How to do sensitivity analysis for residual confounding ?

[Kyra]

I have fit separate logistic regression models with head and neck cancer as outcome , job title as exposure ( I have 23 models each with different jobs). I adjusted each of them for age, sex, race, region, tobacco, smoking.

I wonder how to do sensitivity analysis for residual confounding in SAS. 

[PGStats]

How is Job exposure coded in your models?

[Kyra]

Thank you for the reply.

I have coded job exposure as 1 vs 0.

Below I have posted codes for two of my job exposures: cook and painter. I have modeled 20 more jobs similarly. And for all the job exposures the 0 or reference is same group of people( people who were never exposed to all of the job of interest.

proc logistic data=red.merged;

class studyname ageq(ref="<40") sex(ref="Female") Education_levelmi1(ref="No education") race region drink_dayq(ref="0 (Non-drinkers)") tobaccoyearq(ref="0 (Never smoker)") cook(ref="0")/param=reference;

model oropharynx= studyname ageq sex Education_levelmi1 race region drink_dayq tobaccoyearq cook/ scale=none aggregate;

run;

proc logistic data=red.merged3;

class studyname ageq(ref="<40") sex(ref="Female") Education_levelmi1(ref="No education") race region drink_dayq(ref="0 (Non-drinkers)") tobaccoyearq(ref="0 (Never smoker)") painter(ref="0")/param=reference;

model hypopharynx= studyname ageq sex Education_levelmi1 race region drink_dayq tobaccoyearq painter/ scale=none aggregate;

run;