Weight bias and support of public health policies

Abstract

Objectives

Public health policies have been proposed to help address prevalent Canadian obesity rates. Along with the increase in obesity prevalence, explicit weight bias is also rampant in Western society. This paper aimed to assess the association between explicit weight bias attitudes and Canadian public support of these policy recommendations.

Methods

Canadian adults (N = 903; 51% female; BMI = 27.3 ± 7.0 kg/m²) completed an online survey measuring explicit weight bias, using the three subscales of the Anti-Fat Attitudes Questionnaire: Willpower (belief in weight controllability), Fear of fat (fear of gaining weight), and Dislike (antipathy towards people with obesity). Whether these subscales were associated with policy support was assessed with logistic regression. Analyses were adjusted for age, race, gender, and income.

Results

Public support of policy recommendations ranged from 53% to 90%. Explicit weight bias was primarily expressed through a fear of weight gain and the belief that weight gain was within the individual’s control based on willpower. Although the Dislike subscale was associated with lower support for several policies that enable or guide individual choice in behaviour change, the Willpower and Fear of fat subscales were associated with greater support for similar policies.

Conclusion

This study contributes to evidence-informed public health action by describing public support of public health policies and demonstrating an association between explicit weight bias and public support. A higher total explicit weight bias score increased the odds of supporting primarily less intrusive policies. However, dislike of individuals with obesity was associated with decreased odds of supporting many policies.

Introduction

Obesity is an international public health issue. Weight bias (the tendency to associate negative or stereotypical beliefs and attitudes with an individual because of their weight) is highly prevalent (Puhl & Heuer, 2010; Washington, 2011). When individuals are linked to an “undesirable” characteristic (in this context, their excess weight), a rationale for rejecting and devaluing them is constructed and termed ‘weight bias’ (Link & Phelan, 2006). Explicit weight bias, a specific type of weight bias, involves the conscious belief that individuals with obesity lack willpower and self-control (Brewis, 2014; Puhl & Brownell, 2001). Attributing obesity to causes within the individual’s control (e.g., lack of willpower to exercise regularly) contributes to weight bias (Blaine & Williams, 2004). With the increased prevalence of obesity, one might hypothesize that weight bias attitudes would decline. However, explicit weight bias has grown unabated (Puhl & Heuer, 2010). For instance, the majority (55%) of a Danish representative sample (N = 1141, aged 20–70 years) agreed that “If fat people really wanted to lose weight, they could”(Lund, Sandøe, & Lassen, 2011). Research has documented a range of adverse psychosocial and physical health consequences of weight bias (Hatzenbuehler, Keyes, & Hasin, 2009; Major, Eliezer, & Rieck, 2012; Puhl & Suh, 2015). These adverse consequences interfere with the quality of life of individuals with obesity and can impede efforts to improve their overall mental and physical health (Latner, Barile, Durso, & O’Brien, 2014; Puhl & Suh, 2015).

To address prevalent obesity rates in Canada, the federal, provincial and municipal levels of government have been implementing initiatives, laws and regulations (e.g., mandatory calorie labelling on menus) to promote healthy changes in public dietary and physical activity behaviours. In March 2016, Canada’s federal government released: Obesity in Canada: A Whole-of-Society Approach for a Healthier Canada (“the Obesity Report”) (Ogilvie & Eggleton, 2016). The Obesity Report contained the expert testimony of Canadian and international stakeholders presented to the Standing Senate Committee on Social Affairs, Science and Technology. It concluded with 21 policy recommendations in addressing the obesity epidemic.

