II. Approach
- See Contingency Grid for description of setting up this grid
- Given- We are studying whether an event is less likely to occur in an experimental group, then in a control group
 
- Create a hypothetical grid- Label the grid top- Event (or E+)
- No Event (or E-)
- Total patients
 
- Label the grid left- Control (or T-)
- Experimental (or T+)
- Total patients
 
- Fill in the next row (Control Row)- Event = a
- No Event = b
- Total = a + b
 
- Fill in the top row (Experimental Row)- Event = c
- No Event = d
- Total = c + d
 
- Fill in the bottom row (Total row)- Event = a + c
- No Event = b + d
- Total = a + b + c + d
 
 
- Label the grid top
- Calculations- Control event rate (CER)- Risk of event from control intervention = a / (a+b)
 
- Experimental event rate (EER)- Risk of event from experimental intervention = c / (c+d)
 
- Relative Risk (RR)- RR = EER / CER
- Hazard Ratio (HR) is a similar concept related to survival
 
- Relative Risk Reduction (RRR)- RRR = 1 - RR
 
- Absolute Risk Reduction (ARR)- ARR = CER - EER
 
- Number Needed to Treat (NNT)- NNT = 1 / ARR
 
 
- Control event rate (CER)
III. Example: Dabigatran versus Warfarin in CVA Prevention
- See Contingency Grid for description of setting up this grid
- Given- Example: Does CVA occur less often in Atrial Fibrillation with Dabigatran (experimental) than with Warfarin (control)
- Connolly (2009) N Engl J Med 361:1139-51 [PubMed]
 
- Create a grid- Label the grid top- CVA (or E+)
- No CVA (or E-)
- Total patients
 
- Label the grid left- Warfarin (or T-)
- Dabigatran 150 mg (or T+)
- Total patients
 
- Fill in the next row (Warfarin Row)- CVA = a = 199
- No CVA = b = 5823
- Total = a + b = 6022
 
- Fill in the top row (Dabigatran Row)- CVA = c = 134
- No CVA = d = 5942
- Total = c + d = 6076
 
- Fill in the bottom row (Total row)- CVA = a + c = 333
- No CVA = b + d = 11765
- Total = a + b + c + d = 12098
 
 
- Label the grid top
- Calculations- Warfarin control event rate (CER)- Risk of CVA on Warfarin = a / (a+b) = 199/6022 = 0.033 (3.3%)
 
- Dabigatran experimental event rate (EER)- Risk of event from experimental intervention = c / (c+d) = 134/6076 = 0.022 (2.2%)
 
- Relative Risk (RR)- RR = EER / CER = 0.022 / 0.033 = 0.667
 
- Relative Risk Reduction (RRR)- RRR = 1 - RR = 1 - 0.667 = 0.333
 
- Absolute Risk Reduction (ARR)- ARR = CER - EER = 0.033 - 0.022 = 0.011
 
- Number Needed to Treat (NNT)- NNT = 1 / ARR = 1 / 0.011 = 90.9
 
 
- Warfarin control event rate (CER)
IV. References
- Hersh (2014) Evidence Based Medicine, AMIA’s CIBRC Online Course
