In Canada, the panels are the same for all grounds for drug testing 1) because the sole objective is to mitigate risk for injury/death due to drug use by those in safety sensitive roles, & 2) to underhandedly screen out candidates who may have an underlying medical condition is illegal in this country.
My employer does perform job fitness testing to ensure candidates can physically perform the fundamental requirements of the position (some roles in our industrial setting are very physically demanding: heavy duty mechanic and other heavy trades, fire fighter, etc) but that is only done after a conditional offer of employment is made. By law we cannot ask about medical conditions in an interview, as we cannot sneakily test for any in order to screen out candidates who may be at high risk for sick leave.
Does all of that stop some candidates from still being suspicious? No, unfortunately.
Sorry; I didn't mean to imply that's a real suspicion, but I am speculating that it's a natural fear, dispelled by the truth that LauraLu shared above. Are pre-employment screens the same panel of tests as are used in for-cause inquiries?
Hence the question, some wonder, about what shows in the pre-employment tests, and if an employer might screen out those candidates whose pharmaceutical profile suggests a future involving higher-than-normal use of sick leave, excess medical claims costs, etc.
....."drug test their EMPLOYEES", not employers. You get the drift. Oh, and in addition to the above grounds, we too also do pre-employment drug testing.
My employer (before LTD) does reasonable cause, post-incident and random testing for safety-sensitive positions. The substances we test for are very specific, and are only ones which have the potential to cause impairment (eg: cannabinoids, cocaine, opiates, etc) thus increasing the probability of the employee to cause an incident resulting in injury or death. In Canada, there is no other legally justifiable reason for employers to drug test their employers. It remains somewhat controversial and thus the parameters around testing are very specific.
Thanks for following up on this important topic (and one I'd never even considered!). Glad there was a nice black and white answer. We can put this one to bed.