America’s Mental Health

COVID has been tough for all of us. It’s been 171 days since the White House declared coronavirus a National Emergency, and life hasn’t been the same since. As with any major life change, there are bound to be mental health effects for a lot of us. The CDC even has an entire page on taking care of your mental health during the pandemic. In fact, the pandemic has indeed coincided with a mental health crisis,and mental health is arguably in the public eye more than ever. 

In the spirit of this, I thought I would do an article on mental health in America. I use a pre-pandemic survey of roughly 500,000 Americans called the Behavioral Risk Factor Surveillance System (BRFSS) collected in 2018-19. Importantly, I don’t measure mental health by rate of psychiatric disorders (depression, anxiety, etc.) in a population. Rather, I measure well-being based on the answer to the survey question: 
 Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?”
This is self-reported, so it definitely has flaws (namely, measurement error). But I think it has a number of advantages over the alternative. For instance, it includes the uninsured who cannot go to a doctor for an official clinical diagnosis and those who come from cultural backgrounds that preclude labels like “depressed”. It’s also a sliding scale that measures intensity, while the other metric is binary (depressed or not). I divide this variable by 30 and call it “probability of having a mental health day.” Think of it as your chance of having a “bad day.” Now let’s have some fun. 
Here are links to the various sections of the blog post:
Let’s start with the obvious. Does the age-old adage “money doesn’t buy happiness” really apply? The survey-takers divided the response to the income question into less-than-ideal buckets, so I had to work with what I got. See below: 

I don’t know about you guys, but I wasn’t exactly surprised. We see a perfectly positive relationship between money and happiness up to $75,000/year. In simple terms, each step is large and statistically significant. Survey participants making $10,000-$15,000/year were 14 percentage points more likely to have depressed or stressed days than participants making $50,000-$75,000/year. To contextualize that, assuming an adult lifespan (foregoing child years) of 55 years, that’s an additional 6 years worth of “bad days” over one’s life.


Race and Income

Of course, not everyone values the almighty dollar equally. Different people, cultures, and countries place different emphasis on economic well-being as it relates to mental health. To that end, I sought to find out if the relationship between income and mental health differed by race in the United States. The results were pretty interesting: 

First, the obvious. The relationship between income and mental health is strong for whites. Each step is statistically significant and large. In fact, the relationship is stronger for whites than it is for the general population. For minorities, the picture is less clear. The chart for the black population most resembles the white population – each step indicates that as income rises, the probability of a mental health day decreases. Even then, 4 of the 7 steps are not statistically significant, indicating that we cannot statistically conclude that the difference between the two buckets isn’t merely due to chance. For Hispanics and American Indians, we note that for many steps, there is no change in mental health. This indicates a much weaker relationship between income and mental health for these races, although generally, we do see a relationship. 
The really interesting chart is the one for the Asian population. Here, we see a very weak relationship between income and mental health. Many times, higher income buckets have a higher probability of mental health days than lower income buckets. The only step that is statistically significant is the jump from $20,000-$25,000 to $25,000-$35,000.


Race and Income: Young Adults

The above applies to survey participants of all ages. I found even more interesting results when I looked at this relationship for young adults (aged 18-29). First, let’s observe the relationship for the entire population: 

The relationship between income and mental health is much weaker for young adults than the general population. When we delve into this relationship by race, we see why: 


As we can see, the only race for which there is a relationship between income and mental health in young adults is whites (and maybe to a lesser extent, black people). The other minorities, namely American Indians, Asians, and Hispanics, display no relationship between income and mental health in young adults. To me, this was remarkable. Personally, it seems counterintuitive that Hispanic young adults making more than $75,000 had more depressed or stressed days a month than those making $10,000-$15,000, just given the financial stresses associated with an income that low. 
In sum, these charts tell us three things: 
  1. Young adults are more depressed and stressed than the general population
  2. The relationship between income and mental health is weaker for young adults than it is for the general population. 
  3. For both young adults and the general population, the relationship between income and mental health is largely driven by whites, who have the strongest relationship between the two and account for the greatest proportion of the sample. 


