What would be significant is if the proportion of soldier suicides is greater than that of the similar group of age and sex in the general population.
That suggests to me that the structure of the military would generally have significant effects of reducing depression, or maybe just the type of person joining the military is/was less likely to be depressed, but now it seems reasonable that maybe the nature of our military services today causes potentially more stable men and women to be more likely than ordinary to commit suicide.
http://www.nih.gov/news/health/mar2014/nimh-03.htm
"Although historically, the suicide death rates in the U.S. Army have been below the civilian rate, the suicide rate in the U.S. Army began climbing in the early 2000s, and by 2008, it exceeded the demographically matched civilian rate (20.2 suicide deaths per 100,000 vs. 19.2). Concerns about this increase led to a partnership between the Army and the NIMH to identify risks."
Also need to follow the gravy train. Once someone starts getting a paycheck for a problem, that problem is never going away and on paper will always expand. Seventy years ago an infantryman who dived on a live grenade to save his whole squad of buddies was simply a dead hero. Today, if the dude is a white upper middle class straight guy from a stable family with no addiction problems and a decent credit score and stable romantic relationships, then, aside from being almost a unicorn, he would still be a (dead) hero in that grenade situation, but any demographic discrepancy could indicate to someone with a counseling / research paycheck depending on finding a problem, that racism in the .mil might have driven that poor downtrodden soldier to suicide, etc. This is probably a very small effect, but if you don't follow the money you're not doing due diligence.
Suicide is an extremely complex issue. I don't think it's fair to reason about the causes of suicide. Hearing from suicide survivors would likely permit bias to be reduced. These are not simple issues. Cause and effect are woven together with thousands of tiny little knots, when complexities like these develop.