In accordance with Nuffield’s Council of Bioethics’ Inter-vention Ladder framework, these policy recommendations cover the spectrum of less intrusive to more intrusive policies (Nuffield Council on Bioethics, 2007). The Intervention Ladder framework is a measure of governmental input which considers the degree of public policy intrusiveness on individual autonomy (Kongats, McGetrick, Raine, & Nykiforuk, 2020; Nuffield Council on Bioethics, 2007). For instance, less intrusive policy recommendations enable individual choice in behaviour change or guide choice by changing norms/standards to healthier options. On the other hand, more intrusive policies guide choice through the use of disincentives/incentives and restrict or eliminate population freedom to choose (Nuffield Council on Bioethics, 2007). In order to garner public support, more intrusive policies require stronger justification (i.e., produce a desired outcome which outweighs losses of individual liberty) (Kongats et al., 2020). As other interpretations as well as other frameworks can be applied to this study, the impact of our decision to utilize the Intervention Ladder framework with these interpretations is further described in our Discussion section.

Research assessing public support of these policies is crucial as it has the potential to influence policy creation, implementation and adoption (Burstein, 2003; Mazzocchi et al., 2015). However, to our knowledge, only a few studies have examined Canadian public support of public health policies (Bhawra et al., 2018; Lange & Faulkner, 2012; Morin & Roy, 2013; Potestio, McLaren, Vollman, & Doyle-Baker, 2008; Puhl et al., 2015; Reid, 2011). To date, no studies have specifically examined support of or opposition to the 2016 Obesity Report Federal policy recommendations. As the implementation of the recommendations will impact the health behaviours of the entire Canadian population, improving our understanding of public support of these recommendations is needed.

The purpose of this study was to examine the association between explicit weight bias and policy support of the 2016 Obesity Report Federal policy recommendations in a large public sample of Canadian adults. As individuals who perceive obesity to be beyond individual control have been reported to support more intrusive public health policies, we hypothesized that individuals with higher explicit weight bias scores would be less likely to support more intrusive policy recommendations and more likely to support less intrusive policy recommendations (Barry, Brescoll, Brownell, & Schlesinger, 2009).

Methods

Procedure and participants

Cross-sectional data were obtained from a large public sample recruited by Survey Sampling International (SSI), a market research company. Individuals registered as survey participants with SSI who were over 18 years old were eligible to participate in this online study. A total of 42,080 eligible participants were invited to complete the survey. These individuals were informed via email of the study purpose, length of the survey, and incentivization. Quotas based on age, gender, and province of residence allowed for a close approximation of Canadian census demographics (Census Profile, 2016 Census, n.d.). All SSI participants were also members of SSI partner organizations, which allowed for personalized incentives. Approximately 1865 expressed initial interest by clicking on the survey invitation link. Of these participants, 903 completed the survey (response rate: 48%). The 20-min survey was hosted on SurveyMonkey. This research study received ethical approval from a Research Ethics Board (Ethics certification number: 30009752). All participants completed an informed consent form.

Measures

Explicit weight bias

To assess explicit weight bias, Crandall’s 13-item Anti-Fat Attitudes Questionnaire (AFA) was used (Crandall, 1994). The AFA consists of three subscales: Dislike (n = 7 items), Fear of Fat (n = 3 items) and Willpower (n = 3 items). The Dislike subscale assesses antipathy towards individuals with obesity (e.g., I really do not like fat people much). The Fear of Fat subscale considers emotions towards weight gain (e.g., I feel disgusted with myself when I gain weight), and the Willpower subscale assesses perceptions that weight gain is within the individual’s personal control (e.g., some people are fat because they have no willpower). Responses were rated on a 10-point Likert scale (0 = very strongly disagree, 9 = very strongly agree). Higher scores in each subscale are indicative of greater weight bias attitudes. In this study, Cronbach’s alpha for the Dislike, Fear of fat and Willpower subscales were 0.89, 0.85 and 0.81, respectively, and 0.88 for the total scale. These scores indicate strong internal consistency for each subscale as well as for the entire questionnaire.

