how to compare the trend of risk ratios across age groups, between cases and controls

[Branu Ags]

Hi,

I used binomial regression to calculate the risk ratio of outcomes (e.g. diabetes, renal failure, heart failure) comparing cases and controls.
I did this for patients in 4 age groups
And this gave me the following table:

	18-29 years	30-39 years	40-49	50+	p value for the trend
Renal failure	5.60	2.30	3.50	2.20	?
Diabetes	1.70	1.70	1.70	1.50	?
Heart failure	85.40	29.50	19.90	8.20	?

Now my questions is, how do I calculate the p value of the trend in these risk ratios across the age groups? Its obvious that the difference in prevalence of diabetes between cases and controls stay the same across age groups and decrease dramatically for heart failure. How do I show if that difference is statistically significant?

Thank you

[Branu Ags]

Sorry I meant to say how do I calculate the p value for the trend in rish ratio of each outcomes and did not mean to say "How do I show if that difference is statistically significant?"

[Branu Ags]

Any suggestions? Thanks and please let me know if the above question is unclear and needs clarification.