Table 11.

Classification of study quality

Good quality Low risk of bias and no obvious reporting errors; complete reporting of data. Must be prospective. If study of intervention, must be RCT.

Fair quality Moderate risk of bias, but problems with study or paper are unlikely to cause major bias. If study of intervention, must be prospective.

Poor quality High risk of bias or cannot rule out possible significant biases. Poor methods, incomplete data, reporting errors. Prospective or retrospective.

RCT, randomized controlled trial.