We assessed Canadians’ awareness of known cancer risk factors and beliefs in cancer myths. Similar to previous studies that examined both myths and known risk factors [19, 37], respondents were better able to identify known cancer risk factors over cancer myths. For example, in the UK, 52% of known risk factors and 36% of cancer myths were accurately identified, compared to 70% of risk factors and 49% of myths in this study [19]. Variations in the average percentage of correctly identified cancer facts and myths were observed by demographic group, particularly for income, ethnicity, and age, consistent with previous studies [19, 23, 38, 39]. Differences in beliefs about cancer myths and risk factors by thinking style were observed, suggesting that those who are more actively open-minded and prefer effortful thinking may be better able to identify cancer myths and risk factors, while those who are more close-minded and prefer intuitive thinking are less able. When examining the correlations between the thinking styles and individual cancer myths in particular, more items were significantly correlated with actively open-minded thinking– i.e., the tendency to question beliefs according to evidence - compared to the three other thinking styles (although only small to modest correlations were observed throughout) [40, 41]. Additionally, people who were correct in their responses regarding cancer myths and facts had slightly higher mean active open-minded thinking (AOT) and preference for effortful thinking (PET) scores than those who were incorrect. People who provided correct answers had slightly lower mean preference for intuitive (PIT) and close-minded thinking (CMT) scores than those who were incorrect. The small differences in thinking styles means and relatively small effect sizes for the relationships between the thinking styles and the individual myths and facts may suggest that the scales overlap in the space of health misinformation somewhat. However, our work showed that people with higher AOT scores in particular were more consistently correct in their identification of cancer myths. This is demonstrated by the higher negative value Spearman correlation coefficients seen in Additional File 2 for AOT scores and, for example, eating GMO foods. The higher the AOT score, the less likely a respondent was to agree with the statement that GMO foods increase the risk of cancer, which is a correct interpretation. People with higher AOT scores were also more likely to correctly identify correct facts about cancer prevention, such as practicing sun safety or getting the HPV vaccine. Higher PET scores followed similar trends as the AOT scores, though the results were not as strong. Conversely, people with higher CMT and PIT scores were less likely to answer questions about facts or myths correctly. For example, one of the strongest positive correlations we detected with PIT scores demonstrated that people with higher PIT scores were more likely to believe that drinking fluoridated water, living near power lines, and eating GMO foods increase the risk of cancer. Thinking dispositions and cancer myths and risk factor knowledge, specifically, have not been studied to our knowledge; however, the broader literature has shown evidence that being actively open-minded, thinking more analytically, and having a need for cognition (or, preference for effortful thinking) are significant factors when it comes to the discernment of fake news and health misinformation [29, 30, 32, 31–44]. In other words, those that are willing to question their beliefs based on evidence, and those who enjoy and are willing to dwell and think through information are more likely to identify cancer myths. These findings may have implications on how and with whom information should be shared in the future in order to increase cancer risk factor knowledge [45]. The 4 intuitive-analytic thinking styles that we used in our study are a very new tool for research of this nature, and further use of them in the design of potential interventions is needed as their use evolves and increases [46].

Study participants had variable awareness of known cancer risk factors and cancer myths. Tobacco continues to be a well recognized risk factor for cancer (with over 90% of participants correctly identifying tobacco smoking and exposure to second-hand smoke), but less than 35% were able to correctly identify low vegetable and fruit intake (less than 5 portions of vegetables or fruit per day) and low physical activity (less than 30 min of physical activity, 5 times a week) as risk factors for cancer. These findings were consistent with results found in the UK [19]. However, in Newfoundland and Labrador, being over 70 years of age, drinking more than one unit of alcohol per day, and having a diet low in fibre were the least identified risk factors (≤ 65% correct) [23]. The results on relatively poor recognition of alcohol use and low fibre diets may not be as surprising as the fact that older age was poorly recognized as a factor; for most cancers, age is the most significant predictor of cancer risk [3]. The most commonly endorsed myths among our participants were consuming additives (61%), feeling stressed (52%), consuming artificial sweeteners (49%), and being exposed to electromagnetic frequencies (42%). These are consistent with the findings from the UK study, where the top three myths were stress (42%), food additives (41%), and exposure to electromagnetic frequencies (35%) (19). Although a different population was surveyed, these findings suggest potentially persistent beliefs in these myths across countries.

Recommendations to counter misinformation include targeting the “undecided majority” over the “unswayable minority” who hold more extreme views that are particularly resistant to change [47]. For this reason, it is important to examine the myths and risk factors with a high percentage of “neither agree nor disagree” responses. By doing so, incorrect beliefs where there is more room for further communication and for which awareness initiatives may be more successful can be identified and targeted. The greatest uncertainty for cancer myths was presented for glyphosate as a risk factor for cancer, and alkaline diets as a form of prevention (approximately 60% for both). The high degree of uncertainty may relate to real uncertainties about glyphosate exposure and cancer risk. To date, messaging and media reports on hazardous potential and/or human health risks of glyphosate remains mixed with conflicting conclusions by major agencies, including the International Agency for Research on Cancer, the European Union, and the US EPA [48–54]. We were unable to identify published literature that examines myths related to alkaline diets and cancer and that could help to explain such a high level of uncertainty in this myth, but it may suggest that people have heard of these diets but are not sure about any conclusions made on the topic. It is also possible that respondents were unsure about what these substances or practices are, and so they were less sure in their responses.

