Apparently they have controlled for this obvious possibility: In each of the three resulting models, our independent variable was the prevalence of T. gondii i

Apparently they have controlled for this obvious possibility: In each of the three resulting models, our independent variable was the prevalence of T. gondii in a given country (the proportion of people who tested positive for T. gondii exposure) – indicated by published seroprevalence data. (https://figshare.com/articles/Supplementary_Methods_and_Analyses_from_Risky_business_linking_i_Toxoplasma_gondii_i_infection_and_entrepreneurship_behaviours_across_individuals_and_countries/6809468 https://figshare.com/articles/Supplementary_Methods_and_Anal...)