Actually, suicide stats indicate that "I do not intend to kill myself" is not how it works. In general, suicidal episodes seem to be short and having access to means to kill yourself increases the chance you'll actually do that. Not having the means available increases the survival rate - even small barriers like selling painkillers in smaller boxes show significant reduction in suicide rates. Another considerable factor are accidents - nobody intends to kill himself in a gun accident either, it just happens. So I think it's quite correct to include those numbers in the risk profile.