Support of public health policies

Based on the original 21 Obesity Report recommendations, participants indicated their level of support of 15 public health policy recommendations. The other six recommendations were excluded if they were redundant and/or required specialized knowledge from participants. A policy was identified as redundant if the recommendation referred to a similar policy action as another policy. For example, two different recommendations suggested revisions to the Canadian food guide. Rather than including both recommendations, we reworded one recommendation to encompass the premise of two recommendations. Since some policy recommendations were verbose in the original Obesity Report, some policies were rephrased to increase readability and reduce participant burden. All survey items were then pilot tested in three phases, with three different audiences: first, university students (n = 4) completed paper copies of the survey during the first pilot test; a second sample of Canadian adults (n = 12) then completed the survey online using the SurveyMonkey platform; and last, a sample of our target population (n = 84 Canadian adults) recruited by SSI completed the online survey in the third pilot phase. Policy support was assessed on a 4-point Likert scale (1 = strongly oppose, 4 = strongly support). For analytic purposes, responses indicating support (“strongly support”, “support”) were collapsed together and compared with responses indicating opposition (“strongly oppose”, “oppose”).

Demographic questionnaire

The demographic questionnaire included items assessing height, weight, age, gender, race, and annual household income.

Data analysis

Statistical analyses were conducted using SPSS version 24. Since a standard cut-off for high vs. low explicit weight bias scores is not established in the current literature, the explicit weight bias data were analyzed as continuous variables. Spearman correlations were conducted to examine the relationships among the three explicit weight bias subscales. Logistic regressions were conducted to investigate the association between explicit weight bias and support for each policy recommendation. Explicit weight bias was assessed in two separate sets of models: (1) as a total score, and (2) as three predictors (Willpower, Fear of fat, and Dislike subscales) in a single model. All logistic regression models were adjusted for age (18–44 vs. >45 years), race (white vs. other), income (<$24,999 vs. >$25,000) and gender (male vs. female) since these co-variates were associated with either the dependent variable (policy support) or the independent variable (explicit weight bias). Since the adjusted variables were measured categorically, we were unable to analyze them as continuous variables. A sensitivity analysis was additionally conducted where we adjusted for BMI. However, since BMI was not significantly associated with policy support, nor did its inclusion change any of the main findings, it is not included in the models presented here.

Results

In total, 903 participants completed the survey during the period of October 15–26, 2018 (Table 1). Rated on a 0–9 Likert scale, the mean explicit weight bias total score was 3.4 (1.7). The Dislike subscale mean score was 2.4 (1.9), while the Willpower and Fear of fat subscale mean scores were 4.6 (2.2) and 4.6 (2.7), respectively. The Willpower subscale was moderately positively associated with both the Dislike (r = 0.43; p < 0.001) and Fear of fat (r = 0.42; p < 0.001) subscales, while the Dislike subscale was weakly positively associated with the Fear of fat (r = 0.28; p < 0.001) subscale.

Table 1.

Demographic characteristics of study sample

Characteristic	Total sample (N = 903)a
Gender
Female	464 (51.4)
Male	439 (48.6)
Age
18–44	420 (46.5)
45 or older	483 (53.5)
Race/ethnicity
White	684 (75.8)
Other	219 (24.3)
BMI, kg/m2
<18.5	37 (4.1)
18.5–24.9	311 (34.4)
25–29.9	291 (32. 2)
>30	247 (27.4)
Annual household income
<$24,999	167 (18.5)
>$25,000	736 (81.5)
Explicit weight bias scores, mean (SD)
Total	3.4 (1.7)
Willpower	4.6 (2.2)
Fear of fat	4.6 (2.7)
Dislike	2.4 (1.9)

