Ref (type)	Population	Outcome, Interventions	Results and statistical analysis	Effect size	Favours
All-cause mortality
[81] Systematic review	4909 people 8 RCTs in this analysis	Rate of mortality 459/2428 (19%) with implantable cardiac defibrillator (ICD) 695/2481 (28%) with usual care	RR 0.74 95% CI 0.67 to 0.82	Small effect size	ICD
[81] Systematic review	2946 people with evidence of heart failure or coronary artery disease (primary prevention) 5 RCTs in this analysis	Rate of mortality 260/1494 (17%) with ICD 391/1452 (27%) with usual care	RR 0.72 95% CI 0.63 to 0.84 The magnitude of absolute-mortality benefit increased with increasing baseline risk of sudden cardiac death There was significant heterogeneity among RCTs because 3 RCTs were in people at high risk of heart failure and 2 RCTs were in people at moderate risk of heart failure	Small effect size	ICD
[81] Systematic review	1963 people who had survived sudden cardiac death or had unstable ventricular rhythm (secondary prevention) 3 RCTs in this analysis	Rate of mortality 199/934 (21%) with ICD 304/1029 (30%) with usual care	RR 0.76 95% CI 0.65 to 0.89	Small effect size	ICD
[82] Systematic review	2110 people with heart failure caused by non-ischaemic cardiomyopathy 7 RCTs in this analysis	Rate of mortality with ICD with usual care Absolute results not reported	RR 0.69 95% CI 0.56 to 0.86	Small effect size	ICD
[82] Systematic review	1457 people with heart failure caused by non-ischaemic cardiomyopathy (primary prevention) 4 RCTs in this analysis	Rate of mortality with ICD with usual care Absolute results not reported	RR 0.74 95% CI 0.58 to 0.96	Small effect size	ICD
[82] Systematic review	256 people with previous resuscitated cardiac arrest or symptomatic ventricular tachycardia (secondary prevention) 2 RCTs in this analysis	Rate of mortality with ICD with usual care Absolute results not reported	RR 0.69 95% CI 0.39 to 1.24 The number analysed may have been too small to detect a significant difference		Not significant
[83] Systematic review	934 women with heart failure and reduced left ventricular ejection fraction 5 RCTs in this analysis	Mortality with ICD with usual care Absolute results reported graphically	HR 1.01 95% CI 0.76 to 1.33 P = 0.95		Not significant
Cardiac mortality
[81] Systematic review	4909 people 8 RCTs in this analysis	Rate of cardiac mortality 124/2428 (5%) with ICD 339/2481 (14%) with usual care	RR 0.43 95% CI 0.35 to 0.53	Moderate effect size	ICD
[81] Systematic review	2946 people with evidence of heart failure or coronary artery disease (primary prevention) 5 RCTs in this analysis	Rate of cardiac mortality 57/1494 (4%) with ICD 177/1452 (12%) with usual care	RR 0.37 95% CI 0.27 to 0.50 The magnitude of absolute-mortality benefit increased with increasing baseline risk of sudden cardiac death There was significant heterogeneity among RCTs because 3 RCTs were in people at high risk of heart failure and 2 RCTs were in people at moderate risk of heart failure	Moderate effect size	ICD
[81] Systematic review	1963 people who had survived sudden cardiac death or had unstable ventricular rhythm (secondary prevention) 3 RCTs in this analysis	Rate of cardiac mortality 67/934 (7%) with ICD 162/1029 (16%) with usual care	RR 0.50 95% CI 0.38 to 0.66	Moderate effect size	ICD