How We Perceive Health Risks, and the Implications for the Healthcare Industry

YAN, Dengfeng | SENGUPTA, Jaideep

Healthcare is one of the fastest growing industries in the world. It consumes more than 10 per cent of GDP in most developed countries, and in the US is expected to reach 19.6 per cent in 2016. How people perceive their health risks will influence consumer decisions, and also economic and public policy. Getting a better understanding of these perceptions therefore is essential.

Dengfeng Yan and Jaideep Sengupta have addressed this issue by looking at how health risk perceptions are influenced by two things: “self-positivity” (underestimating one’s risk when the prevalence, or base rate, is quite high, but case information such as symptoms are low), and “self-negativity” (overestimating the risk when the base rate is low but case information is high). This is one of the few studies to include the latter in the context of health-risk perceptions.

The authors conducted six experiments that showed how self-positivity and self-negativity kicked in depending on participants’ psychological distance from the health risk question. Psychological distance here was tested in terms of social distance (for instance, assessing your own risk or that of “others”) and temporal distance (doing things today, like smoking, that could cause health problems in future). Greater psychological distance was associated with higher base rates and lower case information, reflecting a self-positivity bias. The opposite happened with closer psychological distance, in line with a self-negativity bias.

For example, in one experiment university students were asked to assess their risk for osteoporosis, either currently or when they reached their 60s, after reading a passage that for one group said smoking and heavy drinking were major causes of the disease and for another group said not drinking milk was a major cause. Osteoporosis affects up to 55 per cent of older people in the US, mostly female. Female students who were presented with the smoking and heavy drinking causes and asked to assess their current risk, tended to underestimate their risk, reflecting a self-positivity bias. But male participants presented with the milk-drinking cause, which was a more common behaviour, overestimated their current risk, reflecting a self-negativity bias.

“Prior research has shown that there are adverse consequences to self-positivity bias, such as an unwillingness to take necessary preventive action and a consequently higher rate of infection. Self-negativity bias may also have significant downsides, such as mistakenly diagnosing oneself as possessing a serious disease, which is particularly likely given the ease of information access on the Internet.

“By documenting both types of biases, our work provides straightforward applied implications for consumers, health practitioners and public policy experts in the health domain.”

One implication could be how policymakers fashioned messages to encourage behavioural change. For instance, in the case of osteoporosis, “highlighting relatively infrequent behaviours is unlikely to be as effective as focusing on common risk behaviours. In fact, messages that focus on such uncommon behaviours as smoking and heavy drinking may actually produce a counterproductive effect among those for whom the base rate of the disease is high – in this case, women. It’s likely to induce a self-positivity effect, causing them to downplay the current risk.”

Overall, the authors said self-positivity could be lessened by increasing perceptions of case risk, such as emphasising frequently-occurring risk behaviours. Self-negativity could be lessened by increasing perceptions of the disease base rate, such as highlighting that symptoms are likely related to a common disease rather than a rare one.

“In addition, our findings suggest seeing a doctor rather than engaging in self-diagnosis makes sense not only because the doctor is an expert, but also because, as the patient is an “other”, the doctor will be more likely to take into account the base rate of the disease rather than be solely influenced by case symptoms. This is likely to reduce the negativity bias for relatively common diseases.”


Synergis – Geoffrey YEH Professor of Business, Chair Professor