To find the absolute risk in the in intervention group using the below formulas.
From the intervention risk we calculate the absolute difference by a simple subtraction.
-------- Relative risk ---------
RR*Baseline risk
-------- Odds ratio -------
We have taken the calculation we use from the Cochrane handbook, chapter 12.5.4.3 . It calculates the number fewer per 1000 (the difference), and we calculate the absolute intervention risk by adding the difference (whether more or fewer) to the baseline risk.
The calculation from the Cochrane handbook uses the baseline risk (called 'assumed control risk (ACR) in the calculation):
{{ "http://magicapp.uservoice.com/assets/106588467/Skjermbilde%202016-08-05%20kl.%2020.54.44.png" | embed }}
As an example, suppose the odds ratio is OR = 0.73, and a control risk of ACR = 0.3 is assumed. Then the effect on risk is 62 fewer per 1000:
{{ "http://magicapp.uservoice.com/assets/106589100/Skjermbilde%202016-08-05%20kl.%2021.05.42.png" | embed }}
------- Hazard ratio ---------
Hazard rates are calculated assuming your outcome is an event (e.g mortality), and not a non-event (e.g survival). We do this as having the outcome stated as an event is more common than a non-event.
That means that we use the inverse of the formula explained by Patrick Bossuyt in an article from 2011 (1).
We use:
Intervention rate of events =1- (1-baseline rate of events)Hazard ratio
While the formula explained in the Bossuyt article it is references as:
Intervention rate of events = baseline rate of eventsHazard ratio
If you use a non-event as your outcome, you will have to calculate the absolute risk manually
________________________________________________________________
1) Survival and Hazard Rate Ratios, written by Patrick Bossuyt in 2011
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article