1A) Using the data set, examine whether dichotomized age is acting as a confounder in the relationship between food insecurity and diabetes using the modeling (logistic regression) technique. We will use a 10% difference between the crude and adjusted OR as a guideline for presence of confounding.Fill in the table provided. Remember to only round the estimate at the last step. Estimates should be presented to two digits after the decimal.

Measure	Estimate	95% CI
Crude OR
Age-Adjusted OR
% difference

1B) Examine the results from the table you completed in question 6 assessing confounding by dichotomized age, with an adjusted logistic regression model.Compare the unadjusted and adjusted Odds Ratio. Is there evidence of confounding by dichotomized age of the relationship between food insecurity and diabetes, based on the percent difference guideline? Explain.

1C) You realize you should have examined your data with stratification using two by two tables, before you ran your adjusted logistic regression model. To better understand what is happening in your data set, you examine whether dichotomized age is a potential confounder in the relationship between food insecurity and diabetes. You use the two by two table stratification approach. Fill in the table. Remember to only round the estimate at the last step. Estimates should be presented to two digits after the decimal. Use the exact 95CI% test, rather than Woolf.

Measure	Estimate	95% CI
Unadjusted OR
Stratum specific OR
Less than 45
45 and Older

2. Compare the stratified two by two table Odds Ratio estimates to the crude Odds Ratio estimate (this is utilizing the data from the previous question).

A. Is there evidence of confounding or EMM by dichotomized age of the relationship of between food insecurity and diabetes?

B. Provide a rationale for your decision.

C. Interpret the confidence intervals.

D. Does the stratification two by two table conclusions agree or disagree with your modeling/logistic regression conclusion?

E. Based on your decision about whether dichotomized age is a confounder or EMM, how would you present the results to your colleagues?