Objective The evidence regarding the effectiveness of promoting physical activity (PA) in primary care is varied. The present study systematically reviews the literature pertaining to primary care providers' perceptions about PA counselling to identify the barriers and enablers to PA counselling in clinical practice.
Design A systematic literature review (through 2011) of quantitative and qualitative studies was conducted. Articles were included in the review if the study population consisted of primary care providers and the study evaluated providers' attitudes and perceptions pertaining to PA counselling.
Results Nineteen articles met the inclusion criteria. Most primary care providers believe PA counselling is important and that they have a role in promoting PA among their patients. However, providers are uncertain about the effectiveness of counselling, feel uncomfortable providing detailed advice about PA, and cite lack of time, training and reimbursement as barriers. Providers are more likely to counsel their patients about PA if they are active themselves, or if they feel their patients' medical condition would benefit from a lifestyle change.
Conclusion Primary care providers are receptive to the notion of PA promotion in the clinical setting, yet numerous individual and organisational barriers need to be addressed to integrate PA counselling into primary care effectively.
Statistics from Altmetric.com
Insufficient physical activity (PA) is a serious public health concern, with multiple studies demonstrating an association between higher levels of PA and reduced morbidity and mortality.1 ,2 Increased levels of PA have been found to decrease the risk for premature mortality as well as numerous chronic diseases such as cardiovascular disease, diabetes, hypertension, osteoporosis, depression and several types of cancer.3 PA has been shown to improve metabolism, lower blood pressure and increase high density lipoprotein levels in the blood.4 In addition, regular PA can help strengthen bones and muscles, and prevent falls in older people.5 To achieve these health benefits, the US Department of Health and Human Services and the UK Department of Health recommend that adults obtain at least 150 min of moderate-intensity PA or 75 min of vigorous-intensity PA a week; however, reports show that less than half of adults in the USA and the UK meet these guidelines.6 ,7
Healthy People 2020, the 10-year health agenda released by the US Department of Health and Human Services, identifies PA as a leading health indicator. There are 15 health objectives related to PA, the first of which is reducing the proportion of adults who engage in no leisure-time PA by 10% from 36.2%. Other objectives include increasing the number of adults who meet the current aerobic PA guidelines of 150 min per week and increasing the proportion of adults who engage in muscle strengthening activities.6
Because of the serious medical consequences of physical inactivity, a number of agencies have suggested PA counselling, which typically takes the form of advice and discussion to encourage patients to increase and maintain PA. Healthy People 2020 recommends that physicians provide PA counselling to patients with chronic conditions (eg, cardiovascular disease and diabetes).6 Likewise, the National Institute for Clinical Excellence in the UK recommends that primary care practitioners should identify inactive adults and advise them to aim for 30 min of moderate activity on 5 days of the week (or more).8 The basis for these recommendations is due to the health benefits associated with regular PA, rather than the evidence on the effectiveness of PA counselling. Recent studies suggest that there may be some short-term benefits to counselling, but the long-term impact remains uncertain. The most recent review for the US Preventive Services Task force by Lin et al demonstrates that medium to high-intensity counselling efforts were associated with moderate, short-term increases in patient-reported PA; however, the efficacy of low-intensity counselling (defined as 30 or fewer minutes of contact) efforts varied. Counselling interventions with sustained results after a year were high-intensity efforts characterised by repeated follow-up contact with the patient.9
Although the health benefits of PA are well-established, the evidence regarding the long-term effectiveness of PA counselling by primary care providers is mixed. Many studies have investigated primary care providers' attitudes on PA counselling in a clinical setting, but to date, no study has systematically summarised and evaluated their results. Thus, the primary objective of this review is to systematically summarise and evaluate primary care providers' perceptions and attitudes about PA counselling. More specifically, the review will evaluate providers' attitudes on the importance of PA counselling, whether or not they feel providers have a role in PA counselling, their confidence in counselling patients, the effectiveness of PA counselling on changing patient behaviour, and barriers and enablers to providing PA counselling in the primary care setting.
