of enrolled subjects who are above age 65. We expect that more than 20 included articles will report change in weight; however, fewer articles are likely to report on the other outcomes of interest. Thus, we expect to be able to assess the question of the threshold of 10,000 steps per day on change in weight, but we may be underpowered to assess this association for the other outcomes of interest.

This study is a seed project for the Center on the Demography and Economics of Health and Aging.

Contact
Dena Bravata

Funding provided by
• Center on the Demography and Economics of Health and Aging (CDEHA)