We speak with Judith Prochaska, author of the study ”Likelihood of Unemployed Smokers vs Nonsmokers Attaining Reemployment in a One-Year Observational Study“ that came out April 11 in JAMA internal medicine.
ResearchGate: In your recent study you found that smokers have a harder time getting jobs. How did you come to this conclusion? What were your methods?
Judith Prochaska: We conducted a study of 251 jobseekers with interviews out to 12-months follow-up. About half were daily smokers and half were nonsmokers. We found that the smokers were at a serious disadvantage for finding work and among those who did, they earned, on average, $5 less per hour. Our methods were self-report surveys with bioconfirmation of recent smoking using breath samples to test for carbon monoxide.
RG: Smokers are on average less educated, poorer, and more likely to come from a minority background. All of these factors would make it harder for them to get a job. How did you rule them out?
Prochaska: You are correct about the general profile of smokers, and we saw this pattern in our sample. To account for these factors, Drs. Michael Baiocchi and Eric Daza on our team ran sophisticated data analyses to account for the observational design and the pre-existing differences found in the groups.
Specifically, using propensity score analysis we sought to equate the groups on key factors—i.e., duration of time out of work, age, education, race/ethnicity, and perceived health status in the first model and then added in access to transportation, criminal history, prior alcohol/drug treatment, sex, and housing stability.
Further, we trimmed the data to remove extreme cases—i.e., smokers who were so different on these variables from nonsmokers that there was no overlap in the distributions, and similarly nonsmokers who appeared to be so characteristically different from smokers. You can think of this as a mathematically precise way of “comparing apples to apples.”
