if a study reports a risk of 4 cases per 1000 person months can that be converted into an annual risk per 100,000 people? if so, how? Thanks
$\begingroup$
$\endgroup$
3
-
$\begingroup$ The solution to questions like this is offered by dimensional analysis, which reduces it to the multiplications and divisions needed to convert months to years and thousands to hundreds of thousands. $\endgroup$– whuber ♦Commented Jul 16, 2018 at 13:02
-
$\begingroup$ I've seen other studies that indicate that the risk for a certain disease is 70 cases per 100,000 people. And, as mentioned before, I have seen a study that reports 4 cases per 1000 person months for the same disease. How do I compare these two different estimates of risk? How do I know if they are similar? Thank you very much. $\endgroup$– MHuckCommented Jul 16, 2018 at 13:46
-
$\begingroup$ Are the risks as reported these two ways based on the same population? $\endgroup$– EdMCommented Jul 16, 2018 at 18:51
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
The units of the reported value are $(1000 person months)^{-1}$.
The units you want to convert to are annual risk per 100,000 people. Annual risk means you are measuring it on a per year basis, so $(100,000 person years)^{-1}$.
Since you want it per 100 times the people and 12 times the interval your multiply by 1200, so the final value is an annual risk of 4800 per 100,000 people.
-
$\begingroup$ Thanks. The person-month risk is reported to be 4 per 1000 person months. Converting the person- month estimate to an annual rate leads to an estimate of 4800 per 100,000? As I mentioned in my original question, the reported annual rate is estimated at 70 per 100,000. Huge difference. Something is wrong. $\endgroup$– MHuckCommented Jul 16, 2018 at 15:29
-
$\begingroup$ That isn't in your question and the comment hadn't appeared when I answered. That is a huge difference, look at the methodology for the studies. Any clues there? $\endgroup$– ReneBtCommented Jul 17, 2018 at 7:35