A key question needed to interpret the results of a clinical trial is whether the measured effect size is clinically important. Three commonly used measures of effect size are relative risk reduction (RRR), absolute risk reduction (ARR), and the number needed to treat (NNT) to prevent one bad outcome. These terms are defined below. The material in this section is adapted from Evidence-based medicine: How to practice and teach EBM by DL Sackett, WS Richardson, W Rosenberg and RB Haynes. 1997, New York: Churchill Livingston.
Consider the data from the Diabetes Control and Complications Trial (DCCT-Ann Intern Med 1995;122:561-8.). Neuropathy occurred in 9.6% of the usual care group and in 2.8% of the intensively treated group. These rates are sometimes referred to as risks by epidemiologists. For our purposes, risk can be thought of as the rate of some outcome.
Relative risk measures how much the risk is reduced in the experimental group compared to a control group. For example, if 60% of the control group died and 30% of the treated group died, the treatment would have a relative risk reduction of 0.5 or 50% (the rate of death in the treated group is half of that in the control group).
The formula for computing relative risk reduction is: (CER - EER)/CER. CER is the control group event rate and EER is the experimental group event rate. Using the DCCT data, this would work out to (0.096 - 0.028)/0.096 = 0.71 or 71%. This means that neuropathy was reduced by 71% in the intensive treatment group compared with the usual care group.
One problem with the relative risk measure is that without knowing the level of risk in the control group, one cannot assess the effect size in the treatment group. Treatments with very large relative risk reductions may have a small effect in conditions where the control group has a very low bad outcome rate. On the other hand, modest relative risk reductions can assume major clinical importance if the baseline (control) rate of bad outcomes is large.
Absolute risk reduction is just the absolute difference in outcome rates between the control and treatment groups: CER - EER. The absolute risk reduction does not involve an explicit comparison to the control group as in the relative risk reduction and thus, does not confound the effect size with the baseline risk. However, it is a less intuitve measure to interpret.
For the DCCT data, the absolute risk reduction for neuropathy would be (0.096 - 0.028) = 0.068 or 6.8%. This means that for every 100 patients enrolled in the intensive treatment group, about seven bad outcomes would be averted.
The number needed to treat is basically another way to express the absolute risk reduction. It is just 1/ARR and can be thought of as the number of patients that would need to be treated to prevent one additional bad outcome. For the DCCT data, NNT = 1/.068 = 14.7. Thus, for every 15 patients treated with intensive therapy, one case of neuropathy would be prevented.
The NNT concept has been gaining in popularity because of its simplicity to compute and its ease of interpretion. NNT data are especially useful in comparing the results of multiple clinical trials in which the relative effectiveness of the treatments are readily apparent. For example, the NNT to prevent stroke by treating patients with very high blood pressures (DBP 115-129) is only 3 but rises to 128 for patients with less severe hypertension (DBP 90-109).