^aN (%) unless indicated otherwise

The percentage of participants who supported each of the 15 public health policy recommendations is presented in Table 2. In this table, cut-off values were used to organize the different policies in terms of levels of public endorsement from most to least. These cut-offs were derived from Roger’s Diffusion of Innovations Theory, a framework for understanding policy adoption, and have been used in previous similar literature in this field (Nykiforuk, Eyles, & Campbell, 2008; Raine et al., 2014). Canadian public support of the 15 policy recommendations ranged from 52.9% to 91.9%. The majority of policies (n = 10, 66.7%) received strong endorsement (84.0–97.5% support). All strongly supported policies (n = 10) were policies categorized as less intrusive, such as encouraging improved training for physicians regarding diet and physical activity (91.9%) and encouraging the use of nutrition labelling on menus and menu boards in food service establishments (90.5%). The remaining five policies received moderate endorsement (50.0–83.7% support). The majority of these moderately endorsed policies (n = 4) were categorized as more intrusive (e.g., prohibiting the advertising and promotion of food and beverages to children (66.7%) and taxing sugar and artificially sweetened beverages (52.9%)).

Table 2.

Canadian public support of public health policies

Policy	Policy category	Support (%)a
Strong endorsement (84.0–97.5%)
Encourage improved training for physicians regarding diet and physical activity	Enable choice	91.9
Require that the daily intake value for protein be included in the Nutrition Facts table	Enable choice	91.6
Implement campaigns to increase public awareness of healthy active lifestyles and healthy eating	Educate	90.8
Encourage the use of nutrition labelling on menus and menu boards in food service establishments	Enable choice	90.5
Revise Canada’s food guide to include meal-based guidelines	Educate	90.5
Implement breakfast and lunch programs at school and childcare facilities and programs that improve physical activity, and nutrition literacy	Enable choice	89.4
Promote the Canadian Physical Activity Guidelines of 60 min of moderate to vigorous-intensity physical activity per day for children and 150 min of moderate to vigorous-intensity physical activity per week for adults	Educate	89.0
Change the infrastructure and designs of communities to encourage physical activity	Enable choice	87.4
Promote physician counseling and the use of exercise in prescriptions	Enable choice	85.3
Mandate the use of front-of-package nutrition labelling	Enable choice	84.2
Moderate endorsement (50.0–83.7%)
Prohibit the use of partially hydrogenated oils to minimize trans-fat content in food	Eliminate choice	82.3
Strictly limit the use of permitted health claims and nutrient content claims on packaged foods	Enable choice	77.1
Develop taxes and subsidies to help Canadians of lower socio-economic status choose healthy lifestyle options	Incentives	72.8
Prohibit advertising and promotion of food and beverages to children	Eliminate choice	66.7
Taxation of sugar and artificially sweetened beverages	Disincentives	52.9

^aSupport is defined as the percentage of respondents who selected “strongly support” or “support” for a specific policy. Policy categories are based on the Nuffield Council of Bioethics’ Intervention Ladder

Table 3 presents adjusted logistic regression results. A higher total explicit weight bias score (indicative of greater weight bias attitudes) was associated with increased odds of supporting six policies. Five of the six policies were less intrusive policies, with taxation of sugar and artificially sweetened beverages being the only exception (odds ratio (OR) = 1.18, 95% confidence interval (CI) = 1.09–1.28).

Table 3.

Logistic regression analysis results of weight bias and public health policy support

