Table A6: GRADE Evidence Profile for Comparison of Predischarge Planning Care and Usual Care.

No. of Studies (Design)	Risk of Bias	Inconsistency	Indirectness	Imprecision	Publication Bias	Upgrade Considerations	Quality
Readmissions
2 systematic reviews of RCTs	Some serious limitations (-1)a	No serious limitationsb	No serious limitations	No serious limitations	Undetected	None	⊕⊕⊕ Moderate
Length of Stay
1 systematic review of RCTs	Some serious limitations (-1)c	No serious limitationsd	No serious limitations	No serious limitations	Undetected	None	⊕⊕⊕ Moderate
Mortality/Survival
1 systematic review of RCTs	Some serious limitations (-1)e	No serious limitationsf	No serious limitations	No serious limitations	Undetected	None	⊕⊕⊕ Moderate
HRQOL
1 systematic review of RCTs	Very serious limitations (-2)g	Some serious limitations (-1)h	No serious limitations	No serious limitations	Undetected	None	⊕ Very Low
Patient Satisfaction
1 systematic review of RCTs	Very serious limitations (-2)i	Some serious limitations (-1)j	No serious limitations	No serious limitations	Undetected	None	⊕ Very Low

Abbreviations: EPOC, Effective Practice and Organization of Care Group; No., number; RCT, randomized controlled trial

^aAverage EPOC Risk of Bias score in studies included in systematic review by Hansen et al was 5 out of 9.

The systematic review by Shepperd et al focused on discharge planning and excluded RCTs evaluating interventions where discharge planning was not the main focus of a multifaceted package of care. It was not possible to assess how some components of the process compared between trials (e.g., inclusion of caregivers and the extent of their care). Adequate sequence generation and allocation concealment were reported in 14/21 and 12/21 trials respectively.

^bShepperd et al found a significant difference in readmission favouring discharge planning versus usual care. Hansen et al did not conduct a meta-analysis due to heterogeneity among the included studies and could not make a conclusion as to which comprehensive discharge bundle/package was most effective compared with usual care.

^cThe systematic review by Shepperd et al focused on discharge planning and excluded RCTs evaluating interventions where discharge planning was not the main focus of a multifaceted package of care. It was not possible to assess how some components of the process compared between trials (e.g., inclusion of caregivers and the extent of their care). Adequate sequence generation and allocation concealment were reported in 14/21 and 12/21 trials respectively.

^dShepperd et al found a significant difference in hospital LOS favouring discharge planning. Phillips et al (11) did not find a significant difference in hospital LOS between the comprehensive discharge planning and postdischarge follow-up and usual care. Not all studies in the systematic reviews reported on hospital LOS.

^eThe systematic review by Shepperd et al focused on discharge planning and excluded RCTs evaluating interventions where discharge planning was not the main focus of a multifaceted package of care. It was not possible to assess how some components of the process compared between trials (e.g., inclusion of caregivers and the extent of their care). Adequate sequence generation and allocation concealment were reported in 14/21 and 12/21 trials respectively.

^fShepperd et al did not find a significant difference in mortality between study arms. No significant heterogeneity in summary statistic.

^gThe systematic review by Shepperd et al focused on discharge planning and excluded RCTs evaluating interventions where discharge planning was not the main focus of a multifaceted package of care. It was not possible to assess how some components of the process compared between trials (e.g., inclusion of caregivers and the extent of their care). Adequate sequence generation and allocation concealment were reported in 14/21 and 12/21 trials respectively. HRQOL was a secondary endpoint in 3 studies that reported this outcome and measured using different scales in subgroups of patients.

^hA meta-analysis was not conducted by Shepperd et al for the HRQOL outcome due to the heterogeneity and diverse measurement techniques used by the 3 individual studies. One study reported no significant difference between the study arms. Another study only provided HRQOL data for baseline measurements. A third study showed a significant difference between study arms at 26 weeks follow-up in favour of the control group.

ⁱThis outcome was reported in 3 studies in the systematic review by Shepperd et al and a meta-analysis was not conducted. Satisfaction was reported as a secondary outcome and performed on subgroups of patients using different measurement scales.

^jTwo studies reported a significant difference between study arms, one study did not.