Participants were also uncertain about drinking fluoridated water, using hair dyes, and sustaining injuries or traumas as cancer risks. They were also uncertain about drinking red wine, consuming cannabis, taking vitamins and supplements, eating organic foods, and having a base tan as being protective. While some of these are mostly harmless, they may distract from known protective factors, such as increasing overall vegetable and fruit intake, or physical activity. For example, organic foods may also be prohibitively expensive for people with lower household incomes, and fear of consuming pesticides on non-organic fruits and vegetables could lead to the unintended consequence of avoidance of these foods, which would be detrimental to one’s overall health and cancer risk. Furthermore, some of these beliefs are outright harmful. Drinking any amount and type of alcohol, including red wine, increases the risk for at least 9 different types of cancer [55]. Having a base tan is not protective against skin cancer [56], and indoor tanning, which is used to acquire base tans, actually increases skin cancer risk [57]. Fluoridated water is an important public health measure that reduces the risk of dental caries among children [58], and attempts to minimize exposure to fluoridated water could negatively impact oral health while increasing disparities across socioeconomic lines as many lower income households may not be able to maintain regular dental care.

For known risk factors, approximately 20–30% of participants were uncertain whether the HPV vaccine, reducing radon exposure at home, and eating 5 portions or more of vegetables and fruits a day reduces the risk of cancer. Similarly, approximately a quarter to a third of participants were uncertain whether eating less than 5 portions of vegetables or fruit and HPV infection increases cancer risk, suggesting that social marketing campaigns to educate the public about HPV, including the HPV vaccine and how it can prevent certain types of cancer, as well as the importance of vegetable and fruit intake, could be beneficial. In particular, with nearly 100% of cervical cancers being due to HPV infection, there are concerted efforts to increase HPV immunization with the goal of eliminating cervical cancer in Canada in Canada by 2040 and these findings can be used to inform this work [59, 60]. Furthermore, radon is a known lung carcinogen and the leading risk factor for lung cancer in people who don’t smoke tobacco [61, 62], yet it appears that a large proportion of respondents do not understand the risk of radon exposure. This is especially important in Canada which has among the highest home-based exposures to radon in the world [63]. Promotional campaigns are needed now to inform the public about the risks related to radon exposure and to encourage residents to test their homes. Given that people can only make decisions within the environments that they live, campaigns should be complemented with healthy public policies and supportive environments to reduce barriers to cancer prevention [64].

Several questions were presented for their presence or absence in both fact and myth categories, such as eating or not eating five servings of fruits and vegetables a day or getting/not getting 30 min of exercise five days week. We saw that for fruit and vegetable consumption, respondents agreed that meeting the recommendations was preventive against cancer, but did not agree as much that not meeting the recommendations actually increases cancer risk. Similarly, respondents indicated high awareness that meeting exercise recommendations reduces the risk of cancer, but low awareness that not meeting the recommendations is a risk factor for cancer. These findings may offer important information on framing for cancer prevention campaigns.

This is the first national study that assessed Canadians’ knowledge/awareness of risk factors and cancer myths, with consideration of respondents’ thinking dispositions. By recruiting participants through Prolific [65], we were able to obtain a large and diverse sample of participants. Although a sex-balanced sample was achieved, the sample for this study was relatively young relative to the Canadian population. Furthermore, Prolific is prone to selection bias in terms of how Prolific is advertised, who chooses to participate in Prolific surveys [66] and which surveys are chosen for completion by participants. In order to reduce participant burden, we presented only a random selection of myths and risk factors to each participant. While this allowed us to increase our sample size, we were unable to conduct any predictive modelling for the participants’ overall awareness of cancer myths and risk factors.

Future analyses should examine individual risk factors and myths in more detail, with particular attention to the undecided majority and predictors of beliefs. In addition, Canadians’ awareness of other aspects of the cancer control continuum, including cancer treatment options should be explored, since this is an area ripe with misinformation [67, 68]. Furthermore, people with cancer have been shown to have high rates of complementary alternative medicines use [69, 70].

There were several limitations to our approach. Though our search strategy for survey design (regarding which questions to ask the participants) was broad, it did not necessarily locate all possible cancer myths and facts that are in circulation in the published and grey literature or on social media. Though the survey was developed with careful thought and reference, it was not directly evaluated for clarity/comprehension using a cognitive interview approach. We did run a small pilot of the survey before launching, but no concerns were identified about some of our questions which may have been less clear. Without a vetting process for the survey, we may have overlooked potential challenges. For example, the responses regarding glyphosate may be attributed to lack of understanding of what it is, rather than a misunderstanding of whether it causes cancer. The options did not include an “I don’t know” or “I don’t know what that is” option, so people who may have fallen in those categories had to mark “neither agree, nor disagree”. Our question about whether cancer can be attributed to “having bad luck” was included due to media coverage of a particular study several years ago that pervasively made its way into many media sources that were not well-reported, but respondents may have interpreted ‘bad luck’ to mean genetic mutations, which is indeed a true risk factor for cancer. When collecting demographic information, we only collected information on income and not highest education in order to limit the number of questions for respondents and because these two variables are often highly correlated. Therefore, it is certainly possible that income is acting as a proxy for higher education in our study and may explain why higher income respondents were more accurate at identifying cancer myths and facts. Further, for findings to be representative, it would have been beneficial to have used multiple sources for recruiting research participants to have a more diverse sample to reduce sampling bias.

In summary, we found that while Canadians were able to identify some known cancer risk factors well, many myths are believed and many risk factors are still not recognized. Larger, multi-year campaigns, such as those for smoking cessation, seem to have worked as many respondents understand the risks related to tobacco smoking. Future work should aim to increase awareness of lesser known but established risk factors, such as radon, vegetable and fruit consumption and HPV vaccination while debunking cancer myths so that evidence-based information can be prioritized. The results of this study can be used by relevant stakeholders to prioritize myths for debunking and lesser-known risk factors for social marketing campaigns with complementary shifts in effective healthy public policies.