When Cancer Doesn’t Stop People From Smoking

Why do some people keep smoking, even when a loved one has become seriously ill because of tobacco? A study based on 200,000 people in France shows that these “family shocks” have very little impact on behaviour. The findings raise questions about the effectiveness of prevention policies.

According to Santé Publique France, tobacco killed around 68,000 people in France in 2024, nearly 186 deaths per day. Worldwide, it is responsible for more than 8 million deaths each year, according to the World Health Organization. Despite the scientific consensus on its harms, and decades of prevention measures such as awareness campaigns, plain packaging, graphic warning labels, and price increases, nearly a quarter of French adults still smoke. Why is it so difficult to encourage a change in behaviour?

Beyond public health messaging, lived experience, such as a loved one falling ill or a parent diagnosed with a smoking-related cancer, might be expected to act as a wake-up call and permanently change behaviour. But is that really the case? Researchers Sylvie Blasco, Eva Moreno Galbis, and Jeremy Tanguy examine the impact of a “family shock,” such as a relative’s diagnosis, on behaviour. Their study offers important insights for public health policy.

The researchers drew on data from the Constances cohort, a large-scale public health program tracking more than 220,000 adults in France over several years. Participants report their lifestyle habits annually, including tobacco use, along with medical check-ups and family health histories. This allows researchers to reconstruct individual smoking trajectories over time.

©Serhii / stock.adobe.com

Mixed findings in earlier research

The study fits into a broader body of work in health economics on risk behaviours following a parent’s diagnosis. Earlier research has produced mixed results. Some studies1  have found that health shocks can shift individual preferences. For example, American economists Michael Darden and Donna B. Gilleskie2 observed a slight decline in smoking, mainly among daughters, after a father experienced a cardiovascular event. Other research on the intergenerational transmission of behaviour points to strong continuity and generally concludes that the effects are weak or even non-existent.

Thanks to the richness of the Constances cohort, the researchers were able to examine whether a diagnosis of lung cancer or another smoking-related cancer in parents affects their children’s smoking behaviour. They also explored how the impact varies depending on the child’s age at the time of the diagnosis.

To do so, they used tools from health economics and econometrics to isolate the effect of a specific event on behaviour over the long term. The aim was to determine whether an “information shock,” in this case a parent’s diagnosis of a tobacco-related cancer, alters individual decision-making independently of other factors already at play, such as family transmission or risk perception. This approach helps limit confounding biases and comes as close as possible to identifying a causal relationship.

• 1

  Decker S., Schmitz H., “Health shocks and risk aversion”, Journal of Health Economics, Volume 50, 2016, Pages 156-170.

• 2

  Darden, M. and Gilleskie, D. (2016). The Effects of Parental Health Shocks on Adult Offspring Smoking Behavior and Self-Assessed Health. Health Economics, 25(8):939–954.

©Samantha Peralta / Unsplash

Adolescence, a pivotal moment

The findings may come as a surprise. Counterintuitively, in most cases, seeing a parent develop cancer does not lead to any significant change in the behaviour of their children once they reach adulthood. The effects are small and rarely statistically significant. The shock appears too indirect to disrupt deeply ingrained habits.

There is one exception. When the diagnosis occurs during adolescence, between the ages of 15 and 18, the likelihood of smoking decreases. The authors show that this is a critical period when many individuals decide whether to start smoking. When a parent’s diagnosis coincides with this window, the information shock is more likely to influence behaviour.

How can we explain the gap between intuition and observed outcomes? Several explanations from behavioural economics offer insight. First, risk often remains abstract. Even when confronted with a loved one’s illness, many people still believe they themselves will never be affected. Economist Christian Bünnings, 3 who has studied decision-making under known risks, shows that individuals need information that feels personal and tailored to them in order to truly adjust their perception of risk. Risk assessment is gradual and often incomplete.4 Knowing the dangers is not enough. People must also feel personally concerned.5

The study directly challenges current prevention strategies. Many public health campaigns are based on the idea that exposing people to consequences is sufficient to change behaviour. These findings suggest the limits of that approach and point to the need to rethink intervention strategies, especially among younger populations. Prevention is most effective when it occurs before habits become firmly established.

• 3

   Bünnings, C., 2017,Does new health information affect health behaviour? The effect of health events on smoking cessation”. Applied Economics, 49(10):987–1000.

• 4

  Khwaja, A., Sloan, F., and Chung, S., 2006,”. Learning about individual risk and the decision to smoke. International Journal of Industrial Organization”, 24(4):683–699.

• 5

  Khwaja, A., Silverman, D., Sloan, F., and Wang, Y., 2009“Are mature smokers misinformed?” Journal of Health Economics, 28(2):385–397.