Relative risk reduction

Guidelines for Chemical Process Quantitative Risk Analysis (CPQRA Guidelines) builds on the earlier work to show the engineer how to make quantitative estimates of the risk of the hazards identified. The quantitative estimates can identify the major contributors to risk. They can also help to define the most effective ways to a safer process by indicating relative risk reduction from proposed alternate process safeguards and measures. 

The Safety of Tirofiban in Acute Ischemic Stroke (SaTIS) trial examined 250 patients 6-22 hours after stroke onset treated with tirohban infusion or placebo for 48 hours. No increase in ICH was reported in the active group. Although no beneht in early functional recovery was observed, 5-6-month mortality was lower in the tirohban-treated group (relative risk reduction (RRR) 27%, 95% Cl 0.08-0.95, p = 0.03). 

Calculating the effect size of a therapeutic intervention is central (step 3 in Box 3.3). Different ways to calculate effects sizes can be applied as described in Table 3.2. All statements in this box actually describe the effect sizes correctly. Is the efficacy higher for drug A than for drug B Probably not since the relative risk reduction is not identical. Instead the result probably reflects other differences such as higher morbidity (blood pressure, other risk factors, or diseases) in case A. 

Reduced mortality by 50% Relative risk reduction From 100 to 50 deaths... 

The CURE study involved 12,562 patients randomized to receive Plavix (300 mg loading dose followed by 75 mg daily) or placebo and were treated for up to a year. Patients also received aspirin or other standard treatment such as heparin. The results showed that Plavix had a 20% relative risk reduction compared with placebo (582 cases of cardiovascular death, myocardial infarction, or stroke) versus 719 cases for placebo. 

Recent publications on major clinical trials whose implications will involve a recommendation to change clinical practice have included summary statistics that quantify the risk of benefit or harm that may occur if the results of a given trial are strictly applied to an individual patient or to a representative cohort. Four simple calculations will enable the non-statistician to answer the simple question How much better would my chances be (in terms of a particular outcome) if I took this new medicine, than if I did not take it . These calculations are the relative risk reduction, the absolute risk reduction, the number needed to treat, and the odds ratio (see Box 6.3). 

Moriarty PM. Using both relative risk reduction and number needed to treat in evaluating primary and secondary clinical trials of Upid reduction. Am Cardiol 2001 87 1206-8. 

In this study, CRT reduced the primary endpoint by 37% (hazard ratio 0.63 P < 0.001) and all-cause mortality by 36% (hazard ratio 0.64 P < 0.001). Although the relative risk reduction for mortality at first appears larger than that which trended in COMPANION (24% versus 36%), the mortality benefit was not evident in the early portion of the trial, but the benefit grew over time (see Fig. 4.3). Therefore, much, if not all, of the difference can be attributed to the longer follow-up (29.4 months in CARE-HF versus 14.8-16.5 months in COMPANION). 

The relative effect of treatment reflects the proportional difference between treatment groups in the incidence of disease events. In the Systolic Hypertension in the Elderly (SHEP) trial, the incidence of major CHD events over 4.5 years in patients assigned active treatment was 4.4% while in those assigned placebo it was 5.9%. This represents a relative risk of 0.73 or a relative risk reduction of 27%. 

The best predictor of absolute treatment effects for any individual patient will be provided by application of the estimate of the relative risk reduction from trials to an estimate of the absolute disease risk for the individual in question. 

In similar ways to the above we can obtain confidence intervals for a relative risk and for a relative risk reduction. Confidence intervals for NNT are a little more complicated see Grieve (2003) and Altman (1998) for further details. 

The estimated risk of dying in the first 12 months is then 0.114 in the cavedilol group compared to 0.185 in the placebo group. This enables the calculation of a relative risk at 12 months as 0.114/0.185 = 0.62 and the relative risk reduction is 38 per cent. Similar calculations can be undertaken at other time points. 

The CURE trial investigated the efficacy and safety of clopidogrel in 12,562 patients when administered together with aspirin in patients with ACS (UA or non-Q-wave Ml). The combination demonstrated a 20% relative risk reduction in the combined endpoints of Ml, stroke, or cardiovascular death compared with placebo (31). 

In the ESSENCE trial, the LMWH enoxaparin led to a relative risk reduction of 15% to 16% in the rate of death, Ml, or refractory ischemia as compared to unfractionated heparin at 30 days in UA/NSTEMI patients (38). Nadroparin [FRAXIS study (39)] and dalteparin [FRIC study (40)] did not demonstrate superiority against unfractionated heparin. Human pharmacokinetic data indicate that these differences in clinical efficacy might be explained by different elimination half-lives of antifactor Xa activity (dalteparin 2.8 hours, nadroparin 3.7 hours, enoxaparin 4.1 hours) (41). 

Several direct thrombin inhibitors have been studied in NSTEMI and STEM I patients and were compared to unfractionated heparin. In the GUSTO lib- and OASIS-2 trial (42,43), hirudin was studied versus heparin in patients with ACS. Despite early benefits, no statistical significance could be demonstrated at 30 days. Together with the OASIS-1 data, a combined analysis indicated a 22% relative risk reduction in cardiovascular death or Ml at 72 hours, 17% at 7 days, and 10% at 35 days (42). 

The major features of treatment guidelines concern indications for treatment, the particular treatments to be used, and in the case of biomedical risk factors, target levels following intervention. Figure 2 shows how these aspects are informed by clinical trials. In particular, net benefit depends on the absolute risk reduction (related to both baseline risk and relative risk reduction) and safety of the treatment. The greater the risk of the patient group or individual for future events, the greater is the absolute risk reduction with therapy (Fig. 3). Health policy decisions are informed not only by outcome data but also by cost-effectiveness analyses, Cost-effectiveness in turn relates to the absolute benefits that are observed. 

A second meta-analysis was based on data from individual patients in five long-term randomized trials, which assessed ACE inhibitors in 12,763 patients with left ventricular dysfunction or heart failure. The relative risk reduction of death, with a follow-up period exceeding 4 years, is 20% (13-26%). This means it is necessary to treat 26 patients for 5 years (95 % Cl 19-41) to avoid one death. If the combined end point used to assess the results is death plus reinfarction plus hospitalization for heart failure, it is necessary to treat 14 patients for 4 years to avoid one such event (95% Cl 12-18). 

The main argument in favor of a beneficial effect of ACE inhibition on coronary heart diseases comes from the pooled results of the SOLVE) treatment trial, the SOLVE) prevention trial, and the SAVE, AIRE, and TRACE studies, which indicate a 21% (95% Cl, 11-29%, p <. 001) relative risk reduction for myocardial infarction associated with ACE inhibitor therapy. Enalapril (SOLVE)) significantly reduced hospitalization for unstable angina, and captopril (SAVE) reduced revascularization procedures (291). In patients treated for 38 to 42 months with enalapril or captopril and selected on the basis of a reduction in ejection fraction with or without heart failure, it is necessary to treat 49 patients to avoid one myocardial infarction (95% Cl 32-117). 

---