COVID, Mental Health and Suicide in Youth: What does it mean? (part I)
There is much currently being written in the popular press about the catastrophic impact that the COVID19 pandemic is having on mental health especially the mental health of young people. The argument has been made about mental health in general and specifically centering on suicide risk. This is regularly being used as an argument to open schools to in-person learning as soon as possible-and it is worth noting that the CDC has just released detailed guidance for how schools might be able to open safely -given the right parameters and resources. But what do we actually know about mental health impacts of the COVID pandemic? Are distress and mental illness the same things? Have suicide rates changed?
COVID and mental health
There have been several self-report surveys of mental health during COVID which have reported (in some cases) significant upsurges in depression and anxiety “symptoms” since the pandemic has begun. Frequently when covered by the press and media (and often within the reports themselves) these are portrayed as “silent pandemics of mental illness” especially in young people. Are feelings of anxiety, sadness, anger, and loss during COVID19 actually mental illness or something else? Does it matter?
In an essay and accompanying video that I contributed to Psychiatric Times during the early days of the pandemic I proposed an analogy and an approach to this issue which is, I think, a helpful way of thinking about this. Consider someone walking a long distance carrying heavy bags on a hot day. This person will experience an increased heart rate and body temperature and maybe rapid breathing and, their (I am using the current plural/non-gendered convention) muscles may start to ache. If it is warm enough and they continue, they may begin to feel lightheaded and faint-especially if they don’t have access to fluids along the way.
Is this bag carrier-who is manifesting quite a few serious physical findings-ill? Most of us would not suggest that the pathological physical findings reflect an illness-rather, this is the normal reaction of their body in adapting to a physical STRESS. Whether or not this becomes a problem or potentially dangerous situation will depend on many external and internal factors. These would include internal issues like age, physical condition, underlying pre-existing illnesses or vulnerabilities and even potentially attitudes about or emotional context of the task (consider how different it would feel to go on a leisurely jog as opposed to running while being chased by a bear); along with many external factors-how hot it is, access to water, distance, terrain, weight of the packages among many others. Of course, our bag carrier may suffer a serious medical complication from this challenge-a heart attack, injured joints or muscles, heat stroke, among others. But the point is that the physical reactions to the stress-things we otherwise see as signs of illness-are not signs of illness in this context-recognizing context is essential! Finally, it is also probably true that our bag carrier would be quite happy with someone offering a ride to the destination or offering to help carry the bags.
I would like to suggest that our emotional response to the pandemic is natural and completely to be expected and we make an important conceptual error in using language that pathologizes it. We are facing a long trudge indeed and are carrying a lot of baggage. It is completely natural to feel sad, anxious, angry, and frustrated when we are facing a situation that is dangerous and unpredictable, is having massive impact on our work and social lives and may kill us or our relatives. We would never imagine telling someone who has recently lost a close relative or friend not to be sad, or someone facing a momentous exam or challenge not to be anxious. Like our bag carrying hiker, whether or not our feelings and reactions become dangerous or pathological has lots to do with internal and external context and conditions. Internal circumstances would include things like age and prior psychological, developmental, physical and mental health history and many external factors- physical illness, loss of a loved one to COVID19, living circumstances, loss of job, economic downturn, housing or food insecurity.
The context of consequences and experiences will bear on the psychological impact and question of whether a treatable condition might emerge. As with our bag carrier, even if a mental illness or treatable condition might not emerge, it would in all likelihood still be really great to have people help carry the load-providing emotional support. This pandemic is made all the more challenging since the length of the road and the terrain ahead are uncertain.
Do we have data?
Given this (unfortunately long-sorry) introduction, what are we to make of the many reports about deteriorated youth mental health during the pandemic? Do we have reliable data about the emotional impact of the pandemic? First, let me suggest that whenever you read a report about “deteriorating mental health” during COVID, it is important to go back and examine how the survey was done. Many times, surveys will ask whether “COVID has negatively impacted your mental health?” In fact, the question means to be asking whether the pandemic has impacted your emotional state-treating emotions of sadness, anxiety, or anger as synonymous with mental illness or ill health. The structure of the survey question is forcing the respondent into this “off the mark” response (it is like asking our bag carrier whether they feel sick-they might not feel great during the hike but not really “ill” and would not know how to respond accurately).
