Hi there - I have a dataset of individuals, where each individual is a close contact of an index case. So, individuals are clustered at the level of the index case. The outcome (infection) is yes/no - acquired infection from the index case.
Sample of dataset
IndexCase contact ContactCharacteristic IndexCaseCharacteristic infection
1 1 a a yes
1 2 b a yes
1 3 c a no
2 1 a b no
2 2 b b no
Question is: what are risk factors for contacts acquiring infection for the index case? Risk of infection for contacts depends on contact characteristics and index case characteristics (i.e. individual-level and cluster-level characteristics).
Is the coding for this model correct? i.e. is it ok to put the cluster-level characteristic of interest (indexcharacteristic) in the model?
proc genmod data = dataset;
class indexcase;
model infection = ContactCharacteristic IndexCharacteristic/ dist=binomial link=logit;
repeated subject=indexcase / type = exch;
run;
Thanks!
From your description, that seems like a reasonable model.
Join us for SAS Innovate 2025, our biggest and most exciting global event of the year, in Orlando, FL, from May 6-9.
Lock in the best rate now before the price increases on April 1.
ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.
Find more tutorials on the SAS Users YouTube channel.