Public health policy	Willpower		Fear of fat		Dislike		Total explicit weight bias score
	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Less intrusive: enable choice
Implement breakfast and lunch programs at school and childcare facilities and programs that improve physical activity, and nutrition literacy	1.07	0.95–1.21	1.06	0.97–1.17	0.83	0.73–0.95	0.96	0.84–1.09
Change the infrastructure and designs of communities to encourage physical activity	1.21	1.08–1.35	1.06	0.97–1.16	0.97	0.84–1.10	1.26	1.11–1.43
Mandate the use of front-of-package nutrition labelling	0.97	0.88–1.07	1.06	0.98–1.15	1.00	0.90–1.13	1.05	0.94–1.17
Encourage improved training for physicians regarding diet and physical activity	1.16	1.0–1.34	1.14	1.02–1.27	0.88	0.75–1.03	1.20	1.03–1.40
Require that the daily intake value for protein be included in the Nutrition Facts table	1.19	1.03–1.36	1.18	1.05–1.32	0.77	0.66–0.91	1.12	0.97–1.30
Promote physician counseling and the use of exercise in prescriptions	1.10	0.99–1.21	1.05	0.97–1.14	0.97	0.86–1.09	1.13	1.00–1.27
Encourage the use of nutrition labelling on menus and menu boards in food service establishments	1.19	1.04–1.35	1.04	0.94–1.15	0.85	0.74–0.99	1.06	0.92–1.22
Strictly limit the use of permitted health claims and nutrient content claims on packaged foods	1.02	0.94–1.11	0.98	0.91–1.04	1.11	1.00–1.23	1.09	0.99–1.21
Less intrusive: educate
Implement campaigns to increase public awareness of healthy active lifestyles and healthy eating	1.14	1.00–1.30	1.04	0.94–1.16	0.86	0.74–1.00	1.02	0.89–1.18
Revise Canada’s food guide to include meal-based guidelines	1.07	0.91–1.17	1.07	0.96–1.18	0.90	0.78–1.03	1.0	0.87–1.15
Promote the Canadian Physical Activity Guidelines of 60 min of moderate to vigorous-intensity physical activity per day for children and 150 min of moderate to vigorous-intensity physical activity per week for adults	1.16	1.03–1.31	1.10	1.00–1.21	0.90	0.79–1.04	1.18	1.04–1.35
More intrusive: incentives and disincentives
Taxation of sugar and artificially sweetened beverages	1.00	0.94–1.09	1.00	0.94–1.06	1.19	1.09–1.29	1.18	1.09–1.28
Develop taxes and subsidies to help Canadians of lower socio-economic status choose healthy lifestyle options	1.03	0.95–1.12	1.06	0.99–1.13	0.97	0.88–1.06	1.06	0.97–1.17
More intrusive: eliminate choice
Prohibit advertising and promotion of food and beverages to children	0.93	0.86–1.00	1.07	1.00–1.14	1.16	1.06–1.27	1.15	1.06–1.26
Prohibit the use of partially hydrogenated oils to minimize trans-fat content in food	1.11	1.00–1.23	1.15	1.06–1.24	0.94	0.84–1.06	1.23	1.11–1.38

Logistic regression analyses exploring the relationship between explicit weight bias and public health policy support (support/strong support vs. oppose/strongly oppose) after adjusting for age (18–44 vs. >45 years), race (white vs. other), income (<$24,999 vs. >$25,000) and gender (male vs. female). OR odds ratio, CI confidence interval. All significant results (p < 0.05) are shown in bold. Willpower, Fear of fat and Dislike were all included in a single model as predictor variables while Total explicit weight bias score was included in a separate model

In terms of the three explicit weight bias subscales, a higher Willpower mean subscale score was significantly associated with greater odds of support of 6/15 policies, five of which were less intrusive (e.g., changing infrastructure to encourage physical activity, OR = 1.21, CI = 1.08–1.35). Higher Fear of fat subscale scores were significantly associated with an increased odds of supporting two less intrusive and two more intrusive policies (e.g., prohibit advertising and promotion of food and beverages to children, OR = 1.07, CI = 1.00–1.14). The Dislike subscale was significantly associated with five less intrusive policies and two more intrusive policies. In contrast to the other two subscales, a higher Dislike score was associated with a decrease in the odds of policy support. For instance, there was lower support for implementing breakfast and lunch programs at school and childcare facilities and programs that improve physical activity and nutrition literacy (OR = 0.83, CI = 0.73–0.95).