Let’s look at a recent data set of college students from the Healthy Minds Study. This is a large, yearly multi-campus survey of college student (not just students seen at counseling services but college students in general) mental health. The survey includes both several screening tools for depression, anxiety, eating disorders and inquires about past year suicidal ideation, non-suicidal self-injury, medication and drug use, among other things.
In exploring the data from fall 2020 as compared to the data for earlier years it is striking how similar the findings are. In fact, in many cases, the differences were larger between 2016 to 2018 than they were between 2018 and 2020. For example, positive screens for major depression in the PHQ9 (a primary care screening tool for depression, were 14% in 2016, 18% in 2018 and 21% in 2020. Again, “any depression” was 31% in 2016, 36% in 2018, 39% in 2020. Non suicidal self-injury went down slightly between 2018 (24%) and 2020 (23%). Suicidal ideation in this group also dropped slightly during COVID: 11% in 2016, 14% in 2018 and 13% in 2020. These reports are concerning since the positive rates are all quite high, but do not support the claim that there is a mental illness pandemic driven by COVID-at least not among college students.
What does this mean? First, it is not obvious that mental health problems are necessarily substantially nor dramatically worse during COVID. In fact, as I’ve repeatedly said, context is vitally important. Young people currently attending college during the pandemic might be, on average, a relatively well supported segment of the population. While many college students do struggle with finances, housing, and food insecurity, it is likely many of those students are not completing surveys. The point is that any survey and report must be scrutinized thoroughly and understood in its particular context. Common wisdom is sometimes wrong or at the very least incomplete or misleading.
Along similar lines, Dr. Paul Nestadt and colleagues examined suicide rates in Maryland during the first months of the pandemic (national suicide rates reporting typically lags by two years) and found that suicides in general had gone down as compared to previous years but were dramatically higher in Black men-known to have been disproportionately impacted both in terms of health and employment/economics-by the pandemic. Again, the context in which people experienced the pandemic had dramatic impact that would not have been recognized had they not segmented the data.
To be clear, I am not suggesting that the pandemic is not stressful, difficult, and challenging, simply that we should not be automatically calling those reactions or feelings “mental illness”.
Why does this matter?
Language and messaging matter. When we call our natural reactions to stress “mental illness” we are suggesting that the reactions: 1) might be pathological, unhealthy, or inappropriate and 2) since we tend to (mistakenly) believe that most mental illnesses are chronic conditions, we might mislead people into thinking they are experiencing an illness which will become chronic-ironically adding to their stress and worry.
When we misread the data and/or accept the common wisdom as scientific truth, we also run the risk of making bad policy decisions on this basis. Take for example the recent NY Times article which tied a series of student suicides to schools being closed. While any suicide is a tragedy, we don’t actually know that these suicides were actually precipitated or to what extent they have been influenced by school closures to in-person learning. There have certainly been clusters of school suicides in towns and cities before the pandemic. So, while there are many good reasons to try to get schools open for in-person learning as soon as this can be safely done, we should not let anecdotes-as tragic and powerful that may be-based on incomplete assumptions or analyses, drive our policy making.
What do we do?
The pandemic has certainly been very stressful. People and families have fallen ill, died in unimaginably large numbers, jobs have been lost, people have had long periods of isolation. The nation is rightly in mourning. For some, the impacts have been worse than others owing to both internal factors like preexisting poor physical or mental health, old age, and external factors-isolation, economic and employment status, living conditions. We need to bolster supports for everyone in as many ways as possible. We need to remain vigilant for severe or extremely disturbing emotional states and remain on the lookout for those who might have past history or emerging problems and connect them to or bolster care. If someone is in mourning and sleeping very poorly, we might prescribe a short-term sleep aid to provide some symptomatic relief. We should treat extreme symptomatic distress in this pandemic setting as well.
Finally, consider how most every religious tradition responds to loss and mourning. Families and friends come together and share social space and rituals-most often around meals-our most basic way of showing support, care, and love. The pandemic has clearly shown us that we need a national healthcare and social safety net system that strategically provides care and support for those most in need and most vulnerable.
In part 2 of this post, soon to follow, I will discuss how we might understand the consistently, alarmingly high (and possibly increasing) rates of suicidal thoughts and attempts in young people and what impact COVID may be having on these phenomena.