Extract from Statistical Questions in Evidence-based Medicine by Martin Bland and Janet Peacock.
We hope that the topic will be useful in own right, as well as giving a flavour of the book.
Age 30-49 years at start | Back pain | No back pain | |
---|---|---|---|
Cause of death | (n=1274) | (n=586) | P value |
Ischaemic heart disease | 18.1 | 3.4 | 0.02 |
Stroke | 0.8 | 0.0 | 0.68 |
Other cardiovascular disease | 3.9 | 5.1 | 0.49 |
All Causes | 56.5 | 44.4 | 0.32 |
Age 50-66 years at start | Back pain | No back pain | |
Cause of death | (n=1212) | (n=576) | P value |
Ischaemic heart disease | 54.5 | 72.5 | 0.15 |
Stroke | 7.4 | 8.1 | 0.99 |
Other cardiovascular disease | 22.3 | 26.2 | 0.74 |
All Causes | 169.6 | 203.6 | 0.10 |
In women no association between back pain and any vascular disease was found. The author concluded that back pain may be an early manifestation of atheroschlerosis, which leads to ischaemic heart disease (Penttinen 1994).
1. Which subjects are missing from the table?
Check answer 1
2. How many tests of significance do you think the author
did?
Check answer 2
3. How many significant relationships would you expect to
find by chance in that number of tests if all the null hypotheses were true?
Check answer 3
!
4. What would you conclude from the significant
difference in younger men?
Check answer 4
+
5. How could we use these P values so as to test the null
hypothesis that back pain is unrelated to mortality?
Check answer 5
Reference: Penttinen, J. (1994) Back pain and risk of fatal ischaemic heart
disease: 13 year follow up of Finnish farmers. British Medical Journal
309 1267-8.
You can read the full paper
on the BMJ website. Use the browser's back button to get back to this page.
Back to Statistical Questions in Evidence-based Medicine contents.
Back to Martin Bland's home page.
This page maintained by Martin Bland.
Last updated: 30 June, 2005.