Surveys were distributed to 591 PCPs (n=538 family physicians; n=53 nurse practitioners). Of these, 103 (17.4%) completed the survey. Online completion rate was 81.7%. The demographic characteristics are shown in Table 1. Where applicable, values are presented as mean ± standard deviation (SD).
Male participants were significantly older than females (mean age 54.2±12.6 vs. 44.0±11.7 years, p<.001), and had been in practice significantly longer (25.21±13.5 vs. 13.9±11.7 years, p<.001). PCPs in a non-interprofessional practice (IPP) were significantly older than those in an IPP (53.9±13.1 vs. 45.0±12.1 years, p=.004). Similarly, those in a non-IPP had been in practice significantly longer compared to those in an IPP (23.6±13.6 vs. 15.5±12.7 years, p=.012).
Part A: Plans of Care for Patients with Class II and III Obesity
Surveyed PCPs estimated that 11.6±9.8% of patients in their practice would qualify for MSWLI. PCPs serving a rural population estimated that 14.2±10.9% of their patients would qualify for MSWLI, which was significantly higher than PCPs serving an urban population (9.9±8.5%, p=.049). Overall, 53.3% (49/92) of PCPs were not aware of contemporary international guidelines regarding the referral of patients with class II and III obesity and type II diabetes for BS. Overall, 57.5% of respondents (50/87) did not feel competent prescribing weight management programs to their patients with class II and III obesity, and 87.1% (74/85) disagreed that they are usually successful in helping patients with class II and III obesity lose weight without BS.
Part B: Referrals for Bariatric Surgery
Overall, 69.8% (50/86) of respondents agreed they had ‘good’ knowledge of referral criteria for BS. Males were significantly more likely to agree that they are aware of the guidelines for BS referral compared to women (p=.018). Overall 95.4% (82/86) of PCPs have referred patients for BS, however, 60.9% (53/87) reported that they had referred 10% or fewer of their patients who would qualify for BS. Figure 1 summarizes the most frequently cited reasons for referral for BS.
When asked about reasons for BS referral, 37.9% (33/87) of PCPs agreed that patients initiated conversations about BS as a potential treatment for their obesity, whereas 44.8% (39/87) agreed that they most often brought it up. PCPs in practice for 0-10 years were significantly less likely to initiate discussions about referral for BS than PCPs practicing for 11-20 (p=.002), 21-30 (p=.016), and 31+ years (p=0.013). Additionally, there was a significant weak positive correlation between age of the PCP and likelihood of initiating discussion about referral for BS (r=0.363. p=.003). 61.2% (52/85) of respondents agreed that patients with obesity often seek consultation with them for the purpose of receiving information about BS.
Part C: Reservations about Bariatric Surgery
Overall, 22.2% (19/85) of study participants agreed that they are hesitant to refer patients for BS, with the two most common reasons being concerns about postoperative surgical complications and risks associated with surgery. Figure 2 identifies the specific reservations of PCPs about referring their patients for BS.
PCPs serving an urban population were significantly more likely to report that they had reservations about referring their patients for BS due to the lack of long-term data on the effects of surgery on obesity-related comorbidities as compared to those serving a rural population (39.4% (13/33) vs. 11.5% (3/26), p=.012). Female PCPs were significantly more likely to report that they had reservations about referring for BS due to past negative experiences as compared to male PCPs (22.7% (10/44) vs. 6.7% (1/15), p=.037).
Part D: Perceptions of Follow-up after Bariatric Surgery
Nearly every PCP surveyed (97.7%, 84/86) agreed that long-term follow-up is required after BS; 37.9% (32/87) felt that follow up should be the responsibility of the bariatric surgeon, 25.0% (21/84) were comfortable providing long-term follow up themselves, and 25.6% (21/82) reported having resources necessary to provide good-quality long-term follow-up after BS. Only 18.4% (16/87) of PCPs felt competent in addressing medical complications that may arise after BS.
Part E: Future Treatment of Class II and III Obesity
Overall, over half of PCPs (58.8%) agreed that the future treatment of patients with class II and III obesity must be based primarily on lifestyle intervention and behavioural modification; 43.4% (36/83) and 34.5% (29/84) believed it should be based primarily on BS with behavioural and dietary modifications and medical management with dietary restriction, respectively. Several differences between groups are summarized in Table 2.
Part F: CPD in management of patients with class II and III obesity
Only 39.1% (34/87) of PCPs had participated in education on management of patients with class II and III obesity in the past five years. Overwhelmingly, 88.5% (77/87) of PCPs believed there is a need for education on this topic, with PCPs in their first 10 years of practice significantly more likely to agree compared to those who have been in practice for 11-20 years (p=0.033). Overall, 79.3% (69/87) believed that it is very important to be knowledgeable about medical treatment options for obesity; with PCPs serving a rural population significantly more likely to rate this importance higher compared to those serving an urban population (p=0.013). Finally, 63.2% (55/87) believed that it is very important to be knowledgeable about BS for patients with class II and III obesity.