We conducted a systematic review of quantitative observational studies (cross sectional, longitudinal) and qualitative studies examining primary care providers' attitudes and perceptions related to factors affecting their PA counselling practices. A systematic literature search of relevant articles was conducted using four databases: MEDLINE (1948 to 5/2011), PsycInfo (1806 to 5/2011), CINAHL Plus with Full Text (1981 to 5/2011) and the Cochrane Library. The MeSH terms included, ‘primary health care,’ ‘attitude of health personnel,’ ‘counselling,’ ‘health promotion,’ and ‘exercise.’ Using these terms as a starting point, the ‘explode’ command was used to include the most inclusive version of each subject heading. Then, other variations of each term were included in the search using ‘OR’ to obtain the maximum number of potentially related articles. Once all potential terms were identified, four categories of keywords emerged: primary care providers, exercise and PA, counselling, and attitudes and opinions of care providers. These four categories were combined using ‘AND’, and once several pertinent articles were found, the ‘find similar’ function and a hand search of the references of relevant articles were also used to search for additional results. All results were limited to the English language.
A total of 110 potential articles were identified from the search. Studies were included if (1) the study population consisted of primary care providers (ie, general practice physicians, physician assistants, nurses and nurse practitioners), and (2) the study evaluated providers' attitudes and perceptions pertaining to PA counselling. Both quantitative observational studies (eg, cross sectional, longitudinal), and qualitative or mixed methods studies were considered. Articles were excluded if (1) they were intervention studies (eg, randomised controlled trials or quasi-experiments) which evaluated the impact of a specific intervention, since the primary aim of the study was to examine general perceptions of providers pertaining to PA counselling, irrespective of a specific intervention strategy; and (2) perceptions of PA counselling were not specifically discussed—for example, lifestyle counselling was discussed without specifically focusing on PA counselling.
Abstracts of the 110 potential articles were carefully assessed to evaluate whether or not they met the inclusion criteria. Based on the initial review, 86 articles were excluded based on an abstract review, leaving 24 potential articles for a review of the full text (figure 1). Two reviewers independently examined hard copies of the final group of articles. After the full text review, an additional five articles were excluded, leaving a total of 19 articles included in the review.
Data synthesis and analysis
Data were extracted into a standardised form developed and tested by two authors. The extraction form included author, study year, practice setting, type, and location, study aim, study design, type of provider, provider characteristics (including age, gender and training), patient characteristics, sample size and response rate, sampling method and survey instruments. In addition, the main independent and dependent variables, covariates, the method of analysis (eg, bivariate, multivariate or qualitative), and main results of the study (both descriptive and analytic) were also extracted. Once the primary variables were extracted, specific results related to the review's aim were organised into tables that summarise relevant barriers to PA counselling, enablers to PA counselling, discussion of the importance of PA counselling, providers' role in PA counselling, perceived effectiveness of PA counselling and providers' confidence in providing counselling. Because most of the studies were descriptive and the analytical studies vary widely in terms of measurement methods and variables analysed, a narrative review of the results was conducted.
Study characteristics of the 19 included studies were summarised, including study design, sample size and response rate, sampling method, provider characteristics and major results (see appendix). All included studies were cross sectional, the majority of which were quantitative studies that used written questionnaires to obtain descriptive and numerical data regarding providers' attitudes and practices related to PA counselling. Only one study was purely qualitative,10 and two were mixed methods that included semistructured interviews or focus groups in addition to a written questionnaire.11 ,12 Five of the 19 studies evaluated the opinions of providers practicing in the USA13,–,17; the others were conducted in Canada, Europe and Australia. The providers in the included studies consisted primarily of physicians practicing in a primary care setting; however, seven studies investigated the attitudes of primary care nurses or nurse practitioners.12 ,14 ,18,–,22 Sampling methods in the included studies varied, and only five used random sampling. In all studies, the survey instrument was designed by the authors; three were pilot-tested, and only one provided information about reliability and validity. The survey instruments in the quantitative study consisted primarily of closed-ended questions with Likert scale style measurements, and the three studies with qualitative components used semistructured interviews and/or focus groups.
