Clinical instability on discharge and hospital readmissions in adult patients hospitalized with community-acquired pneumonia: a systematic review with meta-analysis

Aims: To assess the association between clinical instability at discharge and readmission for hospitalized adult patients with community-acquired pneumonia. Method: We performed a systematic review with meta-analysis of aggregated data. Primary studies were identified by searching Medline® and Central® databases from inception to February 15, 2016 and by screening references of potential eligible records. Eligible studies of all designs included at least 30 adult patients hospitalized for pneumonia. Pooled odds ratio was estimated by random effect model. Results: Of the 874 records retrieved from our search process, five studies were included in the meta-analysis totaling 3575 participants. Association between clinical instability and readmission was statistically significant (OR=1.39, 95% confidence interval 1.01 to 1.92, P=0.045). No evidence of between studies heterogeneity (I2=0.0%) was found despite different clinical instability definitions. This result was robust in sensitivity analysis by leave-one-out method. Data at disposal did not allow us to consider avoidable readmission apart. Conclusions: Presenting at least one criterion of clinical instability at discharge is associated with higher risk of readmission in adult patients hospitalized for community acquired pneumonia. However, as in any meta-analysis of data from observational studies, confounding effects cannot be discarded. A meta-analysis of individual participant data could allow the discussion of clinical stability best definition by considering each criterion relevance and threshold.