A review of the use of the 5 A’s model for weight loss counselling: differences between physician practice and patient demand (2024)

Search for other works by this author on:

Abstract

Introduction

Obesity is an epidemic problem in the United States. Approximately 68% of US adults are overweight or obese (1). To compound this problem, obesity has more than doubled in children and tripled in adolescents in the past 30 years (2). Primary care physicians and practitioners (PCPs), including nurse practitioners and physician assistants, are at the forefront of addressing this epidemic and need tools to work effectively with their patients around weight loss.

The 5 A’s is an emerging tool used by PCPs to encourage behaviour change among overweight and obese patients. The use of the 5 A’s as a behavioural change model has been gaining momentum as an all-purpose counselling model that can be used by providers to produce many types of desired behavioural changes, including smoking cessation, dietary change and reducing alcohol consumption and weight (3–5). The 5 A’s may be implemented in approximately 5–10 minutes, allowing for integration into most office visits (4,6).

The 5A’s is an acronym for ‘Assess, Advise, Agree, Assist and Arrange’. Originally modelled as the Four A’s construct (Ask, Advise, Assist, Arrange) by the National Cancer Institute to assist with smoking cessation, ‘Agree’ was eventually added as the fifth ‘A’ by the Canadian Task Force on Preventive Health Care (7). In the existing evidence-based literature, there are multiple variations of the 5 A’s model recommended for use in the primary care setting. The version referenced by the United States Preventive Services Task Force (USPSTF) consists of the aforementioned ‘Assess, Advise, Agree, Assist and Arrange’ (7). Another common variation on the 5 A’s construct includes ‘Ask, Assess, Advise, Assist and Arrange’ (8–10). A third version, in a recently published review article, suggests a ‘Modified 5 A’s’, using ‘Ask, Assess, Advise, Agree and Assist’ for obesity counselling in primary care (11). Refer to Table 1 for an in-depth explanation of each ‘A’, and for a summary of the differences among the models. The remainder of this paper will use the 5 A’s referenced by the USPSTF, as this is the national-level model that has been adopted by Medicare for intensive behaviour therapy for obesity (12).

There is evidence that physicians can be effective in helping overweight/obese patients lose weight using aspects of the 5 A’s counselling model. For example, the majority of overweight and obese patients are not told by health care providers that they are overweight; however, the ones who are told so (‘Assess’) are almost nine times more likely to perceive their weight as damaging to their health compared with those who were not informed (13). In 2010, Singh and Lopez-Jimenez (14) demonstrated that patients who had been told by a physician or health care provider that they were overweight were more likely to have an increased awareness of their weight, to desire to weigh less, to attempt weight loss and to have succeeded in losing weight. Additional research has shown that physician diagnosis of obesity (‘Assess’) is a significant predictor of weight loss attempts and successful weight loss (15).

Thus physicians who ‘Assess’ a patient’s weight status or ‘Advise’ the patient on how to lose weight can have an impact. However, the use of an additional ‘A’ may add to the efficacy of weight loss counselling. In 2010, Jay et al. (16) found that the number of 5 A’s used by physicians was correlated with an increase in patient motivation and intent to lose weight, make dietary changes and increase exercise. Additional research has shown physician use of ‘Assist’ and ‘Arrange’ are related to patient diet improvement and use of ‘Advise’ is associated with increased patient motivation and confidence to change dietary fat intake and lose weight (8).

Despite the demonstrated efficacy of using the 5 A’s counselling strategies, it is clear that current practice is not optimal to support patient weight loss. Only 37%–43% of overweight or obese patients are told by physicians that they are overweight or that they have a weight problem (14). Furthermore, only 20%–29% of obese patients receive a diagnosis of obesity as part of their medical record and only 23% of obese patients have a documented obesity management plan (17,18).

There are likely many reasons why weight loss counselling has traditionally been a challenging area for physicians. Many physicians view their efforts as generally ineffective, reporting inadequate time during routine appointments (19), lack of resources and limited patient motivation as major barriers to effective weight loss counselling (20). In addition, physicians may feel reluctant to address the topic of obesity in the first place, with the fear of offending patients or confirming feelings of failure at weight loss (21,22). Despite physician fears of addressing the topic, it is clear that many obese patients are trying to lose weight (23), with 63% of obese patients reporting attempts to lose weight in the previous year. It is encouraging that almost half of patients view weight loss counselling in a positive manner and list it as one of the possible solutions to obesity (24).

Incorporating patients’ ideas and beliefs regarding their medical care is becoming of greater interest to physicians and is a key component of ‘patient-centred care’, the provision of care that is ‘respectful of and responsive to individual patient preferences, needs, and values’ (25). A systematic review article by Chewning et al. (26) found that among patients with chronic medical conditions, patient preference for participation in the medical decision-making process has increased since the 2000. Thus it is important to have a greater understanding of the type of support and counselling patients feel would be effective in encouraging their weight loss attempts. Identifying patient preferences related to weight loss counselling techniques can benefit both physicians, who may feel hesitant to address the topic, and patients, who may be trying ineffective measures to lose weight.

