Childhood trauma and adulthood physical health in Mexico

The proportion of women in the sample was higher than it should have been (55%) according to Mexican census data. Preliminary analyses indicated that the bias occurred at the point of selection for the psychological interview, although the reason for this is not clear. Fieldwork supervisors reviewed audiotapes of each interview and verified that the interviewer selected the appropriate adult (the one with the most recent birthday) for the psychological interview regardless of who gave the sociodemographic interview or who was home at the time of that initial interview. Because information was collected about all household members during the initial interview, it was possible to compare selected men and women to the larger “populations” from which they were chosen on several variables. Analyses of the household demographic data indicated that female participants were quite representative of the larger population of women, but among male participants, younger, lower-income, and less-educated men were underrepresented. However, the magnitude of this bias appeared to be relatively small, with effect sizes ranging between .09–.12.

It is important to note that before undertaking our epidemiologic study we conducted preliminary research to determine whether PTSD was a relevant construct for Mexican trauma survivors. Results from qualitative interviews determined that Mexican respondents mentioned 14 out of the 17 specific PTSD criterion symptoms with little or no prompting (Norris et al. 2001b). In a subsequent quantitative study conducted with samples of disaster victims from the US and Mexico, a four-factor measurement model representing the accepted multi-criterion conceptualization of PTSD fit the data of the US and Mexican samples equally well (Norris et al. 2001a). Both studies implied that PTSD is a relevant and measurable construct in Mexico.