Double jeopardy
A study with 90000 people examining the relationship between a mild pain killer and reduced risk of insomnia reports a relative risk of 0.89 with a 95% confidence interval of 0.78 to 0.97. How would you caution interpreting this finding in the context of this study.
it is statistically significant, but large sample sizes can make almost all relationships significant.
If the sensitivity of a test is 65% and the specificity of the test is 59%, is the test good at detecting potential cases of disease?
The accuracy of the screener test is low, so there is need for caution in using this test as it is not often identifying those with the disease, nor is it often identifying those without the disease and could result in substantial negative consequences such as undue stress, medical costs, etc.