Online Program

A Review and Meta-analysis of 15 Studies Reporting Reactive Arthritis from Non-typhoidal Salmonella Infections

*Zhouyang Weng, University of Cincinnati, Department of Environmental Health 
Robert Herrick, University of Cincinnati, Department of Environmental Health 
Barbara Kowalcyk, The Center For Foodborne illness Research&Prevention 
Linda Levin, University of Cincinnati, Department of Environmental Health 
Susan Pinney, University of Cincinnati, Department of Environmental Health 

Keywords: Reactive Arthritis Salmonella Infections Meta Analysis

Objective: Rather than relying on personal opinion, a consensus of experts, or available research, today’s policy makers tend to base their decisions on the results of a systematic, quantitatively based synthesis. Meta analysis, as the chief research synthesis method for quantitatively integrating prior evidence, has been widely applied to the process of public health policy making. Reactive arthritis (ReA) is a well-recognized sequellae of Salmonella infection; however, population-based incidence estimates are needed to quantify the long-term burden of disease. The objective of this study was to review population-based clinical studies and to quantitatively and qualitatively determine the incidence rate of ReA at least six months after a foodborne non-typhoidal Salmonella infection using a meta-analysis. Methods: A comprehensive search of computerized databases from identified articles was performed. Abstract reviews were used to exclude studies outside the scope of the review and categorized potentially relevant studies from either population based case-control or cohort studies. Articles were qualitatively assessed for a homogeneous definition of ReA. Point estimates and confidence intervals for proportions of ReA cases among the patients with Salmonella infections were developed using binomial data. Results: 15 primary research studies of ReA related to Salmonella infection were identified. The findings indicate that incidence rate of ReA cases among patients with acute Salmonella infection is 5.51% (95% CI=0.03-0.12) and 5.30% (95% CI is 0.01-0.09) estimated from random effects models based on binomial data. Bayesian Analysis also confirmed the previous findings with an incidence rate of 5.51% (95% CI is 0.02-0.13) based on Normal prior. Conclusion: These findings present the first estimate of ReA incidence at least six months subsequent to Salmonella infection.