Stephen - To address your questions: (1) We did not look at other methods individually; however, you'll note that our focus on % of suicide deaths resulting from firearms mitigates concerns on that front. Specifically, in our tests of indirect effects, the numbers demonstrate the impact of the laws on the overall suicide rate was indirect through the proportion of suicide deaths resulting from firearms. This means that overall rates across methods are impacted and that this impact largely stems from reducing percentage of those deaths in which firearms were the chosen method. This does not mean other methods are meaningless in these relationships, but it points towards a clear outcome and mechanism. (2) Yes, our numbers indicate that the method does really matter. The notion that "those who want to kill themselves will do so" is simply not supported by the data. Method substitution is an argument that has been put forth to push against restriction of access to firearms, but the data simply don't support it. Our longitudinal data in particular back this up. (3) Suicide by cop is not included in the CDC suicide numbers as far as I know. It is unclear, however, how this would change our results. It's an interesting question, but it's difficult to get official numbers on that particular outcome. One of the main reasons we chose the variables we did is because all the data are publicly available, thereby optimizing transparency and reducing confusion based upon data source. (4) My lab does a lot of work on death by suicide in the military, so I agree that it's an interesting question. Veterans are far more likely to die by suicide using firearms than are other populations, which would lead one to believe that these effects would actually be stronger for those folks. That being said, although their rates are higher, the proportion of suicide deaths that are veterans is not enormous, so the likelihood that military status is spuriously influencing our results seems rather small. Certainly an interesting question, but it's not clear how military status would better explain the results or minimize the importance of regulation (e.g., if results are more pronounced in military members, this would not diminish the urgency for regulation). In terms of rural states, we controlled for population density, which by definition accounts for much of that variability. In a follow-up paper, we added several more covariates (race, age, education) when examining handgun laws that we believe are more relevant (and the data support this). Given those results, which are embargoed by the journal that accepted the paper for the next several months, I am not concerned that our data in any way overlooked such important factors. On your point about shotguns, I agree that they are possible as a method. The data simply indicate they are much less frequently used, which is why research has repeatedly supported a role for handguns but not long guns in suicide rates. I appreciate the tone of this conversation, by the way. Although you and I disagree fairly strongly on regulation, the focus here has been on suicide and the data, which is incredibly helpful. Thank you.

Ah - I see. Yes, we did look only at overall rates. My rationale there is that showing a decrease in one particular method does not necessarily speak to a decrease overall. Although the evidence does not support the notion of method substitution, if we didn't look at overall rates, we would leave ourselves vulnerable to that counterpoint. Like you said, the assumption is that firearms suicides are a meaningful component of overall rates and the decrease in the latter is likely indicative of a decrease in the former, but we did not examine that directly. I can easily take a look at that though.

Steven - Unless I'm misunderstanding your question, we actually did include that information. Both states enacted laws in 2003. We published rates before and after that year. Given the low base rate of law changes, we weren't able to run the types of analyses I'd like to more clearly demonstrate the effects, which is why we urged caution there. That being said, given the evidence from the cross-sectional analyses (and in our forthcoming follow-up paper), these provide a reasonably useful alternate angle for considering the overall narrative. Happy to assist in getting you a full copy of the paper if you want to email me directly (michael.anestis@usm.edu)

I can certainly try. Having a hard time finding opportunities to write for the blog much these days!