Discussion

The present study assessed explicit weight bias and examined public perceptions of the 2016 Obesity Report policy recommendations in a large public sample of Canadian adults. As all policies were either strongly or moderately endorsed, we speculate that the overall level of policy support reported by the Canadian public reflects perceptions that obesity is a significant health issue that warrants some governmental intervention in the form of public policy. Since a higher total explicit weight bias score increased the odds of supporting six policies, five of which were less intrusive, our hypothesis of a positive association between explicit weight bias and support of less intrusive policy recommendations was supported by the results. However, explicit weight bias subscales were differentially associated with the extent to which the public wants the government to intervene. Importantly, although weight bias was in general associated with supporting less intrusive policies, a higher Dislike subscale score was associated with decreased support of many policies.

The findings of this study are consistent with the existing literature assessing support of general public health policies aimed at addressing obesity (Barry et al., 2009; Lange & Faulkner, 2012; Mata & Hertwig, 2018; Oliver & Lee, 2005). Previous research involving Canadian youth and young adults, 16–30 years of age, also reported similar high support of less intrusive nutrition policies as compared with more intrusive policies (Bhawra et al., 2018). Another study reported lower Canadian young adult (N = 521, mean age 20) support of more intrusive policy recommendations, ranging from 37.8% to 78.9% (Lange & Faulkner, 2012). Like the Canadian public, Canadian key policy influencers also report weak endorsement of restrictive, more intrusive policies and nearly unanimous support (80–99% support) for less intrusive individual-focused policies (Raine et al., 2014). Our study is an extension of previous literature as it not only assesses specific federal policy recommendations, but, through our assessment of the different explicit weight bias subscales, provides insight into the specific attitudes that are associated with these trends in perceptions of public health policies.

Previous studies have reported that one of the strongest predictors of public support of policies aimed at addressing obesity was beliefs about the etiology of obesity (Barry et al., 2009; Oliver & Lee, 2005). Attributing obesity to causes within the individual’s control contributes to explicit weight bias (Blaine & Williams, 2004). In our sample, explicit weight bias was expressed through a fear of weight gain and the belief that weight gain was within the individual’s personal control based on willpower, rather than through dislike of individuals with obesity. The Willpower subscale scores were among the highest, demonstrating that the public does believe that weight gain is attributable to individual behaviours that are based on willpower (e.g., exercise). Support of public health policies that place more responsibility on the individual (i.e., less intrusive policies) reflect Canadian perceptions that weight gain is within the individual’s personal control.

Similarly, an estimated 55% of Canadian key policy influencers in the literature view obesity as a personal responsibility (in contrast with the 1.7% who view obesity as a societal responsibility) (Raine et al., 2014). Perhaps Canadians who attribute weight gain to a lack of willpower do not favourably weigh improvements in population obesity rates against the loss of liberty that comes with implementing more intrusive policies (Nuffield Council on Bioethics, 2007). This speculation is also supported by the results observed with the Dislike subscale. Dislike of individuals with overweight and obesity is also important to consider when speculating on public perceptions of policy. Unlike the Willpower subscale, Dislike as a predictor of public health support has not often been investigated. In our study, Dislike was associated with lower support of several less intrusive policies and higher support of two more intrusive policies. Based on these results, those who dislike individuals with obesity do favourably weigh improvements in population obesity rates against the loss of liberty that comes with the implementation of more intrusive policies (Nuffield Council on Bioethics, 2007). Those who dislike individuals with obesity are willing to tolerate infringement upon their personal freedoms as a means to address prevalent obesity rates.

Study strengths and limitations

This paper has several strengths. To our knowledge, this study is the first to utilize a large Canadian sample to assess public support of the 2016 Senate obesity recommendations from the Canadian federal government. Study results suggest the three subscales (Willpower, Dislike, and Fear of fat) contribute unique information to the total weight bias score. The assessment of all three subscales of explicit weight bias provides a comprehensive understanding of negative attitudes towards people living with obesity and the role each subcomponent plays in the relationship between explicit weight bias and public policy support. Previous studies often examined only one subscale of explicit weight bias (Barry et al., 2009; Chambers & Traill, 2011; Lange & Faulkner, 2012; Oliver & Lee, 2005). Nevertheless, further research on interactions among subscale scores and how they may predict policy support is needed due to the multidimensional nature of weight bias. The categorization of the policies into less intrusive and more intrusive provided insight into the category of policies most supported by the Canadian public. However, more research is needed to better understand how and why public opinions are formed regarding obesity-related policies and to understand Canadians’ reluctance towards more intrusive public health policies that may effectively address obesity at a population level.

