Table 7.

 Analysis by mean summed score of items for reporting guideline checklists, endorsing versus non-endorsing journals*

Reporting guideline†	No of evaluations‡	No of studies (total)	Effect estimate (99% CI)
CONSORT extension for harms, 2004	1§	25 v 77 (102)	Mean difference 0.04 (–1.50 to 1.58)
PRISMA, 2009	3¶	63 v 80 (143)	Standardized mean difference 0.53 (0.02 to 1.03)
STARD, 2003	3**	23 v 65 (88)	Standardized mean difference 0.52 (–0.11 to 1.16)
STRICTA, 2002	1††	17 v 130 (147)	Mean difference 1.42 (–0.04 to 2.88)
STROBE, 2007	1§	9 v 38 (47)	Mean difference 1.55 (–3.19 to 6.29)

*Individual forest plots depicting these summary data are shown in appendix 7.

†QUOROM (two evaluations) was not estimable because of one study in one comparison arm per assessed evaluation.

‡Only evaluations that calculated summed score for report were included.

§All checklist items summed.

¶Subset of items was summed for one evaluation.

**Subset of items was summed for two of three evaluations.

††Subset of items was summed.