Risk Factors for RSV Hospitalization in Healthy Preterm Infants: A Meta-analysis
We used RevMan software (Version 5.3, The Nordic Cochrane Centre, Copenhagen) to estimate the pooled random-effects and fixed-effects treatment estimates using the inverse-variance method and, more specifically, Mantel-Haenszel fixed effect method, and the DerSimonian and Laird method for random effects.2 We evaluated heterogeneity by comparing study designs through visual inspection of forest plots, the I2 statistic, and the χ2 Cochrane Q heterogeneity test. Primary analyses included all studies with available data. We included 4 cohort studies and 1 case–control study.1
Review of the evidence revealed 3 risk factors amenable to pooling: (1) presence of school-age siblings (herein, “siblings”); (2) born close to start of RSV season or early in the season (herein, “young_age”) and (3) family history of atopy (“herein atopy”).
Meta-analysis of the available quantitative evidence identified from the systematic literature review indicated that the presence of siblings and born close to the start of or early in the RSV season are significant risk factors for RSV infection requiring hospitalization (see Table 1). No relationship between family history of atopy and RSV hospitalization could be established.
Our meta-analyses had some limitations. The first is the small size of the evidence base for our target population, in terms of numbers of included studies and risk factors amenable to pooling. In addition, all data were drawn from observational studies, which are more likely than randomized controlled trials to exhibit systematic bias and imbalanced comparison groups. All study estimates were covariate-adjusted, but there was no adjustment for unknown or unreported confounding factors.