Many studies have analysed the frequency of use of each of the 5 A’s and their efficacy for producing patient weight loss and/or motivation to lose weight. However, few studies have compared physician use of the 5 A’s with the viewpoints of patients and what they believe to be helpful for achieving weight loss. Gaining an understanding of the patient’s perspective on the use of the 5 A’s is essential for physicians interested in practicing patient-centred care. This review will examine current quantitative research to determine which of the 5 A’s, if any, overweight and obese patients would like their physician to use as part of weight loss counselling. This will be compared to which of the 5 A’s are being implemented by physicians during appointments with patients.

Methods

Data sources

We conducted a systematic search of the MEDLINE/PubMed database from 1–8 October 2013 (1946 to October 2013). Three types of keywords were combined to conduct searches in the database. The first group included terms related to the 5 A’s model (i.e. 5 A’s, Five A’s, Ask or Assess, Advise, Agree, Assist and Arrange). The second group consisted of terms related to weight (i.e. obesity, overweight, weight loss, weight loss counselling). The third group included terms related to either the patient or PCP (i.e. patient perspective, patient viewpoint, patient opinion, and patient belief, or physician use and physician practice patterns). The search was restricted to English language articles.

Data extraction and analysis

A search for the first group of terms in the title and abstract resulted in 230 citations; combining the first and second groups of terms yielded 10 citations, and combining the first and third groups of terms yielded 69 citations. One author (ES) reviewed the abstracts of all 230 articles to determine relevance for inclusion in this review. Reference lists of retrieved articles were also reviewed to determine if any additional articles would meet inclusion criteria. Data were extracted and managed in Microsoft Office Excel® (version 2007). The full text of each selected article was reviewed to collect the following variables: author(s), publication date, study design, sample size, participants and physician demographics, percentage of physicians implementing individual aspects of the 5 A’s and percentage of patients with preference for individual aspects of the 5 A’s. When provided in the original paper, results were extracted by patient BMI status (normal weight: BMI 18.5–24.9kg/m², overweight: BMI 25–29.9kg/m² and obese: ≥30.0kg/m²).

Inclusion and exclusion criteria

Primary research articles providing prevalence data for physician use of one or more aspects of the 5 A’s process and studies which quantified patient preferences concerning which of the 5 A’s practices they would like physicians to implement during weight loss discussions were included. Studies were excluded if they were related to another behaviour change (e.g. smoking cessation), if they were studying the use of the 5 A’s in non-primary care providers (e.g. psychiatrists), if they reported only qualitative data, and if they did not list prevalence for physician use of the 5 A’s as a whole or as individual practices or if they did not list patient preferences concerning the 5 A’s practices.

Results

Fifteen articles met the inclusion criteria for this review. Three articles examined patient perspectives, while the remainder focused on physician implementation of the 5 A’s in clinical practice.

Patient perspectives

Three studies reporting on patient perspectives met the inclusion criteria. These cross-sectional studies analysed patient preferences on use of the 5 A’s counselling practices (Table 2) (27–29). None of these studies asked patients about all five aspects of the counselling process, and none of them asked how often patients want to be ‘Assessed’. Each of the three studies included patients that were normal weight, overweight or obese. Only two of the studies reported the perspectives of the overweight and obese patients separately from the non-overweight patients (27,28).

The most popular of the 5 A’s with patients was ‘Arrange’, with Davis et al. finding that about half of overweight and obese patients want their physician to provide on-going support and/or referral for more intensive help, and Tan et al. reporting that approximately three-fourths of patients at any BMI (normal weight, overweight or obese) felt referral to a registered dietician and/or regular review of weight status would be useful or very useful (27,29). Few studies reported patient preference for ‘Assist’, but two studies reported a more generic proxy for this step, finding that 67%–75% of overweight and obese patients believe their physician can help with weight loss in the future (27,28). About one-fourth to one-half of patients in two studies wanted physicians to ‘Advise’ and ‘Agree’ with them during the counselling process (28,29).

Davis et al. (27) found that 32% of overweight patients and 10% of obese patients preferred not to discuss weight issues with a physician. Potter et al. (28) had similar results, finding that 29% of overweight patients and 12% of obese patients do not want weight-related help from their doctor.