Wes - At the time this article came out, we were posting cited article on our references page (link at the top of the page). If they're not in there though, you can use the info within this article to do a quick Google Scholar search. Best of luck to you. - Mike Anestis

First off, thanks for the response. It's been quite a while since I've read the article referenced in this PBB piece, but I'll do my best to respond point-by-point to your comment. (1) I'm okay with you characterizing at least part of my piece here as a rant. It's a blog and, at times, I rant a bit. I'm not sure that using words like "horrific" justify that characterization, but I'm fine with that either way. The opening of the piece raised what I believe to be an interesting question - the difficulty in balancing the first amendment (which I value a TON) and looking out for publications with the potential to cause harm. I offered a potential solution, which was to make these types of potentially controversial articles more of a dialogue either through much more prominent (and open) comment sections or through a system in which replies to original articles are published, thereby creating a public discussion capable of giving folks a chance to develop their own informed opinions. (2) I'm less on board with characterizing reflections on personal experiences as "facts," particularly within the context of an article in which the author is discussing directions for the future of health care. Paul Meehl would have a heart attack were he to read that. (3) Given my discussion of my value of the first amendment, it should be clear that I also respect people's right to their opinions and their perspectives and that I also think they should be able to talk about them. The question isn't whether opinions should be hidden, but whether all opinions should be given the same access to worldwide podiums. Perhaps they should....that was the question I raised. I highly suspect that folks I've worked with as a therapist are aware that I am highly interested in their experiences in treatment, both prior to and while working with me. (4) Nobody said that you need an advanced degree to talk about your own experiences. I do, however, think you need an advanced degree or immense experience with the scientific literature in a subject area if you are going to publish a NY Times piece discussing the relative value of different forms of treatment for mental illness and the path the field should take in the future. These are important topics and readers deserved informed opinions. I get that he was interested in the value of writing beyond the idea of psychotherapy, but his piece spoke to a point I believe to be important. (5) Yes...15 minutes was an exaggeration. Sarcasm, folks. (6) The author referred to psychotherapy only as "talk therapy" and made no differentiation between forms of treatment despite epic amounts of research on various forms of treatment and their varying levels of effectiveness. I see no way in which my conclusion there can be seen as reading into anything. If the author believes differently, he did a horrible job of conveying that point. (7) Other examples of the author characterizing "talk therapy" as I said he did: "A generation ago, when “Annie Hall” won the Oscar for Best Picture, talk therapy occupied a prominent place in our collective imagination, whether or not you partook. If you wanted to spend several hours a week baring your soul to a stranger who was professionally obligated to listen and react, you went into therapy. Today you join a writing workshop." "Like the patient on the analytic sofa, we fixate on particular stories and characters and themes because they speak to the fears and desires hidden within us. Our inventions inevitably take the form of veiled confessions." Can you cite any examples of him characterizing "talk therapy" in a different way? If not, that means his only descriptions of treatment align with my characterization. (8) I disagree with the notion the creative writing offers a suitable alternative to psychotherapy (at least evidence-based treatments). When I refer to facts, I refer to repeated falsification of hypotheses. As I have mentioned countless times on PBB, positive findings are not facts, but repeated non-significant findings represent a repudiation of a claim. Popper pointed this out LONG before this site existed. I merely echo his words. (9) If you think my definition of a "fact" is the conclusion of a research article, you either have not read this site. (10) Happy to read those articles. (11) Which treatments have been incorrectly labeled effective based on poor research practices? How great of a likelihood of readers being led away from effective treatment is acceptable to you? Is there a certain number of lives that can be damaged without the cost of a bad article being considered too high? (12)"I think you can do more to improve the quality of mental health treatment by joining the movement to improve the quality of research in this area than by railing at the NY Times. In other words, quality, like charity, begins at home." PBB is a hobby I do on the side. I'm not than content with my contribution to research and my CV is readily available if you would like to determine for yourself whether I'm doing the work to contribute beyond "railing at the NY Times" (does that phrase qualify your comment as a rant?). You use a pseudonym, so I'm not sure what contributions you're making beyond anonymous internet comments. (13) If you see flaws in studies I describe that you believe I have overlooked, by all means point them out. I'm always looking for the strengths and weaknesses of research and do my best to report them. My comment section is always open and I'm happy to hear disagreements with specific points. I might not agree with you and we might argue back and forth a bit, but that's the point of the site. If the implication is that I intentionally overlook flaws, that's a point we'll have to disagree on. - Mike Anestis