select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
A study published in JAMA Psychiatry looks at folic acid prescription fills, suicide attempts and intentional self-harm among privately Insured US adults.
Prof Keith Hawton, Director of the Centre for Suicide Research, University of Oxford, said:
“The findings of Gibbons and colleagues that individuals receiving folic acid had lower rates of self-harm and suicide attempts is an intriguing and potentially important one, especially given high rates of both suicide and self-harm being seen in many countries. The researchers have clearly controlled for some relevant confounders that might have influenced the results. However, as they acknowledge, whether this represents a finding with significant implications for prevention of suicidal behaviour can only really be demonstrated through a large-scale randomised trial, preferably in individuals potentially at risk. Until then the findings should be treated as of considerable interest but with uncertain implications for prevention of suicidal behaviour.”
Dr Simon White, Statistical Ambassador for the Royal Statistical Society, and Senior Research Associate at the University of Cambridge, within the Department of Psychiatry and Medical Research Council Biostatistics Unit, said:
“The authors use records from a large US medical database and find an associations between folic acid prescriptions and a reduction in suicide attempts or self-harm. The authors themselves acknowledge that these are only associations, there are potentials issues with who is in the database that may account for the association in other ways; their conclusion is a recommendation to perform a randomised control trial to validate this association. Hence this paper does not support any direct change in practice or clinical care – individuals should not change or stop their treatment/medications for folic acid.
“It is important to note that this study includes individuals who are prescribed folic acid, and does not account for over the counter (OTC) purchases. This may limit the applicability of the findings to the UK, where prescription practices may differ from the US. The evidence on the use of folic acid together with standard care for depression, suicidal ideation, and suicidality is mixed; this paper adds to the evidence but is not conclusive.
“For the general population, this paper does not support wide scale use of over the counter folic acid to reduce suicide attempts or self-harm; the paper does not address the impact of long-term prescriptions (limiting follow-up to 24 months). In an unrelated initiative, the UK Government plans to add folic acid to bread flour based on strong public health evidence that folic acid prevents neural tube defects during pregnancy (https://www.gov.uk/government/news/folic-acid-added-to-flour-to-prevent-spinal-conditions-in-babies). Which may further complicate translating these findings from the US to the UK.
“Overall the study is well done and clearly presented, in particular the limitations. Noting that 81% of the sample are female whereas in the UK the rate of death by suicide is 3 times higher in males than females; so there is a substantial mis-match between the US database and those who die by suicide (the paper is concerned with suicide events more broadly, not only death by suicide). The authors account for important confounders such as sex, age, previous history of suicide events, previous history of folic acid prescriptions; however the selection of who is in the database cannot be fully discounted, in particular there was no information on women planning for pregnancy in the database.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“This is a well-conducted study, involving a very large number of patients (nearly 870,000) in part of the US health care system. The use of statistical methods and analyses in the study is, in my view, appropriate for the task at hand. The design of the study is quite ingenious, in that it avoids several potential kinds of bias by treating each patient as their own control, and comparing months when an individual had a folic acid prescription with months when they didn’t have a prescription. It appears to provide pretty strong evidence that taking folic acid is associated with quite a large relative reduction in the risk of suicide attempts or deliberate self-harm. But the researchers aren’t recommending that folic acid be prescribed forthwith for people at higher risk of self-harm – they are instead calling for a randomised clinical trial.
“That’s because they understand well the limitations of a study like this, however large it is in terms of participants, and however well-conducted and well-analysed. Indeed I think it’s best to see this study in the context of a series of different pieces of research. This thread of research began with a large-scale screening exercise, a large and complicated piece of data analysis really, called iDEAS, which examined data on 922 different prescribed drugs to identify those which were associated with changes in the risk of suicide attempts. They found 10 drugs that were associated with increased risk of suicide attempts, and 44 associated with reduced risk. But a broad observational study like that can’t by any means tie down whether these associations were causal, that is, whether the drugs actually cause the increased or decreased risk. All it can really do is to give good leads on which drugs might possibly be worth examining further, in terms of how they might be related to the risk of suicide attempts.
“One of the strongest associations, found in iDEAS, between a taking a drug and reduced risk of suicide attempts was with folic acid, which is actually a version of a B vitamin (folate) that has several important functions in the body. It’s found in certain foods, but can also be prescribed for various reasons (including to help reduce side effects of some other drugs, that tend to reduce folate levels). It is also available over the counter (in the UK as well as the USA). This association between taking folic acid and the risk of suicide attempts had not been predicted by the researchers before it emerged from the iDEAS study, and that appears to have motivated the researchers to examine folic acid in this new study. (The researchers report in the Discussion section of their paper that some previous studies have found associations between either taking folic acid, or folate levels in the body, and depression and/or suicide.)