Interpretation
In this study, we explored the knowledge, experiences, perceptions, and educational needs of PCPs in Southeastern Ontario in managing patients with class II and III obesity. PCPs acknowledged that over 10% of patients in their practice had class II and III obesity and most PCPs agreed that these patients are not likely to succeed in achieving durable weight loss without BS. PCPs in rural locations perceived to have a greater proportion of patients with class II and III obesity as compared to urban locations. These perceptions are in agreement with reported 17.2% of patients with class II and III obesity in primary care in Southeastern Ontario [17]. The perceptions of greater proportion of patients with obesity in rural setting is also in agreement with global findings [26].
Despite an accurate perception of the proportion of patients with class II and III obesity in their practice and the knowledge that BS can prevent long-term medical complications of obesity, more than 60% of PCPs reported referring fewer than 10% of eligible patients for BS. Most PCPs reported initiating the referral for BS following a direct request from a patient, and fewer than half of all PCPs reported being the one initiating conversations about BS with patients. These findings are in agreement with a 2014 Ontario survey [27], which reported that over 70% of physicians have referred no more than 5% of their patients with class II/III obesity for BS. Plausible explanations for the low referral rate include lack of knowledge by PCPs about risks and benefits of contemporary BS, and the role that BS can play in helping patients with obesity improve their quality of life [28].
In our study, more than one third of all PCPs were hesitant to refer patients for BS due to concerns about complications and risks associated with surgery, and medical complications after surgery. In a recent qualitative study of PCPs in Southeastern Ontario, we identified that most PCPs viewed BS as high-risk, with significant short- and long-term post-operative complications [28]. PCPs also viewed BS as a last resort after unsuccessful attempts at all other weight loss interventions [28]. Other studies have highlighted similar concerns of PCPs about associated risks of BS [29-32]. These perceptions are in contradiction with data regarding safety of contemporary BS with an overall complication rate of 11.7% and mortality of 0.16% [33]. In comparison, mortality rate for a cholecystectomy in is nine times higher at 1.36% [34]. This gap in knowledge regarding safety of contemporary BS may contribute to the hesitancy of PCPs to refer their patients, however it can be addressed via ongoing professional development activities.
We identified that PCPs in their first 10 years of practice were significantly less likely to bring up BS with their patients compared to PCPs in all other age groups. We also found a significant positive correlation between age of PCP and likelihood of bringing up BS. This is a surprising finding as BS had been available and covered by a provincial health insurance plan in Ontario for 10 years at the time of this study [35]. Our results do not provide an explanation for this finding; however, similar results have been observed in other studies [27,36]. PCPs who have been in practice fewer than 10 years may not have been in practice long enough to appreciate the long-term benefits of BS. This finding suggests that greater emphasis should be placed on management of patients with obesity in undergraduate and postgraduate medical education programs for PCPs in order to better prepare trainees for transition to independent practice.
We identified that one in two PCPs were not aware of contemporary guidelines that recommend considerations of BS for patients with class II/III obesity and type 2 diabetes [24,25], and that one in two PCPs did not feel competent prescribing weight management programs to their patients. Moreover, we also identified a lack of knowledge in PCPs regarding follow-up and care for patients post BS. Our findings are consistent with other literature [27,37,38] in that that few PCPs report having the knowledge they need to feel comfortable providing quality aftercare for their patients following BS. These findings are concerning given the potential cost-savings to the health care system by recommending BS to patients with type 2 diabetes and class II/III obesity. Studies from the United States [39] and Europe [40,41] demonstrate that BS may lead to cost savings to the health care system.
We identified that fewer than half of PCPs in our study region had participated in continuing education on the management of patients with obesity in the past five years. The majority of PCPs, however, do believe there is a need for more education about MSWLI for patients with class II/III obesity. This was especially true for PCPs in their first 10 years of practice and those in rural locations. However, the most effective pedagogical approach to deliver this professional development is not clear. Lectures and symposia have been shown to have a positive impact on physician knowledge; however, interactive CPD activities that encourage reflection on practice, provide opportunities to practice skills, involve multiple exposures, and are focused on outcomes appear to be the most effective at improving practice and patient health outcomes [42,43]. Future research should focus on optimizing and evaluating the delivery of CPD activities focused on management of patients with obesity.
We must emphasize that PCPs knowledge, experiences, and perceptions about managing patients with class II/III obesity are a few of the many barriers that patients with obesity face in gaining access to treatments for their chronic disease. Other barriers include insurance policies and funding issues. Many patients are unable to afford the $253.00 per week out-of-pocket cost for the meal replacement required to take part in medical weight loss programs, while Health-Canada approved anti-obesity medications are not covered by any of the provincial/territorial public drug benefit programs [16]. These barriers are system-related, and out of the control of PCPs. Nonetheless, PCPs should be aware of these barriers when managing and helping eligible patients access MSWLI options.
Study limitations
The low response rate may have led to selection bias, as only those interested in the topic may have been recruited. Additionally, the results of the survey used in this study have not been previously validated. Our results are also subject to recall bias, and bias associated with self-reported responses. Additionally, due to the anonymous nature of the study, we did not collect IP addresses and therefore participants may have completed more than one entry. Lastly, our study was conducted in one region of Ontario, which may limit generalizability to other contexts.