Data collection and analysis

The two authors will independently assess the studies to be included to identify studies which meet the criteria outlined above. We will resolve any differences of opinion by discussion.

We will extract data onto standardised, pre‐piloted forms. We will contact study authors if necessary for clarification. If there is disagreement this will be resolved by discussion.

Both authors will independently assess the included trials for risk of bias. Assessment of the risk of bias of the included trials will be undertaken independently by JP and DM with the following to be taken into consideration, as guided by the Cochrane Handbook for Systematic Reviews of Interventions (Handbook 2008):

sequence generation;

allocation concealment;

blinding;

incomplete outcome data;

selective outcome reporting; and

other sources of bias.

We will use the Cochrane ‘Risk of bias’ tool in RevMan 5, which involves describing each of these domains as reported in the trial and then assigning a judgement about the adequacy of each entry. This involves answering a pre‐specified question whereby a judgement of ‘Yes’ indicates low risk of bias, ‘No’ indicates high risk of bias, and ‘Unclear’ indicates unclear or unknown risk of bias.

We will perform statistical analysis using Review Manager 5.0 (RevMan 2008). We will use both the Chi² and I² tests to formally assess statistical heterogeneity. It is hoped that data will be pooled for meta‐analysis, however this will only be performed in the absence of clear clinical and statistical heterogeneity. If it is not possible to pool data, we will present a narrative synthesis of the included studies. For dichotomous outcomes we will calculate a risk ratio (RR). We will use the mean difference (MD) or standardised mean difference (SMD) for continuous outcomes as appropriate.

Data allowing, we plan to conduct subgroup analysis of Tinnitus Retraining Therapy (TRT) versus non‐TRT forms of neurophysiological model‐based treatment for tinnitus.