URGENT DEADLINE instrument variable in Cox model

Hello,

I am using a Cox model in order to measure the determinants of the adoption of a new technology. In order to do so I have several variables X2 X3 X4 ... that describe the firm in itself.

proc phreg data=draft2;
model first*censor(0) = X2 X3 X4 ... /ties=exact;
run;

The problem I am facing is the following: in order to fully use this technology, an organizational change within the firm might be necessary. I keep track of the organizational change in a variable firstpro (year the firm changed its organization MINUS year the first first firm changed its organization). If I incorporate the variable firstpro directly in my original model I should face an endogeneity problem as in many cases the adoption of the new technology and the organizational change occurred at the same time (eg some firms adopted the new tech because they were well organized for it, others changed in order to adopt the new technology).

In order to avoid this issue, I would like to use my variable firstpro as an instrument:

proc phreg data=draft2;
baseline out=pred covariates=draft2 cmf=_all_ cumhaz=_all_;id id;
model firstpro*censor(0) = X2 X3 X4 ... /ties=exact;
run;

proc phreg data=draft2;
model first*censor(0) = predicted(firstpro) X2 X3 X4 ... /ties=exact;
run;

Do you think this could be a good idea ? Could this solve my problem ?

Thank you very much for your time and help,