Current physician practice patterns

Twelve studies met the inclusion criteria for physician practice patterns. Five studies were of the cross-sectional direct observation design (Table 3). Flocke et al. (10) observed 300 family medicine appointments for physician use of the 5 A’s. In this study, the 5 A’s were defined as ‘Ask, Assess, Advise, Assist and Arrange’. Flocke et al. (10) found that discussion of weight, diet and/or exercise occurred in 54% (162 of 300) of all visits. Table 3 demonstrates physician use of the 5A’s within the discussions of weight loss, diet and/or exercise. The majority of these discussions were initiated by the observed physicians, and within these discussions, 80%–94% of physicians gave patients advice on weight loss, while about one-fourth to one-half asked about (‘Assess’ in the USPSTF version of the 5 A’s) patient weight or behavioural health risks. A few physicians were observed assisting (14%–17%) or arranging (3%–10%) in these discussions.

Four cross-sectional studies used audio-recordings of appointments to examine current physician practice patterns (Table 3) (8,9,30,31). Two of the four articles also used ‘Ask, Assess, Advice, Assist and Arrange’ to define the 5 A’s (8,9). Following the same trends as the Flocke et al. (10) study, the majority of physicians (52%–63%) gave advice on weight management in discussions concerning weight loss or physical activity. A higher percentage of physicians (77%–91%) assessed or asked about patient weight or behavioural risks, but the percentage of physicians assisting (13%–39%) or arranging (0%–6%) for their patients was similarly low.

Three studies reported the results of cross-sectional surveys asking patients about their physicians’ practices related to the 5 A’s (Table 3) (28,29,32). Evans (32) surveyed obese individuals who had lost ≥ 45kg while attending a commercial weight loss group about physician involvement. Two of the studies surveyed participants in their PCPs waiting room (28,29). The trends were similar to the studies reported above, with ‘Assess’ and ‘Advise’ being commonly used, and ‘Agree, Assist and Arrange’ being rarely used. Additionally, 33% of obese patients reported that their physician did not address weight with them in the study by Potter et al. (28)

Two additional studies surveyed patients on physician use of the 5 A’s using a unique scoring system. Jay et al. (33) compared physician use of the 5 A’s among a control group of physicians and an intervention group receiving a 5-hour-long obesity counselling curriculum based on the 5 A’s, and used post-encounter interviews to record patient report of use of 5 A’s practices. This study used a unique scoring system for the 5 A’s, detailing seven ways to ‘Assess’, four ways to ‘Advise’, three ways to ‘Agree’, one way to ‘Assist’ and four ways to ‘Arrange’, with a total of 18 counselling practices possible within the 5 A’s model. Of 152 patients (BMI ≥30kg/m² or ≥27kg/m² with two or more co-morbidities), 72% received some type of weight loss counselling. The intervention group used more 5 A’s counselling practices than the control group, but the differences were not statistically significant. In both groups, ‘Assess’ and ‘Advise’ were the most commonly used practices in visits where patients were counselled about obesity.

A separate study from Jay et al. (16) used a similar scoring system to assess the use of the 5 A’s. The majority of patients (BMI ≥27kg/m²) reported some type of weight loss counselling, but the overall use of the 5 A’s practices by the physicians was low, with physicians using an average of 5.3 out of the 18 counselling practices. Physicians used ‘Assess’ most frequently, followed by the use of ‘Advise’. Consistent with the other studies summarized in this review, the use of ‘Agree, Assist and Arrange’ was limited.

Two articles used cross-sectional chart audits to assess current physician practice patterns. Davis et al. (27) found that physicians documented providing nutrition, physical activity or weight loss advice (‘Advise’) more often for obese (36%) versus overweight patients (8%). However, they rarely documented arranging follow-up for overweight (4%) or obese patients (17%). Dosh et al. (34) used ‘Ask, Assess, Advise, Assist and Arrange’ to define the 5 A’s and found that ~50% of physicians documented asking about diet and physical activity (Ask) for patients of any BMI that they deemed to be at risk for poor habits. The article states that ‘Advice’ was the next most commonly documented practice and ‘Assist’ and ‘Arrange’ were the least frequently documented of the 5 A’s, but did not list the frequency of physician use of each practice.

Conclusion

Summary

In general, based on the currently available evidence for weight loss counselling, patients would like their physicians to ‘Arrange’, followed by ‘Advise’, ‘Agree’ and ‘Assist’. There was no data regarding how often patients want to be ‘Assessed’. In comparison, physicians most frequently ‘Advised’ and ‘Assessed’, but rarely ‘Agreed’, ‘Assisted’ or ‘Arranged’. This disconnect identifies a significant gap between patient needs and physician practices that likely affects patient care.

Limitations

This review has some limitations related to our methodology. The use of only the MEDLINE/PubMed database may have limited our identification of relevant studies. In addition, the use of ‘physician’ as a search term may have caused us to miss articles relevant to the 5-A-related practices of other primary care practitioners, such as physician assistants and nurse practitioners.

