Americans made 560 million visits to primary care physicians in 2010 (1). Most visits were for prevention and treatment of chronic conditions for which physical activity counseling would be appropriate. Even modestly effective evidence-based physical activity interventions in primary care settings could have powerful health benefits for patients (2).

Yet clinicians face significant barriers to implementing evidence-based interventions, especially for underserved populations that have limited resources and competing demands (3). The implications of clinicians not using evidence-based interventions include worsening health disparities related to inadequate physical activity.

Some barriers to implementing evidence-based interventions could be addressed by the “5As,” an evidence-based clinical counseling framework in which clinicians are encouraged to Ask about (or Assess), Advise on, Agree upon, Assist with, and Arrange follow-up on patients’ behavior-change efforts (4–11). The 5As can be used briefly and during multiple visits (12) and have been endorsed by the US Preventive Services Task Force (13) and others (14–16) and more recently for obesity counseling (17,18).

Despite recommendations to use the 5As, evidence for the feasibility of implementing the 5As in “real-world” settings is limited. In 2010, the US Preventive Services Task Force updated its recommendations on physical activity counseling (19), concluding that the evidence for low-intensity (30 minutes or less) counseling interventions was mixed and that further study is needed to evaluate them. Low-intensity counseling interventions are especially important because they are more likely than high-intensity interventions to be implemented in primary care settings.

The objective of this study was to examine the effectiveness of a clinician-targeted intervention that used the 5As approach to physical activity counseling in a medically underserved patient population.

The study design and protocol are described in detail elsewhere (20). Briefly, the study used a 2-group pragmatic pilot randomized-controlled–trial design; 13 clinicians were randomized to receive the intervention; patients (n= 319) were the primary unit of analysis. For practical and logistical reasons, the clinicians were randomized so that Group 1 clinicians (n = 6) participated in the intervention first; Group 2 (n = 7) participated in the intervention 8 months later. The intervention consisted of a clinician training program designed to increase use of the 5As for physical activity counseling with patients. Outcomes were patient and clinician assessments of the effectiveness of the intervention in increasing use of the 5As for physical activity counseling. Other than the timing, there was no difference in the intervention or the assessment for the 2 groups. The University of Rochester Review Board approved the study protocol. The intervention period started in June 2009 and ended with the final follow-up data collection in October 2011.

Three-quarters of clinicians were women; 66% were white, 25% African American, and 9% Asian. The average age of clinicians was 49 years (range, 31–73 y), and the average length of practice was 15 years (range, 2–33 y). Group 1 and Group 2 patients did not differ significantly by any sociodemographic or health characteristic, so data for both groups were combined (Table 1). Patients’ mean age was 43 years, and 69.9% were African American. Most (61.4%) had public insurance, and the mean body mass index (BMI) was 32.3. The most common weight-related conditions were hypertension (48.5%), chronic pain (42.3%), depression (32.4%), and diabetes (21.3%); 70% of patients had more than 1 weight-related chronic condition, and 51% had 3 or more conditions.

In the mixed model controlling for clinician as a random effect, the PAEI score for both groups combined increased from 6.9 at baseline to 8.6 at postintervention (P = .01) (Table 2). The medium effect size(0.34)was not sustained at 6-month follow-up (mean PAEI score at 6 months, 8.2; P = .09). The average PAEI score did not change significantly at either point for Group 1, whereas the average PAEI score increased at both points (from baseline) for Group 2 (Table 2). The PAEI score at baseline for Group 1 (7.5) was higher than the score at baseline for Group 2 (6.3) because Group 1 patients rated their clinicians higher on Assess and Assist items. However, at 6-month follow-up, Group 2 patients rated their clinicians higher on all PAEI items than did Group 1 patients, especially for Assess and Assist (Table 3). Of the 3 PAEI items added for this intervention, the scores for 2 items increased (Table 3).

From baseline to 6-month follow-up, clinician confidence increased significantly for assessing an exercise history, negotiating an exercise plan, turning setbacks into learning, helping patients cope with barriers, counseling in a cost-effective way, and having knowledge of community resources that could meet patients’ needs (Table 4).