Barriers and enablers to PA counselling
Table 1 summarises the major barriers to PA counselling cited by providers. The most common barrier identified was lack of time, followed by lack of knowledge or training in PA counselling, and lack of success in changing patient behaviour. Other impediments included lack of financial incentive or reimbursement for time spent on counselling, lack of counselling protocols and organisational barriers. In several studies, providers asserted the belief that PA counselling was not a priority or that it was not relevant to the patient consultation.10 ,11 ,14 ,16 ,17 ,23 ,24 In the three studies that evaluated the opinions of both physicians and nurses, doctors were more likely to agree that the described barriers were a major impediment to counselling than were nurses. Barriers were shown to have a clear effect on counselling practice in several studies. In the work of Cho et al,25 physicians who felt unprepared to counsel gave advice less often (OR=5.05, 95% CI 2.91 to 8.76), and the physician's perception of exercise as an important health factor was positively associated with providing advice (OR=1.87, 95% CI 1.04 to 3.35). Sherman et al16 found that the most important predictor of counselling was perceived success at getting patients to start exercising (OR=22.83, 95% CI 8.36 to 62.31). Similarly, Walsh et al17 found that physicians who felt they were successful in changing patients' behaviour were more likely to ask than those who felt unsuccessful, and that physicians with adequate knowledge about PA were more likely to ask patients about PA than those who did not.
Although the focus in most studies on providers' attitudes was directed towards the impediments to counselling, several studies discussed enablers to PA counselling. Beyond the elimination of barriers such as lack of training, there were two primary enablers to counselling. First, Bize et al10 found that providers were more likely to provide counselling if the patient had cardiovascular disease symptoms. Douglas et al18 also reported that physicians and nurses were more inclined to consistently provide PA counselling to patients who were obese or had hypertension. Second, providers were found to be more likely to counsel patients about PA if they were active themselves. In the work of McDowell et al,20 nurses who were active themselves rated lack of resources, protocols and success as less limiting to PA promotion compared with nurses who were not active. Likewise, Abramson et al, Reed et al and Sherman and Hershman found that physically active physicians were more likely to encourage patients to be active than their sedentary counterparts.13 ,15 ,16
Importance of PA promotion
Providers' opinions on the importance of PA counselling were specifically discussed in eight of the studies (table 2). Both physicians and nurses overwhelmingly felt that PA was important, with between 61% and 99% of providers agreeing or strongly agreeing that PA should be promoted in primary care setting. In one study, physicians felt that other types of lifestyle counselling, such as smoking cessation, were more important than PA counselling.11 In another study, where 87% of physicians believed that PA was valuable for all patients, only 51% believed that it was important for all patients to receive counselling.15
Providers' role in PA counselling
The role of primary care providers in PA counselling was examined in six studies (table 3). The majority of providers, both physicians and nurses, felt that PA counselling was one of their responsibilities. In the two studies that used repeated cross-sectional surveys, physicians were asked if they had a responsibility to discuss the benefits of PA with patients and suggest to patients ways to increase PA. In both studies, the percentage of physicians agreeing with these statements increased in the later surveys. Although providers generally agreed that they had a role in counselling patients, over half of physicians and nurses in one study felt that nurses rather than physicians were the most appropriate care giver to provide health promotion to patients.22
Perceived effectiveness of PA counselling
Eight studies investigated providers' perceptions of the effectiveness of PA counselling (table 4). Providers' opinions were varied. Providers felt that counselling was ineffective in changing patient behaviour in five of the studies, with less than half of providers agreeing that it was successful.10 ,15 ,17 ,22 ,23 In three studies, over half of providers felt that PA counselling was effective or very effective.19 ,24 ,25 According to Sherman and Hershman,16 providers' attitudes were stratified by the percentage of patients whom they regularly counselled. In the group of physicians who counselled more than 75% of their patients, 71% felt that counselling was moderately or very successful; yet in the group of physicians who counselled 25% or fewer of their patients, only 10% felt that counselling was successful.
Providers' confidence in promoting PA
Nine studies investigated provider's confidence in promoting PA in the primary care setting (table 5). Between 48.2% and 92% of providers in each study agreed that they felt confident or very confident in their abilities to provide PA counselling. Three studies distinguished between providing general advice or specific advice about PA.26,–,28 In each of these studies, physicians were significantly less confident in dispensing more detailed PA counselling than they were in providing general advice.
