Suicide and pandemic more complicated than it seems

In late March, just as the crest of the first wave of the coronavirus approached, President Donald Trump was pushing to reopen the economy. There would be “suicide by the thousands” within weeks if businesses stayed closed and people were out of work, he argued.

The notion that economic shutdowns themselves could pose a greater threat to the health and safety of communities than the uncontrolled spread of a disease that has now killed more than 1 million people worldwide was and remains an important idea to consider.

But asserting a fact doesn’t make it true. Was Trump right? Did suicides skyrocket in March, April and May? Did a spike in suicide deaths outpace COVID-19 deaths?

As an emergency room physician, I kept an eye out during my shifts in the weeks following Trump’s March 24 statement. It seemed to me that we had fewer suicidal patients than usual. I called a colleague across town at another hospital. He thought he might be observing the opposite in his ER, that there might have been an uptick in patients with suicidal thoughts or attempts.

Along with a team of researchers, I set out to try to find out what was happening. But we would have to wait. Death by suicide takes longer to be reported and finalized than most other causes of death. Every suicide death is investigated and its final cause directly adjudicated by a medical examiner, making the process slower but ultimately more reliable.

It turns out that both I and my crosstown colleague were mistaken. Suicide rates in Massachusetts neither rose nor fell last spring. Suicide rates did not change from expected rates at all. Just to be sure, we performed what researchers call a sensitivity analysis — a fancy way of saying we asked the same question in a number of ways to make sure we were not deluding ourselves. We compared this year’s rates in March, April and May with those from last year and other years.

Whether we considered the months individually or combined, year by year, there was just no change. We ran the numbers again, this time assuming that each of the few dozen deaths in 2020 that occurred during March through May in which the cause of death was yet to be determined were in fact suicides. The scenario was unlikely, but one we had to consider. No matter how we looked, we kept finding the same thing. Suicide rates did not budge during the stay-at-home advisory period (March 23 until a phased reopening began in late May) in Massachusetts, which had one of the longest such periods of any state in the nation.

Studying the effects of stay-at-home advisories is still in its infancy, and what is learned will help inform the decisions of public health officials as they consider measures to address future infectious-disease outbreaks or another COVID-19 spike. Some worries about stay-at-home periods will turn out to be overblown, others not. Concerns about the limitations of remote learning for children increasingly appear to be justified, for instance, and concerted efforts to address the problem are urgently needed.

Many well-informed and well-meaning people fell for the cognitive trap that if something rings true, it must be true — and thus assumed that suicide deaths were destined to rise during shutdowns. Certainly, more study on this subject is needed. The Centers for Disease Control and Prevention recently reported, “During late June, 40 percent of U.S. adults reported struggling with mental health or drug use,” with 1 in 4 people between the ages of 18 and 24 saying they had “seriously considered suicide” in the previous 30 days.

There are legitimate questions to be raised about the pandemic’s toll on mental health. Some of the impact may have more to do with the continuing inability to control the virus, and with the ensuing economic fallout, than with Americans’ staying home for weeks and even months in the spring. That said, a rise in suicides or other suffering resulting from temporary stay-home advisories is neither guaranteed nor inevitable.

To get this right, both now and in the future, we have to keep asking the right questions and awaiting the actual answers — and remember that the questions themselves, no matter how obvious their implications might seem, do not provide the answers. They remain what they are: questions.