Demographic predictors of false negative self-reported tobacco use status in an insurance applicant population

To identify and quantify demographic correlates of false-negative self-reporting of tobacco use in life insurance applicants.

Several studies have assessed the sensitivity of self-reporting for tobacco use in various populations, but statistical examination of the causes of misreporting has been rarer. The very large (488,000 confirmed tobacco users) sample size, US-wide geographic scope, and unique incentive structure of the life insurance application process permit more robust and insurance industry-specific results in this study.

Approximately 6.2 million life insurance applicants for whom both tobacco-use interview questions and a confirmatory urine cotinine test were completed between 1999 and 2012 were evaluated for consistency between self-reported and laboratory-confirmed tobacco-use status. The data set was subjected to logistic regression to identify predictors of false negative self-reports (FNSR).

False-negative self-reporting was found to be strongly associated with male gender, applicant ages of less than 30 or greater than 60, and low cotinine positivity rates in the applicant's state of residence. Policy face value was also moderately predictive, values above $500,000 associated with moderately higher FNSR.

The findings imply that FNSR in life insurance applicants may be the result of complex interactions among financial incentives, geography and presumptive peer groups, and gender.