“The new study is based entirely on the health and prescription records of people in a very large database of insurance records in the USA. Because the researchers didn’t assign people to taking or not taking folic acid, it’s an observational study. In fact the participants were all prescribed folic acid at some point. The outcome that the researchers were interested in was what they call “suicidal events”, which include all suicide attempts or events of intentional self-harm, that involved outpatient or inpatient treatment. Rather than comparing the risk of suicidal events in a group of people prescribed folic acid and another group who weren’t prescribed it, the researchers instead directly compared the risk of suicidal events in months when a participant had a folic acid prescription with the risk in months when they didn’t have a folic acid prescription.
“On this basis, after the statistical adjustments that they made, the researchers estimated that there was a 44% reduction in the risk of suicidal events over a given period of time, in months when folic acid was prescribed, compared to the risk in months where folic acid was not prescribed. Of course, there’s some statistical uncertainty about that estimate – plausibly the reduction could be between 35% and 52%. These are pretty substantial reductions. They are relative measures of the reduction in risk, not absolute. It probably doesn’t make much sense to estimate absolute risk changes from this study, since everyone in it was prescribed folic acid, so to that extent the participants don’t give a good measure of the absolute risks in the whole population which would include many people never prescribed folic acid.
“This approach, of comparing each participant directly only with themselves at a different time, avoids one kind of bias that can occur in observational studies, that there are long-term systematic differences between those who take folic acid and those who don’t, so that those differences could be the cause of any difference in risks, rather than the folic acid itself. But that doesn’t get the study off the observational hook entirely. It could still be the case that there are other differences between the months in which the participants take folic acid, and the months when they don’t, or in how the participants behave or their state of health in those two different types of month. Those other differences that vary over time could still be the real cause, or part of the cause, of any differences in risk associated with being prescribed folic acid. In the jargon, there could still be some kind of confounding that hasn’t been dealt with. Thus it’s still true that we can’t be completely sure, from these results, that the reduced risk of suicidal events in the months when people were prescribed folic acid, compared to the months when they weren’t prescribed it was caused by the folic acid.
“The researchers did build in several aspects of the study design and analysis to establish as far as they could that the reason for the risk differences was the folic acid, rather than something else. They made statistical adjustments for a wide range of characteristics, some of which might be potential confounders, because they could be related to whether people were prescribed folic acid as well as to the risk of suicidal events. They did a further statistical analysis that indicated that, the more months the folic acid prescription went on for, the greater the reduction in risk. And, to check that any findings were not just an artefact of their particular approach in designing and analysing their study, they carried out another exactly similar study, but using the vitamin cyanocobalamin (a human-made version of vitamin B12) instead of folic acid. It is also available over the counter or by prescription, but it does not contain folic acid, and it was not found to have any association with the risk of suicide attempts in iDEAS. In this control study, the researchers found no evidence of an association between being prescribed cyanocobalamin and suicidal events.
“All of this does tend to make it more likely that the cause of the difference in risk was folic acid, at least to some extent. But we still can’t be certain. And there are other potential limitations, as the researchers make clear. For example, the rate of suicidal events in this study is considerably smaller than national US rates, which is probably due to very incomplete reporting (perhaps just because the records are incorrect, but also possibly because not all such events lead to formal medical treatment). Also, there’s no data on whether and when the participants might have taken folic acid that they bought over the counter, perhaps as part of a multivitamin tablet.
“Fundamentally, this remaining uncertainty about cause and effect is a major part of the reason why this study recommends a randomised clinical trial (RCT) that directly examines whether folic acid reduces the risk of suicidal events. In such a trial, people would be allocated at random to take either folic acid, or some control substance (perhaps a dummy placebo tablet, perhaps something like a different vitamin), or perhaps they would take folic acid or the control substance at different times, allocated by the researchers. So, on average there wouldn’t be a difference between those who took folic acid and those who didn’t (or between the times when people took folic acid or not), and the statistical analysis would be able to tell the researchers how far from average things might be. Therefore, if there was a big enough difference in the risk of suicidal events between people taking folic acid and people not taking it, really it would be very likely that the folic acid was the cause. The RCT could perhaps also look at the appropriate dosage, and at whether the effects are different in different subgroups of the population. All this would make it very much clearer whether folic acid really has a role in reducing suicide and self-harm, and if so, how big the effect really is. I hope the trial happens soon.”
‘Association Between Folic Acid Prescription Fills and Suicide Attempts and Intentional Self-harm Among Privately Insured US Adults’ by Robert D. Gibbons et al. was published in JAMA Psychiatry at 16:00 UK time on Wednesday 28 September 2022.
DOI: 10.1001/jamapsychiatry.2022.2990
Declared interests
Prof Keith Hawton: “I am a member of the National Suicide Prevention Strategy for England Advisory Group.”
Dr Simon White: “No conflicts.”
Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee. My quote above is in my capacity as an independent professional statistician.”