This study evaluated the effect of a clinician-targeted intervention on patient- and clinician-reported use of the 5As for physical activity counseling. The intervention used innovative, interactive clinician training techniques and focused exclusively on a medically underserved population not well represented in this type of research. The main finding was that the intervention increased patient-reported improvements in 5As counseling on physical activity immediately postintervention but not at 6-month follow-up, although this improvement was primarily due to higher scores for Group 2 clinicians. Group 1 clinicians had a higher baseline PAEI score, which may have limited the effectiveness of the intervention for them. Also, the community exercise program may have provided a greater incentive for Group 2 clinicians to use the 5As because the program was better established for them than for Group l clinicians.

There are some practical lessons learned from this study. First, the development of the referral process for the community exercise program took time to function effectively, and this delay affected clinician behavior (ie, familiarity with and ease of referral to the program). Research on the effectiveness of community exercise programs for clinician referrals is lacking; others have commented on the potential demand and the difficulty of establishing effective partnerships for referral. A recent meta-analysis of primary care–based physical activity counseling (26) and recent US Preventive Services Task Force recommendations (27) identified evaluation of referral programs as a major gap in knowledge that requires future study.

Second, the intervention might have been more effective in increasing clinicians’ use of the 5As if it had included strategies such as booster trainings, reminders or prompts about the ongoing availability of the community exercise program, and additional staff support to help with problem solving (28). Consistent with their patients’ aggregate ratings, Group 1 clinicians reported challenges in problem solving with their patients; problem solving is necessary to accomplish Assess and Assist and may be related to a clinician quality that needs further study.

This study has several limitations. Although clinicians were randomized for practical reasons, we had inadequate control group data and therefore used a pre–post analysis. Although the patient sample size was larger than is typically reported for this type of research, patients were nested within a small sample of clinicians from a single geographic location. Participants’ awareness of the study topic and objective could have increased the likelihood of patient bias to overreport physical activity counseling. However, the blinding of patients to time point and participation of clinicians in the intervention increased the likelihood that any bias was evenly distributed among time points and clinicians. Finally, by design, this study did not assess patients’ behavior change during the study; instead, it aggregated patient ratings nested by clinician. This trade-off was made for practical and logistical reasons because of the focus on an underserved population, but future research would be enhanced by longitudinal assessments of changes in physical activity among patients.

Strengths of the study are that, to our knowledge, this is the first study to assess the effectiveness of a clinician intervention on use of the 5As for changing physical activity counseling in a medically underserved population. This study is relevant for several reasons. First, it addressed the high prevalence of lifestyle-related chronic conditions in a medically underserved population and the need to eliminate racial/ethnic disparities (20). Second, the intervention was interactive and innovative and had multiple levels. Although it focused on clinicians, it also created tools in the electronic health record for use by the entire clinic, and it initiated a partnership with a community exercise program. Third, this study represents an evaluation of an intervention aimed at translating guidelines into everyday practice, which is not commonly reported in the literature.

This study has 2 main clinical implications. First, clinicians can be taught to improve their physical activity counseling by using the 5As framework. Second, the intervention can improve confidence in counseling skills among physicians, especially by educating them about community resources for physical activity.

This study also has policy relevance. Primary care is undergoing transformation in the United States. The patient-centered medical home (PCMH) initiative has created a resurgence of interest in helping primary care patients change health behaviors (29) to qualify for higher insurance reimbursements (30). The PCMH standards from the National Committee for Quality Assurance include key (“must-pass”) elements for practices to “support self-care” in part by providing educational resources, self-management tools, and counseling to adopt healthy behaviors for least 50% of patients (31). The changes primary care practices are undertaking are complex, however, and many practices struggle with how to best meet the PCMH standards. This study represents 1 strategy to offer training and tools to help clinicians translate evidence-based guidelines into practice and to address the PCMH-relevant goal of helping patients adopt healthy behaviors.

The results of a clinician-directed intervention designed to increase patient reports of physical activity counseling were mixed. Group 2 clinicians (who took part in the intervention after Group 1) increased their use of the 5As more than did Group 1; the difference in use between the 2 groups was due primarily to greater use of Assess and Assist skills among Group 2. Group 2 clinicians improved their awareness of the community exercise program for referral, whereas Group 1 clinicians reported difficulty with problem-solving skills. Future directions are to explore the association between 5As counseling and patient outcomes such as enrollment and participation in community exercise programs.