Assessing risk of bias in a non‐randomized study

Cochrane Reviews often include non-randomized studies of interventions (NRSI). The Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool is recommended for assessing risk of bias in a NRSI. This chapter summarizes the biases that can affect NRSI and describes the main features of the ROBINS-I tool. There is greater potential for bias in NRSI than in randomized trials. A key concern is the possibility of confounding. NRSI may also be affected by biases that are referred to in the epidemiological literature as selection bias and information bias. Furthermore, the authors are at least as concerned about reporting biases as they are when including randomized trials. In some studies measurements of the outcome variable are made both before and after an intervention takes place. The chapter considers uncontrolled studies in which all units contributing to the analysis received the (same) intervention, and controlled versions of these studies.