This review is the first to systematically summarise providers' perceptions and attitudes towards physical activity counselling in the primary care setting, which provides insight into the receptiveness of primary care providers to act as PA promoters in their clinical practices. Although previous reviews have examined the effectiveness of primary care providers in promoting PA, these have primarily been systematic reviews of randomised controlled trials and have not assessed impediments to PA counselling irrespective of an organised intervention. The evidence on the efficacy of PA interventions delivered in the primary care setting has been largely inconclusive, although medium to high-intensity counselling efforts have been associated with moderate, short-term increases in patient-reported PA.9 ,29 Given the abundance of evidence regarding the health benefits of PA, providing counselling on PA is still a goal for health professionals, a fact that is reflected in the newest objectives in Healthy People 2020.6 Currently, however, it is estimated that only 30–50% of physicians in the USA regularly provide counselling on PA to their patients, so it is important to examine their attitudes and perceptions about counselling, and some of the reasons that they may or may not counsel their patients.30
The current review suggests that most primary care providers agree that PA counselling is important and that they have a role in promoting PA to their patients. Despite this fact, providers are uncertain about the effectiveness of counselling; feel only marginally comfortable providing more than general advice about PA, and cite major barriers to counselling such as lack of time, lack of training and lack of reimbursement for their counselling efforts. The evidence in this review suggests that beyond these barriers, providers are more likely to counsel their patients about PA if they are active themselves, or if they feel that their patients' condition, such as cardiovascular disease or obesity, would strongly benefit from a lifestyle change. Physicians and nurses tend to perceive similar barriers to PA counselling; however, fewer nurses than physicians rate these barriers as major impediments to counselling, and fewer nurses feel that they do not have time to incorporate counselling into their patient visit. The included studies represented the opinions of primary care providers from seven different countries, many of whose healthcare systems vary drastically. Regardless of the differences between the countries, the opinions of healthcare providers regarding PA counselling did not appear to differ significantly between locations.
Although no other systematic review has examined the broad opinions of providers on PA counselling, the results of this review regarding barriers to counselling are similar to those found by Eakin et al in their 2005 study which evaluated the extent to which the current literature on PA counselling informs new interventions.31 Eakin et al sought descriptive studies of barriers to the delivery of PA counselling in primary care; however, no specific inclusion or exclusion criteria were described. The results obtained by Eakin et al were consistent with the present review. Lack of time was the most common barrier identified, followed by lack of patient interest and willingness to adhere to advice, lack of reimbursement and lack of practitioner knowledge. The evidence in the current review suggests that primary care providers feel they have a role as health promoters in the primary care setting, and that they recognise the importance of PA counselling. Many physicians cited lack of training or lack of knowledge as a primary barrier to their counselling efforts, and given the importance of preventive care, curriculum changes in medical school to incorporate training for health promotion counselling should be considered. In a 2011 cross-sectional survey of 129 US medical schools, only about 13% of medical schools provided instruction on the health benefits of PA, 87% of the schools did not offer any kind of PA curriculum, and 76% of the schools stated that there were no plans to implement such a curriculum in the future.32 The development of training programmes for medical school students, residents and other healthcare professionals faces several barriers, including a lack of existing educational materials, lack of space in the curriculum and a lack of faculty members trained in PA counselling.33 While it may be difficult to incorporate new material into the medical school curriculum, PA counselling training could be included through web-based courses, discussion of counselling in grand rounds or noon lectures and via mini-clinical exam evaluations.33 It is important to note that while providers reported lack of training as a significant barrier to the provision of PA counselling, the amount of training necessary for providers to feel well prepared or effective is not quantified in their reports. While the introduction of counselling training to the medical school curriculum would be helpful in reducing this barrier, the question of how much training a provider requires to counsel effectively merits additional research.
