The risk difference (RD), excess risk, or attributable risk[1] is the difference between the risk of an outcome in the exposed group and the unexposed group. It is computed as , where is the incidence in the exposed group, and is the incidence in the unexposed group. If the risk of an outcome is increased by the exposure, the term absolute risk increase (ARI) is used, and computed as . Equivalently, if the risk of an outcome is decreased by the exposure, the term absolute risk reduction (ARR) is used, and computed as .[2][3]

Illustration of two groups: one exposed to a risk factor, and one unexposed. Exposed group has smaller risk of adverse outcome (RD = −0.25, ARR = 0.25).
The adverse outcome (black) risk difference between the group exposed to the treatment (left) and the group unexposed to the treatment (right) is −0.25 (RD = −0.25, ARR = 0.25).

The inverse of the absolute risk reduction is the number needed to treat, and the inverse of the absolute risk increase is the number needed to harm.[2]

Usage in reporting

edit

It is recommended to use absolute measurements, such as risk difference, alongside the relative measurements, when presenting the results of randomized controlled trials.[4] Their utility can be illustrated by the following example of a hypothetical drug which reduces the risk of colon cancer from 1 case in 5000 to 1 case in 10,000 over one year. The relative risk reduction is 0.5 (50%), while the absolute risk reduction is 0.0001 (0.01%). The absolute risk reduction reflects the low probability of getting colon cancer in the first place, while reporting only relative risk reduction, would run into risk of readers exaggerating the effectiveness of the drug.[5]

Authors such as Ben Goldacre believe that the risk difference is best presented as a natural number - drug reduces 2 cases of colon cancer to 1 case if you treat 10,000 people. Natural numbers, which are used in the number needed to treat approach, are easily understood by non-experts.[6]

Inference

edit

Risk difference can be estimated from a 2x2 contingency table:

  Group
Experimental (E) Control (C)
Events (E) EE CE
Non-events (N) EN CN

The point estimate of the risk difference is

 

The sampling distribution of RD is approximately normal, with standard error

 

The   confidence interval for the RD is then

 

where   is the standard score for the chosen level of significance[3]

Bayesian interpretation

edit

We could assume a disease noted by  , and no disease noted by  , exposure noted by  , and no exposure noted by  . The risk difference can be written as

 

Numerical examples

edit

Risk reduction

edit
Example of risk reduction
Quantity Experimental group (E) Control group (C) Total
Events (E) EE = 15 CE = 100 115
Non-events (N) EN = 135 CN = 150 285
Total subjects (S) ES = EE + EN = 150 CS = CE + CN = 250 400
Event rate (ER) EER = EE / ES = 0.1, or 10% CER = CE / CS = 0.4, or 40%
Variable Abbr. Formula Value
Absolute risk reduction ARR CEREER 0.3, or 30%
Number needed to treat NNT 1 / (CEREER) 3.33
Relative risk (risk ratio) RR EER / CER 0.25
Relative risk reduction RRR (CEREER) / CER, or 1 − RR 0.75, or 75%
Preventable fraction among the unexposed PFu (CEREER) / CER 0.75
Odds ratio OR (EE / EN) / (CE / CN) 0.167

Risk increase

edit
Example of risk increase
Quantity Experimental group (E) Control group (C) Total
Events (E) EE = 75 CE = 100 175
Non-events (N) EN = 75 CN = 150 225
Total subjects (S) ES = EE + EN = 150 CS = CE + CN = 250 400
Event rate (ER) EER = EE / ES = 0.5, or 50% CER = CE / CS = 0.4, or 40%
Variable Abbr. Formula Value
Absolute risk increase ARI EERCER 0.1, or 10%
Number needed to harm NNH 1 / (EERCER) 10
Relative risk (risk ratio) RR EER / CER 1.25
Relative risk increase RRI (EERCER) / CER, or RR − 1 0.25, or 25%
Attributable fraction among the exposed AFe (EERCER) / EER 0.2
Odds ratio OR (EE / EN) / (CE / CN) 1.5

See also

edit

References

edit
  1. ^ Porta M, ed. (2014). Dictionary of Epidemiology (6th ed.). Oxford University Press. p. 14. doi:10.1093/acref/9780199976720.001.0001. ISBN 978-0-19-939006-9.
  2. ^ a b Porta, Miquel, ed. (2014). "Dictionary of Epidemiology - Oxford Reference". Oxford University Press. doi:10.1093/acref/9780199976720.001.0001. ISBN 9780199976720. Retrieved 2018-05-09.
  3. ^ a b J., Rothman, Kenneth (2012). Epidemiology : an introduction (2nd ed.). New York, NY: Oxford University Press. pp. 66, 160, 167. ISBN 9780199754557. OCLC 750986180.{{cite book}}: CS1 maint: multiple names: authors list (link)
  4. ^ Moher D, Hopewell S, Schulz KF, Montori V, Gøtzsche PC, Devereaux PJ, Elbourne D, Egger M, Altman DG (March 2010). "CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials". BMJ. 340: c869. doi:10.1136/bmj.c869. PMC 2844943. PMID 20332511.
  5. ^ Stegenga, Jacob (2015). "Measuring Effectiveness". Studies in History and Philosophy of Biological and Biomedical Sciences. 54: 62–71. doi:10.1016/j.shpsc.2015.06.003. PMID 26199055.
  6. ^ Ben Goldacre (2008). Bad Science. New York: Fourth Estate. pp. 239–260. ISBN 978-0-00-724019-7.
  NODES
eth 1
see 2
Story 1