Optic Neuritis and Vaccination Investigation: Failure to Consider Significant Sex Differences and Multiple Vaccine Combinations—Reply

In reply

We thank the authors for their letter and appreciate the opportunity to respond to their concerns on our published study evaluating anthrax vaccination and risk of optic neuritis in the US military.¹ As the authors point out, there is a consistent pattern of an increased incidence of optic neuritis for women vs men in the military. Approximately 15% of the active duty military population is female; however, in our study population, women constituted 30.7% of the analyzed optic neuritis cases. We estimate that women are about 2.5 times more likely than men to be diagnosed with optic neuritis in the military, consistent with the authors' figure. Because sex is likely a confounder, we controlled for it by including sex as a matched factor in our matched case-control analysis.

We considered a case-series approach when originally developing the study design. We estimated that vaccine coverage would be approximately 30% and risk–observation period ratio, based on the 18-week risk period, less than 10%. Had we used the self-matched case series method for this study, we would have substantially reduced the study power to detect an association for the 18-week interval, and this would be true even if we had left our risk period indefinite for the study period of 1998 through 2003. The case series controls for all fixed confounders, ie, confounders that don't change over time within a person. We chose to control for sex and several other potentially confounding factors by either matching on the effect or controlling for it in the conditional logistic regression model.

We are particularly aware that the quality of immunization data in the DMSS database is critical for our studies, and we have recently completed a comparative study of immunization data for services other than the Air Force contained in DMSS with immunization data abstracted from military medical charts (J. C. Davila, MPH; D. C. Payne, PhD, MSPH; Y. Zhang, PhD; C. E. Rose Jr, PhD; A. Aranas, MBA, MPH; B. Ruscio, DrPH; M. M. McNeil, MD, MPH; unpublished data, December 2004).² Our optic neuritis and anthrax vaccination study included assessments of vaccine-vaccine interactions and multivariable analyses of the 3 vaccines most likely to be administered at the same time as anthrax vaccine in our database (influenza, smallpox, and hepatitis B vaccines). In addition, we recognize that the topic of health effects due to receiving multiple concurrent immunizations in the military is particularly complex and challenging, and an initial study of this topic by the Vaccine Analytic Unit using the DMSS database has examined some possible health effects.³

In light of these authors' comments and further advances in the use of the case-series method, we are actively exploring the use of the method for subsequent anthrax vaccine safety studies, which will use the DMSS database and are included in the Vaccine Analytic Unit research agenda. We appreciate receiving input from these and other anthrax vaccine safety experts as we continue this challenging series of investigations.

Correspondence: Dr Payne, Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop A-47, Atlanta, GA 30333 (dvp6@cdc.gov).

Financial Disclosure: None reported.