Brian
Thanks for the swift and helpful response
The case-control study was nested in a larger cohort study with 228,660
patients. So I could work out the incidence of heart failure over the
study period. But I am not confident that this would give a meaningful
NNTh. There are not only multiple times for heart failure to occur, but
also withdrawals and exclusions.
Michael
-----Original Message-----
From: Brian Alper MD [mailto:[log in to unmask]]
Sent: 28 February 2008 10:56
To: Michael Power
Cc: [log in to unmask]
Subject: RE: NNT calculation
You cannot calculate NNT directly from a case-control study.
Because the study starts with outcomes (cases) and examines rates of
exposure it does not directly provide the risk for the outcome.
The only way to estimate an NNT would be to obtain an estimate of risk
or prevalence of the outcome from other data, then extrapolate from
"relative risk" data, but remember relative risk may not be similar to
odds ratios.
--------------------------------------
Brian S. Alper, MD, MSPH
Editor-in-Chief, DynaMed (www.DynamicMedical.com)
Medical Director, EBSCO Publishing
10 Estes St.
Ipswich, MA 01938
office (978) 356-6500 extension 2749
cell (978) 804-8719
fax (978) 356-6565
home (978) 356-3266
"It only takes a pebble to start an avalanche."
-----Original Message-----
From: Evidence based health (EBH)
[mailto:[log in to unmask]] On Behalf Of Michael Power
Sent: Thursday, February 28, 2008 4:05 AM
To: [log in to unmask]
Subject: NNT calculation
Is it possible to calculate meaningful NNTs from the following data?
If so, I would be very grateful for advice on how to do this, if
preferably using my 2x2 calculator. I know that it is easy to choose the
wrong numbers for denominators and numerators in very simple statistical
calculations!
From a case-control study: Huerta et al. Non-steroidal anti-inflammatory
drugs and risk of first hospital admission for heart failure in the
general population. Heart. 2006 Nov;92(11):1610-5.
Risk of hospitalization for heart failure associated with use of NSAIDs
Cases Controls
Number 1396 5000
Duration of NSAIDs use:
Non-use 487 2014
1-30 days 49 105
31-365 days 54 173
365-730days 28 62
>730 days 65 184
Thanks very much.
Michael
Clinical Knowledge Author, Guideline Developer and Informatician
NHS Clinical Knowledge Summaries Service www.cks.library.nhs.uk
Sowerby Centre for Health Informatics at Newcastle Ltd www.schin.co.uk
|