Extreme obesity (body mass index or BMI ≥ 35 kg/m
2) in Canada has increased in prevalence by over 400% in three decades and currently afflicts nearly 9% of the adult population [
1,
2]. Currently, bariatric surgery is widely considered to be the most effective treatment for extreme obesity and contemporary Canadian obesity management guidelines recommend that bariatric procedures be considered in patients who are refractory to non-surgical interventions and who have either severe obesity (body mass index [BMI] ≥ 40 kg/m
2) or medically complicated moderate obesity (BMI ≥ 35–39.9 kg/m
2 with a major obesity-related comorbidity such as diabetes or obstructive sleep apnea) [
3,
4]. Surgery markedly reduces obesity-related morbidity and mortality, improves health-related quality of life, and is cost-effective; in Canada, the incremental cost-effectiveness ratios over a lifetime horizon are $9000-12,000 per quality-adjusted-life-year (QALY) gained [
4‐
7].
From 1996 to 2008, the number of publicly funded bariatric surgeries performed annually in Canada increased 12-fold to 1882 procedures [
8,
9]. Despite this increase, patients seeking to undergo a publicly funded bariatric procedure in Canada currently face protracted multiyear wait times that average 5 years [
10]. Furthermore, residents of six of the thirteen Canadian provinces or territories do not have access to a bariatric surgery program within their home province or territory [
9]. These patients must either travel to another Canadian province for publicly funded surgery or pay out-of-pocket to undergo a privately funded procedure.
High demand, driven primarily by the large number of potentially eligible surgical candidates, is thought to be a major contributor to these lengthy wait times [
11]. However, no previous study has formally examined the prevalence of the bariatric surgery-eligible population in Canada and calculated the proportion of patients eligible that
actually receive surgery. Furthermore, no prior study has characterized the surgery eligible Canadian population in comparison to the population considered surgery-ineligible and the population currently receiving bariatric surgery. These data would help inform the current state of bariatric care in Canada, facilitate comparisons with other countries, and assist future health care delivery planning. For example, those responsible for setting volume targets for bariatric surgery in Canada can use these data to understand the current surgical volumes and estimate remaining gaps. Comparing similar data from other countries would allow one to rank surgical volumes in Canada against other health care systems and determine if they require adjustment. To address these knowledge gaps, we linked several contemporary and nationally representative surveys and administrative databases to characterize the population eligible for bariatric surgery in Canada and compare them with the patients currently receiving bariatric procedures.
Discussion
This analysis of nationally representative Canadian data quantifies and characterizes those who are eligible, ineligible and receiving bariatric surgery in this country. Our first major finding is that a very low proportion of bariatric surgery-eligible Canadians is currently receiving surgery annually. The large gap between potential demand and delivered supply is likely a major contributor to protracted wait times for bariatric procedures [
10]. The second major finding of this study is that notable differences in sociodemographic profiles and comorbidities are present amongst individuals that are eligible for compared with those actually receiving bariatric surgery in Canada. Most notably, surgery-eligible individuals exhibit poorer self-reported health status and greater burden-of-illness compared to those considered surgery-ineligible. Conversely, surgical recipients are younger and appear to have a very low burden of comorbidity (with the exception of diabetes), suggesting that it is ’lower risk’ patients are more likely to receive these procedures.
In 2006, 0.4% of the surgery-eligible population underwent a bariatric procedure in the US [
19]. This proportion is 400% higher than our estimates for Canada and may be partially (but not fully) explained by the 40% higher prevalence of extreme obesity in the US [
1]. Most bariatric procedures in the US are performed within the private health care sector [
20]. Privately performed adjustable gastric banding is available in 4 Canadian provinces, but these surgeries are not captured within CIHI databases and thus the total number of bariatric procedures (public plus private) performed within Canada is not known. The results of a recent survey of privately funded bariatric clinics in Canada indicated that private clinics may improve access but provide less comprehensive follow-up care, [
21] although the quality and accuracy of these data have been questioned [
22].
It is clear that only a small minority of surgery-eligible patients is currently able to access a bariatric procedure. The feasibility of increasing the proportion of surgery-eligible patients that receive a publicly funded bariatric procedure in Canada is uncertain and the optimal proportion of eligible patients that should receive surgery remains undefined. Increasing the provision of surgery is contingent upon the availability of surgeons, multidisciplinary bariatric programs, specialized operating rooms and funding. A recent Health Technology Assessment estimated that increasing the provision of bariatric surgery to 5% of eligible Canadians over 5 years would require nearly 32 000 more publicly-funded bariatric procedures at a cost of nearly 500 million dollars [
7]. This estimate does not, however, address the issue of whether or not publicly funded procedures should be reserved for patients with selected characteristics (such as those with greater comorbidity or those that are predicted to incur higher future health care costs). Emerging classification systems that reflect obesity-related comorbidity as opposed to BMI alone may prove useful as a means to stratify potential candidates and reduce the eligible population to a more manageable quantity [
23].
