OBJECTIVE: To systematically search, summarize, and critically appraise the literature to examine whether pet exposure in early life is associated with an increased risk of eczema. DATA SOURCES AND STUDY SELECTION: We searched MEDLINE (1950 to June 2006) supplemented by citation lists in retrieved articles and contact with researchers. No language restrictions were imposed. DATA EXTRACTION: Cohort studies were sufficiently similar to allow pooled analysis. Meta-analysis was not possible for cross-sectional studies owing to differences in methods and populations. MAIN OUTCOME MEASURES: Incidence or prevalence of eczema. RESULTS: Evidence from longitudinal studies showed that previous exposure to cats (pooled odds ratio [OR], 0.76; 95% confidence interval [CI], 0.62-0.92), dogs (pooled OR, 0.68; 95% CI, 0.53-0.87), or "any furry pet" (pooled OR, 0.79; 95% CI, 0.74-0.84) is associated with a lower risk of eczema. However, in the only cohort study adjusted for avoidance behavior, this "protective effect" disappeared (for cats: OR, 0.80; 95% CI, 0.33-1.97). Stratified analysis by family history in 2 birth cohort studies showed that dog exposure was protective in patients with atopic families. For cats, 1 study showed reduced risk in atopic families only; the other study showed no effect. Eight cross-sectional studies evaluated past pet exposure; a protective effect was seen in 3 studies for cat, dog, or any pet; no study demonstrated an increased risk. CONCLUSIONS: There was no clear evidence that early pet exposure is associated with increased risks of subsequent eczema. We found some evidence of a possible protective effect of early pet exposure, but this might be explained by avoidance behavior in high-risk families.