Lack of time continues to be a major barrier, not just with PA counselling, but with all types of preventive counselling. A 2003 study determined that if physicians were to provide all of the preventive services according to the recommendations of the US Preventive Services Task Force, they would have to spend 4.4 h of each working day providing counselling and preventive services to their patients, an amount of time the authors explain is not feasible in the current medical environment.34 Many counselling interventions have focused on reducing the amount of time necessary to deliver an effective message, and future research may help in alleviating this barrier. Providers cited lack of a counselling protocol as an additional barrier to PA counselling, and it is likely that simply adding a structured counselling protocol would reduce the amount of time necessary to advise a patient. There are a number of basic counselling strategies that could be helpful and effective in the primary care setting, such as the Five A’s (Assess, Advise, Agree, Assist, Arrange) model.35 In the Five A’s model, providers first assess the patient's current PA level, as well as any contraindications to PA that they may have and the patient's readiness for change. Providers then advise the patient by providing a tailored counselling message based on the patient's stage of change, and agree with the patient by collaborating on a plan of action. Finally, providers assist the patient by providing educational materials to help them change, and arrange a follow-up visit to help the patient stay motivated and to evaluate their progress.36 Peterson suggests that inadequate time for counselling can be alleviated by using the Five A’s model in conjunction with organisational changes in the office, such as having the patient fill out activity questionnaires in the waiting room, limiting the counselling session to 2–4 min, and providing written material to supplement the counselling efforts.37 Other counselling strategies might include programmes such as the ‘green prescription’ intervention, in which PA and lifestyle goals are written as a prescription for the patient after a brief discussion with the provider. Elley et al found that such interventions have been effective in increasing patient PA and improving health over 12 months.38
Another possibility may be referrals to counselling professionals for health promotion, or the use of other healthcare workers to deliver counselling instead of or in addition to physicians. The review found that nurses did not feel the burden of too little time as strongly as physicians did, so it may be possible to delegate some of the counselling responsibility while still keeping it as a goal within the primary care appointment. This idea is supported by a 2005 study by Tulloch et al in which they examined the literature on PA counselling interventions in primary care to identify the effectiveness of physicians, nurses, and other allied health professionals such as health educators and exercise physiologists. They found that PA interventions delivered by allied health professionals alone or in conjunction with physicians produced the best long-term results in patient behaviour compared with interventions delivered by physicians alone.39 The reason for this, they explain, may be largely due to the fact that allied health professionals have more time to spend counselling the patient and more extensive training in counselling. Recent studies have suggested that interventions involving technology such as web-based counselling or telephone reminders may be an effective way to deliver PA advice.40 ,41 Appel et al found that participants in a web-based weight loss intervention achieved clinically significant weight loss despite having no face-to-face contact with providers.42 In the intervention, primary care providers reinforced participation through the review of progress reports at routinely scheduled visits. Similar web-based PA interventions have been well received and have had positive effects.43 ,44 These innovative strategies may be an effective way to provide PA counselling without increasing time demands of the primary care providers.
This review is unique in that it presents a systematic examination of primary care providers' perceptions and attitudes pertaining to PA counselling. The analysis is strengthened by several factors. First, the review included both qualitative and quantitative data, which allows for a broader examination of provider opinions than would quantitative data alone. Second, the review examined the opinions of all types of primary care providers from multiple practice settings and locations, which makes the results generalisable to a wider variety of providers. Third, the review examines the opinions of PA counselling in general, without limitation to a specific intervention protocol, making it applicable to many different types of PA interventions. This review also has several limitations. First, all non-English studies were omitted from review. Second, the included studies vary widely in design, quality and conclusions, making it difficult to present a generalised conclusion of their results. Third, although the included studies represent the opinions of multiple types of primary care providers, few of the studies examine the attitudes and perceptions of nurses and other non-physician primary care providers specifically, so it may be difficult to draw conclusions of their beliefs based on the results in this review. Finally, the survey instruments in all studies differ, and only one questionnaire had been validated. It is unknown, therefore, if the individual results of each study questionnaire are biased in some way.
In summary, this review is the first to systematically summarise providers' perceptions and attitudes towards PA counselling in the primary care setting. Primary care providers feel that they have a role in the promotion of PA, and feel that promoting PA is important; however, numerous barriers need to be addressed to enable providers to incorporate PA counselling into clinical practice effectively. Non-physician members of the primary care team may facilitate PA promotion either independently or in conjunction with physicians.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Any unpublished data, including detailed data extraction tables and literature search strategy can be made available upon request from the corresponding author.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.