When typing out a lengthy comment on the Internet —in the land where attention spans nanometers—it wouldn't be decorous to not preface this comment with a caveat lector warning of its verbosity. So there.
I think it's much more likely that people who are already depressed will turn to drugs.
Indeed. Both critical thought and scientific concensus seem to corroborate the notion that drug dependency is a sequela (a result) of psychiatric or neurologic disorders, and not the obverse.
I surmise—by approximating the percent of misinformed people with whom I've discussed this topic and therefrom extrapolating that percentage to the broader population—that a majority of people would aver that some number—depending on whom one asks, it may be a few, some, a lot, most, or all—of psychiatric disorders are a corollary of drug dependency or, if you may, substance use disorders.
It's a curious thing that some people can experiment with a drug and not go on to develop cravings or dependence, but some others need only try that same drug once to get hooked.
My own experiences with psychotropic drugs can serve as a great example of this phenomenon. I have used cocaine on several occasions and sometimes more than once or twice in a single day. In spite of this, I am completely free of a cocaine addiction or incessant cravings. If I happen to be offered a line of the stuff while at some gala or clambake, I may accede if the ambiance is right or the coke is pure. The drug is fun, vivifying, has at times made a boring social event enjoyable, and has the attractive potential of magically transforming me from a gauche, reticent, shy, and tactless lummox to a loquacious, garrulous, assertive, and convivial social cynosure.
But unlike some unfortunate others, cocaine has not robbed me of my life savings and lowered my integrity to the point of indiscriminately trading sexual favors to procure another rail.
But then there is alcohol (and a myriad of other pharmacologically-similar drugs, like barbiturates and benzos). While a staggering proportion of Americans and other denizens of the Western World seem to quaff down inebriating libations near daily, I cannot even have a container of liquor within eyesight without inevitably imbibing the whole bottle and transmogrifying into a debauched, sexually aggressive, immoderately petulant, and lawbreaking troglodyte. Indeed, while others drink with temperance, I am a definite dipsomaniac.
Why is there such an extreme inconsistency in a drug's addiction potential? Why does its addictive effects seem so mercurial and different between one individual and another?
I believe the most plausible hypothesis is that an individual's personality and psychopathology are the primary determiner of who is and who is not at risk of developing a dependence to a given drug. That is to say, the type of drug user is of equal or greater influence than the type of drug used, insofar as the probability of addiction is concerned.
It's not pharmacology nor psychiatry in isolation that determines a drug's addictive potential and is therefore considerable; rather, it's their interface—the synergy between the drug's pharmacology (its set of pharmacodynamic properties in combination with the body's set of innate pharmacokinetic responses to its introduction) and the psychopathology of the person it's administered in that is the deciding factor in which lives the drug results in the ruination of or is essentially innocuous.
Earlier in this comment, I mentioned my weakness for alcohol. I reckon my proclivity to excessively drink is a result of alcohol being able to assuage the symptoms of my anxiety disorder and social phobia. In essence, the degree to which I find alcohol addictive is roughly commensurate with the degree to which I find alcohol medicating.
This is not applicable to just myself, but to everyone. People who suffer from certain psychiatric illnesses—especially schizophrenia, bipolar disorder, anxiety disorders like PTSD and social anxiety, and most personality disorders—have a much greater probability of abusing drugs (a form of self-medicating) than people without disabling mental illnesses.
I'm not sufficiently motivated to cull the Internet for apposite references to substantiate my argument. I do invite the incredulous reader to proffer references of their own that may invalid my claims. But I digress.
In conclusion, drug use, in the mentally ill, may in fact deter suicide. However, when the drugs that, say, a schizophrenic has been abusing to distract himself from his anguish lose their efficacy—which will happen with most things done in excessive amounts over a long interval of time—suicidal ideation may begin to replace his drug cravings.