stratified_cox: Runs multiple referent group cox models using interaction...
In buddha2490/BERG: A set of standardized functions for ACS BERG analyses

Description Usage Arguments Details Value Author(s) References See Also Examples

The stratified_cox() function runs multiple referent group interaction cox models. The results will show the effect of our main exposure variable within each level of our stratification variable.

1	stratified_cox(dat,start,stop,outcome,expoVar,strataVar,age,covariates=NULL)

`dat`	data frame used in the analysis
`start`	Character vector naming your start of followup time, can be numeric or date format. Typically this is one of our DTINT variables. For time dependent or late-entry models, this will be the name of your start time variable.
`stop`	Character vector naming your end of followup time, can be numeric or date format. Typically this is one of our DATEFT or DATEDD variables. For time dependent or late-entry models, this will be the name of your stop time variable.
`outcome`	Character vector for the outcome variable used in the analysis, must be coded as 0=control, 1=case
`expoVar`	Character vector for our main exposure variable, may be categorical or numeric
`strataVar`	Character vector for our stratification variable, must be categorical
`age`	Character vector for our age variable used for stratifying on single year of age
`covariates`	Character vector of covariates. Default is NULL (age-adjusted analysis only)

The stratified_cox() function will run your stratified interaction models, format the results into a table, and return all model output in a list format that is suitable for program review. Function will work for standard, time-dependent, or late-entry models, depending on how the data are formatted prior to running the function.

Interaction models are coded using the following formula:

y <- formula(Surv(start,stop,outcome)~ strataVar + strataVar:expoVar + covariates + strata(age))

To calculate a p-interaction, a reduced model is also calculated:

y <- formula(Surv(start,stop,outcome) strataVar + expoVar + covariates + strata(age))

The p-interaction is then calculated:

anova(interaction.model, reduced.model)

The function outputs a 3-level list including all the output from the analysis.

`final`	A data frame containing the organized output of the function. Includes exposure name, categories, case numbers, stratified estimates and p-values, and a p-value for interaction.
`int.model`	All model output for the interaction model
`base.model`	All model output from the base model

Brian Carter

G:/Intramural Research/Epidemiology Research/Analysts EPI/Memos, presentations, resources and code/Interaction/Memo, Interaction Methods.doc

coxph, cox_models, interaction_cox

df <- example_data

foo <- stratified_cox(dat=df,
                      start="dtint92",
                      stop="dateft",
                      outcome="dead",
                      expoVar="smoke92",
                      strataVar="bmicat92",
                      age="age92m",
                      covariates=NULL)