Summary of outcomes from included studies
| Outcome | Trials | Intervention | Comparator | Significance |
| In-hospital mortality | DeVore 2015: n=165 hospitals randomised, 71 829 participants analysed. | Baseline mean (SE): 2.9% (0.26); postintervention absolute change mean (SE): 0.49% (0.44). | Baseline mean (SE): 2.9% (0.28). Postintervention absolute change mean (SE): 0.48% (0.34). | Estimated treatment effect (95% CI): −0.03 (95% CI −1.12 to 1.05); p=0.96. |
| Panella 2009: n=14 hospitals, 429 participants analysed. | No: 12/214, 5.6% (95% CI 2.5% to 8.7%). | No: 33/215. 15.4% (95% CI 10.5% to 20.2%). | P=0.001. | |
| Philbin 2000: n=10 hospitals, 2906 participants analysed. | Baseline mean: 5.9%. Postintervention mean: 5.2%. | Baseline mean: 5.4%. Postintervention mean: 3.7%. | Intervention effect (95% CI): 1.0% (-95% CI −3.0% to 5.0%); p=0.57. | |
| In-hospital medical therapy | Panella 2009: n=14 hospitals, 429 participants analysed. | ACE-I %: 57.9%. BB %: 46.7%. Diuretics %: 95.3%. | ACE-I %: 40.0%. BB %: 10.2%. Diuretics %: 95.8%. | P<0.001. P<0.001. P=0.8. |
| Discharge medical therapy | DeVore 2015: n=165 hospitals randomised, 71 829 participants analysed. | Baseline mean (SE): ACE-I/ARB* %: 92.8% (1.02). BB* %: 95.3% (0.76). AA* %: 16.2% (2.85). | Baseline mean (SE): ACE-I/ARB** %: 93.9% (1.11). BB* %: 95.0% (0.57). AA* %: 19.6% (2.70). | Estimated treatment effect (95% CI): ACE-I/ARB*: 0.75 (95% CI −2.66 to 4.16). P=0.67 |
| Postintervention absolute change mean (SE): ACE-I/ARB* %: 2.11% (1.30). BB* %: 1.54% (0.87). AA* %: −0.18% (1.75). | Postintervention absolute change mean (SE): ACE-I/ARB* %: 1.36% (1.16). BB* %: 0.49% (0.73). AA* %: 3.93% (2.37). | AA*: −4.10 (95% CI −9.95 to 1.74). P=0.17. | ||
| Hayes 2002: n=32 hospitals, 3276 participants eligible. | Difference between baseline and post-intervention %: ACE-I*: 3%. | Difference between baseline and postintervention %: ACE-I*: 3%. | No significance value reported. | |
| Laramee 2003: n=287 participants. | No (%) ACE-I/ARB: 121/141 (86%). No (%) BB: 91/141 (65%). | No (%) ACE-I/ARB: 115/145 (79%). No (%) BB: 89/145 (61%). | P=0.16. P=0.63. | |
| Panella 2009: n=14 hospitals, 429 participants analysed. | No (%) ACE-I*: 48/80 (60%). | No (%) ACE-I*: 5/12 (41.7%). | P=0.23. | |
| Philbin 2000: n=10 hospitals, 2906 participants analysed. | Baseline mean ACE-I*: 79%. Postintervention mean ACE-I*: 78%. | Baseline mean ACE-I*: 79% postintervention mean ACE-I*: 83%. | Intervention effect (95% CI): −7% (95% CI −20% to 5%). P=0.20. | |
| Tu 2009: n=86 hospitals, 17 544 participants analysed. | Absolute change % (95% CI): ACE-I/ARB*: 4.2% (95% CI 0.7% to 7.8%). BB*: 31.7% (95% CI 22.6% to 40.9%). | Absolute change % (95% CI): ACE-I/ARB*: −0.4% (95% CI −7.4% to 6.5%). BB*: 29.4% (95% CI 18.9% to 39.8%). | Absolute difference % (95% CI): 5.9% (95% CI 1.0% to 10.7%); P=0.02 3.5% (-6.1%, 13.1%); P=0.47 | |
| Hospital readmission (up to 90 days after discharge) | Laramee 2003: n=287 participants. | No (%): 49/131 (37%). | No (%): 46/125 (37%). | P>0.99. |
| Rich 1995: n=282 participants. | No (%): 53/142 (37%). | No (%): 94/140 (67%). | P=0.02. | |
| Sales 2013: n=137 participants. | No (%): 5/70 (7%). | No (%): 13/67 (19%). | Absolute risk reduction %: 12%. P=0.04. | |
| 30-day all-cause mortality | Sales 2013: n=137 participants. | No (%): 5/70 (7%). | No (%): 5/67 (7%). | Absolute risk reduction %: 0.4%. P=1.00. |
| Tu 2009: n=86 hospitals, 17 544 participants analysed. | Absolute change % (95% CI): −1.7% (95% CI −3.4% to 0%). | Absolute change % (95% CI): 0.2% (95% CI −1.6% to 1.9%). | Absolute difference % (95% CI): −1.1% (95% CI −3.2% to 0.9%); p=0.26. | |
| Hospital length of stay | Panella 2009: n=14 hospitals, 429 participants analysed. | Mean (95% CI) days: 10.4 (95% CI 9.6 to 11.0). | Mean (95% CI) days: 11.4 (95% CI 10.5 to 12.3). | P=0.028. |
| Philbin 2000: n=10 hospitals, 2906 participants analysed. | Baseline mean days: 8.0. Postintervention mean days: 6.2. | Baseline mean days: 7.7. Postintervention mean days: 7.0. | Intervention effect (95% CI): −1.1 (95% CI −2.9 to 0.7) days. | |
| Patient-level health-related quality of life | Koelling 2005: n=223 participants. | MLHF score baseline mean (SD): 56 (23). MLHF score 30-day mean (SD): 38 (22). MLHF score 180-day mean (SD): 41 (22). | MLHF score baseline mean (SD): 59 (22). MLHF score 30-day mean (SD): 45 (25). MLHF score 180-day mean (SD): 42 (25). | No significance value reported. P=0.049. No significance value reported. |
| Philbin 2000: n=10 hospitals, 2906 participants analysed. | LOL score baseline mean: 6.6. LOL score postintervention mean: 6.5. | LOL score baseline mean: 6.3. LOL score postintervention mean: 6.5. | Intervention effect (95% CI): −0.3 (95% CI −1.6 to 1.0). | |
| NYHA class baseline mean: 2.2. | NYHA class baseline mean: 2.4. | 0 (95% CI −0.4 to 0.3). | ||
| NYHA class postintervention mean: 2.1. | NYHA class post-intervention mean: 2.4. | P=0.88. | ||
| Rich 1995: n=282 participants.† | CHFQ mean (SD) change: 22.1 (20.8). | CHFQ mean (SD) change: 11.3 (16.4). | P=0.001. |
↵*In patients with reduced left ventricular ejection fraction (ranging from ≤40.0% or <40.0%).
†CHFQ assessed in subset of 126 participants.
AA, aldosterone antagonist; ACE-I, ACE inhibitor; ARB, angiotensin receptor blocker; BB, beta-blocker; CHFQ, Chronic Heart Failure Questionnaire (higher scores reflect better quality of life); LOL, Ladder of Life (higher scores reflect better quality of life); MLHF, Minnesota Living with Heart Failure Questionnaire (lower scores reflect better quality of life); NYHA, New York Heart Association.