Lack of diversity in product testing — safe?

Weight and size are key variables in determining passenger safety, and a lag in adjusting test- to real-world conditions can have significant, even fatal, consequences. The same is true for the many other variables, including gender, age, ethnicity or genetics, which can make trial populations non-representative of end-consumers. Where so, caution in underwriting is advised.

Potential impacts

  • Insurers could face product liability and / or product recall claims if the testing population for new products available on the market is not representative of end-user groups.
  • This is especially so if new products are rolled out fast to many users. If a large serial-loss event involving body injuries or even death occurs after rollout, product flaws could also trigger claims in Life and Health policies.

Making products safe for human use is not an easy task, not least because product testing needs to keep up with demographic trends. This includes adjusting test  conditions and results for gender, and changes in weight, height, age and others. Failure to do so can have fatal real-world consequences.

Car-crash test dummies are a case in point. Typically, the test dummies are mostly modelled on the average male, who is not representative of the entire population. This is a factor in explaining why women are 47% more likely than men to incur severe injury in a car accident.1 Accident statistics point to a similar story. For example, Europe’s roads are getting safer, it seems. Between 2010 and 2017 the number of road fatalities went down 20%. At the same time, however, the proportion of elder people injured or killed on the roads had risen.2 A suspicion is that as crash test dummies are modelled on the up-to 65-years of age bracket, they are not representative of older drivers. As more than 20% of the EU’s population will be more than 65 years old soon, “older” test dummies are now in the making.

Age, gender and physical properties are not the only issues. Genetics, lifestyles and other factors also make people different. In certain fields, notably health, these are important considerations. For instance, a recent study showed that women are more than twice as likely to experience adverse side effects to medications than men.3 The side effects range from mild symptoms like nausea and headache to serious problems including cognitive abnormalities, seizures, hallucinations, agitation and cardiac issues. Historically, it has been more usual for men to participate in clinical drug trials than women. However, at rest metabolism in females is typically slower than in males. This means that dosage levels determined by study results on males can be too high for women.

These examples highlight the hidden risk of safety trials, where the trial population does not represent end-users. For insurers, particularly in fields where product use could lead to bodily injury or health impairment, risk assessors should fully investigate the data set used by regulators as the basis for product approval. If test populations are not representative of end-user groups, a more cautious underwriting approach makes sense. This is especially so where potential for serial losses exists due to fast roll-out of new products to many consumers.


1 European Commission, “Gendered Innovations: How Gender Analysis Contributes to Research”, EU Horizon 2020, 10 December 2013,
2 C. Collins, “Crash test dummies based on older bodies could reduce road fatalities”, HORIZON, 11 June 2018,
3I. Zucker and B. J. Prendergast, “Sex differences in pharmacokinetics predict adverse drug reactions in women”, Biology of Sex Differences 32 (2020),


Related SONAR content

SONAR 2021: New emerging risk insights

While the COVID-19 crisis dominates the risk landscape, other emerging risks and trends arise from developments in human-machine interaction, connected infrastructures, and ethics and sustainability.