Explicit change outcomes in the simple analyses were far fewer, grid B, with eight papers reporting mean differences, and 11 using standardised change outcomes (standardised effect sizes or response means).
For the individual studies in the meta-analyses, standardised effect sizes for each outcome measure were calculated as the standardised mean difference (SMD) of the change from baseline to endpoint.
OC analyses produced similar results for statistical significance and standardised effect sizes.
To account for the differing unit of measurement we anticipate across studies, we will calculate standardised effect sizes.
Instead, to assess the strength of the relationship between dependent and predictor variables, we calculated standardised effect sizes (partial r) from the t-values from multiple regressions [ 58].
Therefore with a large number of clusters available and sufficient numbers of individuals per cluster, trials are possible to detect small changes in proportions and standardised effect sizes.
Analyses of covariance (ANCOVA) will use multivariable linear regression models and standardised effect sizes calculated (adjusted mean difference divided by pooled baseline SD), with >0.5 considered a clinically meaningful effect.
Jacob Cohen, who originally defined standardised effect sizes, considered effect sizes of 0.2 or less to be small, 0.5 to be medium, and 0.8 and above to be large [ 20].
To evaluate the direction and size of any relationships in quantitative studies, standardised effect sizes were calculated when sufficient information was available to do so and an overall summary figure calculated (Field, 2005, p. 192).
At 12 month follow-up, mean PHQ-9 score was 1.36 points lower (95% confidence interval 0.07 to 2.64, P=0.04) in participants receiving collaborative care than in those receiving usual care (standardised effect size 0.28 (95% confidence interval 0.01 to 0.52)).
