Associations between Use of Antimalarial Medications and Health among U.S. Veterans of the Wars in Iraq and Afghanistan

Abstract

Mefloquine (Lariam; Roche Holding AG, Basel, Switzerland) has been linked to acute neuropsychiatric side effects. This is a concern for U.S. veterans who may have used mefloquine during recent Southwest Asia deployments. Using data from the , a population-based study of U.S. veterans who served between 2001 and 2008, we investigated associations between self-reported use of antimalarial medications and overall physical and mental health (MH) using the twelve-item short form, and with other MH outcomes using the post-traumatic stress disorder Checklist-17 and the Patient Health Questionnaire (anxiety, major depression, and self-harm). Multivariable logistic regression was performed to examine associations between health measures and seven antimalarial drug categories: any antimalarial, mefloquine, chloroquine, doxycycline, primaquine, mefloquine plus any other antimalarial, and any other antimalarial or antimalarial combination while adjusting for the effects of deployment and combat exposure. Data from 19,487 veterans showed that although antimalarial use was generally associated with higher odds of negative health outcomes, once deployment and combat exposure were added to the multivariable models, the associations with each of the MH outcomes became attenuated. A positive trend was observed between combat exposure intensity and prevalence of the five MH outcomes. No significant associations were found between mefloquine and MH measures. These data suggest that the poor physical and MH outcomes reported in this study population are largely because of combat deployment exposure.

In Response to the Letter by Nevin

“…In summary, we believe Dr. Nevin’s comments 1 on our article 2 are unfounded. The strengths and limitations of our study are noted in the Discussion section.…”

“…In a letter to the editor 1 commenting on our article 2 on the association between antimalarial medications (including mefloquine) and health outcomes in veterans, three concerns with our analyses were presented. We disagree with the premise that the stated concerns diminish our findings as they are largely theoretical and lack empirical support.…”

“…Third, we disagree with Dr. Nevin’s interpretation of our conclusions and the statement that the effects of combat exposure on mental health burden are not supported by our data. We concluded, “While there appeared to be significant elevated odds of poor mental and physical health outcomes among those who reported antimalarial use relative to nonusers, once the effect of combat exposure was adjusted for, significant relationships generally diminished, implying that in this large population-based sample, it is the effect of combat exposure that is driving the mental health burden, not the exposure to antimalarial medication.” 2 The factors we adjusted for in these analyses are those that are well recognized (i.e., deployment and combat exposure) as potent predictors of mental health outcomes in veterans. 4,5…”

“…Our findings on the association between combat exposure and mental health are supported by other research on military and veteran populations 4,5 ; additional evidence for this association is presented in Table 5 of our article, where a monotonic relationship is observed between the level of self-reported combat intensity and the prevalence of screening positive for mental health symptoms. 2 Although some may suggest that regression analysis of cross-sectional data can help to infer causal associations between a dependent variable and selected independent variables, it is a widely acknowledged tenet in epidemiologic research that any reference to causal associations be viewed cautiously to minimize claims of direct associations that may be spurious in nature.…”

Bias and Confounding in Studies of Chronic Mental Health Effects from Mefloquine Exposure

“…I read with interest the article by Schneiderman et al, 1 which examined associations between self-reported mefloquine exposure and recent mental health symptoms among U.S. veterans enrolled in a population-based cohort study. Although the authors’ findings in unadjusted analysis reveal an increased risk that deployed veterans will screen positive for anxiety, depression, and posttraumatic stress disorder (PTSD)–like symptoms with mefloquine exposure, I am concerned that the magnitude of this observed association reflects the influence of information bias, specifically differential misclassification of a more appropriate exposure of interest.…”