Skip to main content

Table 3 Sensitivity analysis estimates of events prevented or postponed by factor, varied sequentially

From: Assessing the impact of heart failure specialist services on patient populations

Varied factor

Original estimate (source)

Varied estimate (source)

New estimate of number of deaths prevented or postponed (% change from previous "best" estimate-33)

New estimate of number of readmissions prevented or postponed (% change from previous "best" estimate-57)

b-blocker uptake

12% (local audit)

34% (ref. [46])

27 (-18%)

53 (-8%)

b-blocker intolerance

5.2% (ref. [29])

29% (ref. [4])

27 (-18%)

52 (-9%)

N-LEI eligibility

85% (informed judgement)

75% (conservative estimate)

33 (0%)

55 (-4%)

Annual risk of death

32% (data-linkage of HES to ONS mortality data)

29% (ref. [47])

30 (-9%)

-

Mean number of readmissions

0.54 (HES data analysis)

1.41 (ref. [5])

-

150 (+280%)

Estimate of number of patients that can benefit from treatment

286 (local hospital statistics data)

358 (assuming 25% of of heart failure patients are not coded with heart failure primary diagnosis)

41 (+25%)

72 (+25%)

RRR of all treatments

As per references [10,11,30,31]

20% reduction in all estimates (assuming that such a reduction in effectiveness might be applicable due to overall reduced baseline risk compared to that reported in original trials)

27 (-19%)

47 (-20%)

  1. HES: Hospital Episodes Statistics; ONS: Office for National Statistics; Ref.: Reference RRR: Relative Risk Reduction