Purpose:

To determine whether counseling adults in primary care settings improves and maintains physical activity levels.

Data Sources:

The Cochrane Registry of Controlled Trials, MEDLINE, HealthStar and Best Evidence databases were searched for papers published from 1994 to June 2001.

Study Selection:

We reviewed controlled trials, case-control studies, observational studies, and systematic reviews that reported behavioral outcomes of counseling interventions aimed at increasing physical activity in general primary care populations. For inclusion, a patient's primary care clinician (physician, nurse, nurse practitioner, or physician assistant) had to perform some component (assessment, advising, counseling, referral, etc.) of the intervention, behavioral outcomes (physical activity) were reported, the study was of "good" or "fair" quality using criteria developed by the U.S. Preventive Services Task Force (USPSTF).

Data Extraction:

We abstracted from each study: information on the design and execution; quality information; details of the providers, patients, setting, and counseling intervention; and self-reported physical activity at follow-up.

Data Synthesis:

Nine trials involving 9,227 adults met the inclusion criteria for this report. Most counseling interventions in the studies were relatively brief (3-5 minutes). Two of six fair to good quality trials reported statistically significant improvements in physical activity for intervention patients compared with patients receiving usual care. The remaining three trials compared two or more interventions (contained no usual care comparison). These trials reported an increased effect: when the patient was given advice in combination with a written prescription; for female patients, when the intervention included behavior counseling and extended phone call support; or when the patient (male or female) set a physical activity goal.

Most studies had at least one of the following limitations: provided limited details on the counseling intervention, had follow-up on only 60-79% of subjects, excluded nonresponders from the analysis, studied selected provider populations, reported differences in physical activity levels at baseline between treatment groups, and/or had uncertain or low provider compliance. It was often difficult to assess whether patients had actually received a physical activity behavioral intervention. Most trials only assessed the patients' activity levels short-term (less than six months). These methodological problems made it hard to rigorously assess the efficacy or effectiveness of the interventions. More research is needed to clarify the effect, benefits and/or potential harms of counseling patients in primary care to increase physical activity.

Conclusions:

Evidence that counseling adults in the primary care setting to increase physical activity is inconclusive.