Some of the limitations of this review stem from the quality of the currently available evidence. Few studies examined patient preferences related to the 5 A’s in a quantitative way, making it difficult to draw sound conclusions about how patients would like their physicians to support weight loss. There was inconsistent assessment of all of the 5 A’s and use of non-standardized definitions of the 5 A’s across studies assessing patient preferences and physician practice, making it difficult to accurately quantify and rank both patient preferences and physician practices according to the model. All of the studies were cross-sectional in nature, precluding the ability to determine if patient preferences or physician practices change over time with consistent use of the 5 A’s. In general, although even minimal contact physician interventions can make a difference in patient behaviours, follow-up assessment and support are necessary for successful behaviour change interventions (7). For example, Medicare allows for the 5 A’s to be used in up to 22 intensive behavioural therapy sessions for obesity over a 12 month period (12). All of the designs employed to assess physician practice have some limitations that may have caused biased estimates, including imperfect patient recall on questionnaire-based studies, a lack of detail and complete documentation in medical charts and the risk of physicians or patients systematically changed their behaviour in response to being directly observed or recorded. Finally, it is important to acknowledge that both a variety of participant characteristics (weight status, gender, age, socioeconomic status, education level, race/ethnicity and motivation) and physician characteristics (weight status, gender, age, race/ethnicity, training, time in practice and physician to patient ratio) could affect patient preferences and physician practices and their intersection. For example, Jay et al. found that the number of 5 A’s counselling practices received by the patient during the weight loss discussion was associated with patient motivation to lose weight and intention to change diet and exercise behaviour (16). Many of the articles reviewed presented characteristics of patients or providers descriptively but did not systematically analyse patient preferences or physician practices by subgroups or by patient–physician characteristic concordance. In addition, we did not identify any studies that examined pediatric patient preferences for use of the 5 A’s or physician practices with pediatric patients, despite the fact that childhood obesity is also a significant concern.

Implications for research and practice

The current evidence for patient preferences and physician use of the 5 A’s is based on observational, cross-sectional studies that did not systematically assess all of the 5 A’s using consistent definitions or conduct analyses based on patient or physician subgroups or patient–physician characteristic concordance. An example of a study design that would provide useful information for both patient preferences and physician implementation of the 5 A’s would be a randomized controlled trial in which one group of physicians would receive additional training on the 5 A’s and a control group of physicians would not receive additional training. Implementation of the 5 A’s among both groups could be tracked via direct observation, audio-recording, or video-recording of each physician–patient interaction. Ideally, physicians would interact with a large, diverse group of patients of all ages, and interactions would be observed over multiple visits with the same patients. Independent, blinded, observers would score physician use of the 5 A’s (using standardized criteria) and patients would be interviewed regarding their experience and preferences after the appointment with the physician. Patient outcomes (e.g. weight loss) would be tracked by physician training status. Results would be reported by BMI category, age group, gender and other important patient and physician characteristics, with a large enough sample in each category to draw conclusions about subgroups.

It is also important to determine if factors related to physician training could improve implementation of the 5 A’s. It is well known, for example, that nutrition education for physicians during medical school and postgraduate training in the United States ‘is limited in scope, quality and duration’, with no formal medical specialty in nutrition currently recognized (35). Much of the preclinical didactic nutrition-related coursework in US medical schools is focused on biochemistry, which has little direct relevance to providing patients with supports for behavioural change related to weight loss (35). This lack of specific education in helping patients with behavioural changes related to weight loss is self-perpetuating within the profession, as medical students and residents are then unlikely to encounter faculty physician preceptors who feel comfortable providing specific training in nutrition, physical activity and weight-relevant history taking, examination and counselling skills (35,36). Intensive short courses and online modules focused on interactive, practical application of nutrition education and curriculum focused on the development of healthy personal lifestyle habits among medical students are three promising options to improve physician skills in this area (35,36). Health promotion among medical students and physicians may be a particularly promising practice, as a recent survey found that confidence in counselling patients about physical activity was higher among physicians and medical students with a normal BMI who were routinely active themselves (37). Interprofessional and team-based education models also show promise in connecting physicians with other professionals, such as registered dieticians, during their training to increase chances of referral to these professionals later in the physicians’ careers (36). An important additional area for future research is to assess the impact of improved physician training in nutrition, physical activity and behaviour change models, such as the 5 A’s, on patient preferences and weight management and health outcomes (36).

When an overweight or obese patient is ready to talk about their weight and receive advice from their physician, it seems likely the patient may have a greater chance of success at achieving weight loss if the physician is able to systematically implement the full 5 A’s model with the patient over time. Future research and training efforts should focus on better understanding patient preferences and promoting routine physician use of all aspects of the 5 A’s for supporting weight management among patients.

Declaration

Funding: none.

Ethical approval: none.

Conflict of interest: none.

References

© The Author 2014. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Issue Section:

Review

Email alerts

Citing articles via

More from Oxford Academic