Although our policy categorization is a strength of our study, we recognize that the Intervention Ladder framework is not the only framework that could be applied appropriately for this study. As illustrated in the literature cited herein, related studies have utilized a range of different categorization frameworks. However, we decided to apply a well-known public health framework (the Intervention Ladder framework) (Kongats et al., 2020) in a novel manner since previous studies investigating Canadian correlates of obesity public health policy support have not considered this framework. Our decision to utilize this as our categorization framework was primarily informed by previous literature and analytic methods (thematic analysis and factor analysis). As identified by the thematic analysis and confirmed with our factor analysis, the Intervention Ladder framework was the best fit for the specific policies included in the Obesity Report, and it met our objective of applying a commonly utilized public health framework to facilitate our identification of the specific types of policies that the Canadian public supported or opposed. Next, our use of a market research company to recruit our study sample increased the likelihood of selection bias. Individuals registered with market research companies and their affiliate organizations may be inherently different from those not registered. This limits the external validity of our study results. Although we took steps to address this limitation using quotas that allowed for a close approximation of Canadian demographics, our sample demographics suggest an under-representation of people with low household income and an over-representation of white Canadians. Another limitation involves our reliance on participants’ self-reported attitudes and beliefs, which are susceptible to socially desirable responding. Next, although the 2016 Canadian Obesity Report included 21 policy recommendations aimed at preventing obesity, our study only assessed 15 of the 21 recommendations. Six of the excluded policy recommendations required participants to have specialized knowledge on obesity or shared similar concepts with policy recommendations included in the study. Last, related to our analysis of the explicit weight bias data, since no standardized cut-off for high vs. low explicit weight bias currently exists, explicit weight bias was analyzed as a continuous measure. Although higher scores indicate greater weight bias attitudes, whether specific explicit weight bias scores are considered high or low is not apparent in current literature.

Conclusion

The present study contributes to evidence-informed public health action by describing public support of public health policies addressing obesity and demonstrating an association between explicit weight bias and public support. Although such evidence informs public policy development and implementation, as accountable governments consider public opinion, the results presented herein do not suggest the exclusive implementation of policy recommendations that received strong public endorsement. It is not only important to consider public acceptability of public policy, but the policies implemented must also be evidence-based and have the potential to effectively improve population health behaviours. In situations where public opinion is not aligned with evidence-based initiatives, policymakers must be willing and able to undertake actions that promote health equity, regardless of public opinion (Raphael & Sayani, 2019). An evaluation of the effectiveness of the implemented public health policies in changing population health behaviours and addressing obesity is warranted in the future.

Author contributions

All authors were involved in the conception and design of this study. Since the results reported in this manuscript were also presented in partial fulfillment of the first author’s M.Sc. degree, similar results are published in a research repository. The first author was responsible for the acquisition of data, conducting the data analysis and interpreting the results with guidance from the second and third authors. The first author drafted the manuscript which was revised and edited by the second and third authors. All authors approved the final version of this manuscript.

Funding

The first author was supported by the France and André Desmarais and the William R. Sellers graduate student fellowships at Concordia University, Montréal. The second and third authors both hold Junior 1 salary awards from les Fonds de Recherche du Québec – Santé.

Compliance with ethical standards

Conflict of interest

The first author received funding for a Mitacs Accelerate Fellowship for another research project, unrelated to the submitted work, under the supervision of the third author. The second author has no conflicts of interest to disclose.