P1: Industrial accidents are more common when some of the people in safety-sensitive jobs have drinking problems than when none do.
P2: Since, even after treatment, people who have had drinking problems are somewhat more likely than other people to have drinking problems in the future,
Conclusion: any employer trying to reduce the risk of accidents should bar anyone who has ever been treating for a drinking problem from holding a safety=sensitive job.
The argument goes as follows:
- accidents are more frequent when people in the safety business have drinking problems
- people who've had drinking problems stand a greater chance of have drinking problems in the future
- in order to reduce the risk of accidents, employers must not hire anyone who has been treated for drinking issues
B is definitely out because it does not touch on the issue of "people with drinking problems". The argument specifically discusses reducing risk of accidents as a result of having fewer employees with a history of alcohol abuse. In this respect, B is somewhat like A: it's an issue that's parallel to the argument itself.
OK so C presents an issue with such an unforgiving policy, i.e. not hiring anyone who's had treatment: this creates an incentive for people with issues to lie. If such a policy were not in place, then people might seek treatment and therefore fewer people with drinking problems will be on the job. However, if you make it clear that anyone who's been treated before is not welcome, such a drastic measure might discourage some from seeking help, which means that they'll do anything in their power to avoid being detected. In other words, such a policy (of not hiring someone who's received help in the past) might backfire because people now have an added incentive to lie about their condition. So C is a good option (not a perfect one).
Now D is one of my favorite options because it presents a different cause for something. It's the jobs themselves that are making people more likely to have drinking problems. So the idea you could derive from this option is that no matter what you do, you won't be able to eliminate the risk of accidents completely, since the job in itself is stressful and causes people to drink. However, you'd be missing a very subtle note here: the phrase that makes this option less convincing is "exacerbate problems that they may have". If you don't have drinking problems to begin with, then you won't necessarily develop them later. If you already do, then it get even worse and the potential for issues later on greatly increases. D actually supports the passage.
The correct answer is definitely choice C, as many have pointed out.
Choice C tells us that current workers may actually have a (current) drinking problem. So, by replacing them with those who have been treated for their drinking problem, we may actually decrease the risk of accidents, thereby weakening the argument.
That some companies' policy is to put drinkers in residential treatment is clealy irrelelvant. Thus, choice A is incorrect.
Because the argument was about reducing the risk of accident through a certain policy, that some accidents are attributable to an alternative explanation is irrelevant. Thus, choice B is incorrect.
We don't care that safety-sensiive job holders are more likely to become drinkers. We care about whether the policy against hiring them will lead to a reduction in the risk of accident. Thus, choice D is incorrect.
Choice E is wrong for the same reason that choice B is wrong: that some accidents are attributable to an alternative explanation is irrelevant.