
Methods: We performed a clustered seroprevalence study of the areas around Vellore in Tamil Nadu, India. All participants completed a risk factor survey, with seropositive and seronegative participants acting as cases and controls, respectively, in a risk factor analysis. After univariate analysis, accounting for clustering, variables found to be significant were subsequently used in a multivariate analysis.
Results: A total of 721 people participated in this study. Of these, 31.8% were found to be seropositive. After accounting for clustering, having fewer people living in a household, having a fewer rooms in a house, having a house that was clustered with other houses, defecating outside, female sex, age >60 years, shorter height, lower weight, smaller body mass index, and smaller mid upper arm circumference were found to be significantly associated with seropositivity. After multivariate regression modeling, living in a house clustered with other houses, female sex, and age >60 years were still associated with scrub typhus exposure, with adjusted odds ratios of 2.47, 1.80, and 3.50, respectively, and p=0.02 for each.
Conclusion: Overall, scrub typhus is much more common than previously thought. Previously described individual environmental and habitual risk factors seem to have less importance in South India, perhaps because of the overall scrub typhus-conducive nature of the environment in this region.

P. Trowbridge,
None
P. Samuel, None
V. K, None
G. Varghese, None