In the case of a case-control study, I can adjust the odds ratio for age, sex, BMI, clinicopathological data, etc., in determining the odds ratio for genotype data with logistic regression/multinomial logistic regression in SPSS. However, in the case of allele data, as an individual can carry two same (wild/variant) or two different alleles (heterozygote), how can I adjust the odds ratio with covariates (age, sex, BMI, clinicopathological data, etc.)?
If I understand your question, the attached paper may be of some interest. Best wishes, David Booth PS recall the definition of odds ratio compared to the form which we have in this paper. Basically the adjustment is as what we do in predictive models converted to odds ratio language. See the attached screenshot search for details
Hello, Mohammad,
As a naming convention, 'Adjusted Odds Ratio' (AOR) is simply the OR computed using multiple 'right hand' variables (aka independent, predictor, etc).
'age, sex, BMI, clinicopathological data' usually are such predictor variables (covariates) so I don't understand how one can adjust the odds ratio for those?
However, if I correctly interpret, the question may be related to the fact that in the second situation you have two outcome variables instead of just one ('two same, or two different alleles' - as you say), so you should employ the multinomial logistic regression, instead of the binary ('simple') logistic regression - allowing for more than two categories on the dependent variable.
(Sorry if I misunderstood your problem!)
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics
- Survey Design
- Questionnaire