Many of the sociodemographic differences seen in our analysis comparing the surgery-eligible and ineligible populations in Canada parallel those reported in the US. Specifically, surgery-eligible individuals in the US are more likely to have lower education levels, income status and self-reported health status [
19,
24]. In contrast, patients receiving surgery within both Canadian and US programs are predominantly of higher socioeconomic status [
19,
25]. In the predominantly privately-funded US healthcare system, these inequities are readily explained by individual variation in access to health insurance or ability to pay for surgery. However, within Canada’s publicly funded system (which espouses universal accessibility), no obvious reasons for these socioeconomic disparities in access to bariatric surgery are apparent and this issue requires further examination.
Our results also demonstrate that sex-related disparities exist among bariatric surgery recipients in Canada, whereby women are 4 times more likely to undergo surgery compared to men. This pattern is also similar to that reported in the US, and may be partly explained by the approximately two-fold higher prevalence of severe obesity in females compared to males that is present in both countries [
1,
19,
26]. A recent survey of US bariatric surgeons found that sex was not a significant factor in candidate selection; thus, surgeon preference does not appear to explain these findings [
27]. Men may be less likely than women to seek treatment because they may be less aware of the health hazards of extreme obesity. Women may be more likely to seek surgery for body image reasons and it is also possible that sex-related differences in the perceived mental and physical health impact of extreme obesity may explain the higher tendency for women to seek surgery [
28].
Obesity-related comorbidities were less common in bariatric surgery recipients compared to the surgery-eligible population. This may be a consequence of candidate preselection, whereby surgeons (or programs) preferentially select healthier patients to undergo surgical intervention or related to known under-coding of chronic stable comorbidities that do not impact surgical risk or length of stay [
16]. The lack of a relationship between greater comorbidity burden and receipt of surgery has also been found in other publicly funded bariatric programs. In a tertiary care, publicly funded Norwegian bariatric surgery program, patients undergoing surgery were younger, heavier and a greater proportion had earlier-onset obesity [
29]. However, a higher obesity-related comorbidity burden was not present in those receiving surgery [
29].
Limitations of our analysis are inherent to the data sources examined and the fact that we compared two different, albeit population-representative, data sources (survey data and administrative data). The CHMS was a voluntary survey and, as with all surveys of this type, the generalizability of the results depends upon the absence of systematic differences between responders and non-responders. Certain obesity-related comorbidities (e.g., sleep apnea) considered to be indications for bariatric surgery were not captured in the CHMS and some comorbidities were self-reported. For these reasons, the number of surgery-eligible subjects may have been underestimated. Three potential limitations of the administrative CIHI data should be noted. First, BMI is not captured within administrative databases; therefore, we assumed but could not directly verify that surgical programs are following current guideline-concordant eligibility criteria when selecting surgical candidates. Published data from the largest publicly funded Canadian bariatric programs suggest that this assumption is justified [
30‐
32]. Second, data from Quebec, which accounts for approximately 40% of publicly funded bariatric procedures in Canada, were not available for the analysis of bariatric surgery recipients and this limits the generalizability of our results to this province. However, published data from the two largest Quebec programs suggest that the demographic characteristics of surgical recipients are similar to our reported results [
30,
32]. Third, as discussed above, administrative data may not fully capture comorbidity prevalence and surgical complication rates resulting in the underestimation of these outcomes. Fourth, we lacked estimates of the number of Canadians undergoing out-of-country bariatric procedures, which are not captured by CIHI data.
In conclusion, our results quantify and characterize the bariatric eligible population in Canada in relation those considered ineligible and those receiving surgery. These data are useful to depict the current state of bariatric surgery delivery in Canada, thus facilitating comparison to other jurisdictions and serving as a benchmark for ongoing and future initiatives related to bariatric health care delivery in this country. Our findings raise a number of questions regarding the adequacy and appropriateness of publicly funded bariatric surgery care delivery in Canada. In this regard, we note that four of the five founding principles of the Canada Health Act (which outlines the objectives of publicly funded health care delivery in this country) are not currently being met. These principles are comprehensiveness (i.e., all services deemed essential are provided), public administration (obviating the need for private service delivery), universality (all residents receive equal care), accessibility (in a timely manner), and portability (across all provinces and territories). We propose that future efforts should focus on fully characterizing these care gaps; clarifying the role of privately delivered care; examining the value of prioritization methods to streamline and optimize care; and ensuring accessibility for suitable candidates seeking bariatric surgery.
Competing interests
RP, SWK, SM and AMS are supported by an alternative funding plan from the Government of Alberta and the University of Alberta. AMS is supported by an Alberta Health Services Chair in Obesity Research and Management. SWK and SM hold salary support awards from Alberta Heritage Foundation for Medical Research and Alberta Innovates-Health Solutions. The authors declare no competing interest with respect to this work.
Authors’ contributions
RP developed the initial study concept, which was refined after input from all of the other authors. HC and RP performed data analysis. RP wrote the initial draft, which was critically revised by all other authors. All authors read and approved the final manuscript.