OTTAWA — The Canadian Medical Association (CMA) has expressed concern that more patients are seeking health advice from artificial intelligence (AI), potentially placing their health at risk. This trend is particularly alarming in a context where many Canadians lack access to a primary care provider, leading them to rely on quickly accessible information online.
According to Dr. Margot Burnell, the CMA's president, the rise of AI as a health information source is troubling due to the prevalence of misinformation, disinformation, and false information in these digital resources. Such inaccurate information threatens to undermine the trust that exists between patients and their physicians.
A survey commissioned by the CMA and conducted by Abacus Data involved 5,001 Canadians, exploring the sources from which they obtain health information and their perceived trustworthiness. The results showed that nearly all respondents search online for health-related information, with a focus on specific conditions, symptoms, or treatment options.
Alarmingly, 64% of those surveyed reported encountering false or misleading health information online at least occasionally, while 29% indicated they experienced this issue rarely. Despite only 27% of participants expressing trust in AI to provide accurate health information, around half admitted to using AI search results from platforms like Google or ChatGPT for health-related inquiries. Moreover, approximately 38% confirmed they utilized ChatGPT to seek treatment advice.
The survey results highlighted a critical concern: individuals who acted on AI-generated advice were found to be five times more likely to report experiencing adverse health reactions or negative effects as a consequence of that advice. Dr. Burnell noted that while patients have historically conducted their own online research before visiting their doctors, AI tools pose a unique challenge. Such platforms often present definitive health recommendations without considering an individual's specific medical history.
Dr. Burnell emphasized that the nature of AI information complicates the verification process, as it can be challenging to identify the source and lack a clear scientific basis. The survey also underscored a widespread belief among Canadians that both social media companies and the government hold responsibility for addressing health misinformation.
Respondents expressed significant concern regarding the quality of health information emerging from the United States. Over three-quarters of those surveyed indicated they were worried about increasing misinformation originating from the U.S. Figures like former President Donald Trump and health secretary Robert F. Kennedy Jr. have been noted for making inaccurate claims regarding vaccines and other health issues.
Dr. Burnell pointed out that much of the misinformation propagated during the Trump administration lacks scientific backing, which can contribute to public skepticism toward other health information sources. Approximately half of the survey participants stated that exposure to misinformation has led them to question the advice provided by healthcare professionals, while 69% reported a loss of confidence in other online health information as a result.
The Abacus survey was conducted in early November and cannot be assigned a margin of error as online surveys do not constitute truly random samples. The CMA maintains a partnership with The Canadian Press that supports health coverage, although the association does not influence editorial content.