Race and Mental Health

Taking income out of the equation, are some races “happier” than others? There is evidence that minorities are more stressed out than whites. Let’s take a look at the relationship between race and mental health: 


It looks like the aforementioned generally holds true — minorities have a higher probability of having a mental health day than whites. That being said, we’ve seen the strong link between income and mental health. And we know that race and income are highly linked. So really, we might not be capturing that American Indians are inherently more unhappy than whites. Rather, what’s more likely is that they are poorer than whites, causing them to be more depressed and stressed. 
We can solve this problem by controlling for income. A simple regression allows us to do this. For the purposes of this particular analysis, we assume that whites are the “baseline” from which we compare the other races. The below chart summarizes our results: 


Here, we see that some of our results flip. Although American Indians still have a higher probability of mental health days than the other races, it is not nearly as drastic once we control for income. Even more interestingly, black people and Hispanics actually have arguably better mental health than whites once income is controlled for. In sum, we’ve learned: 
  • It looks like the phenomenon where minorities are more depressed and stressed than whites seems to be primarily driven by economic factors. In other words, the evidence points towards the “economic” part of socioeconomic status being more important towards explaining this phenomenon than the “social” part. 


I’ll stop here to caveat my conclusion. It’s evident that socioeconomic status is a large part of why minorities are more depressed and stressed than whites. Socioeconomic status is a complicated topic that means a lot more than income. In simple terms, your socioeconomic status is your “place” in society — it’s your education, your job, it’s how the police treat you, it’s how the president talks about your people, it’s how your neighbors treat you when you move into your new house. All of these factors are important in determining mental health, and I don’t want to diminish that. That being said, this particular evidence does point towards income being the dominant one within these factors in determining why minorities are more depressed and stressed than their white counterparts. 



Another potentially interesting topic is whether happiness is distributed evenly across America. Are the sunny beaches of California happier than the cold streets of Boston? I don’t know, I liked living in Boston, and I believe the Celtics just swept the 76ers in the first round of the playoffs (I think that makes 5 series in a row?), so there’s a lot to be happy about. For those interested in a state-by-state breakdown, see below: 

In my opinion, more interesting than looking at individual states is looking at regional differences. The U.S. Census Bureau conveniently splits the country into four regions: West, Midwest, Northeast, and South. Here’s a map to help you visualize: 

Below is a breakdown of results by each region as defined by the Census Bureau: 


Each pairwise-difference is statistically significant, except for the West-Northeast pairing. At the extremes, it appears survey participants from the Midwest enjoyed many fewer mental health days than their counterparts in the South. These differences might not seem large, but multiplied over entire regional populations, we get large amounts of social loss. Consider that 1 bad day might not seem like a big deal, but everyone in the country having a bad day is 300 million bad days. A heatmap helps visualize which states are driving these differences: 

We can walk away from this regional analysis with two conclusions: 

  • The Midwest fares better than the other regions of the United States in the mental health metric that I’ve constructed. This is primarily driven by the Upper Midwest (the Dakotas, Minnesota, Nebraska, Iowa). In fact, the Upper Midwest is the subregion with the lowest incidence of depressed or stressed days. 
  • The South suffers from a much higher incidence of depression and stress than the rest of the country. This is driven by what I’ll call the “Inner South” (i.e. the non-East Coastal South) — West Virginia, Kentucky, Tennessee, Louisiana, Mississippi, and Alabama.


As a parting thought, let’s look at the stark difference between the Upper Midwest and the Inner South: 

It’s evident that there are large regional differences, and areas of the South are faring quite poorly when compared to areas of the Midwest. 


Taking Care of Yourself

Let’s take a pause. We live in uncertain times. There’s a lot to be stressed about out there. It’s certainly been one of the most tumultuous years of my life. It’s especially important in times like these to check in on yourself. When was the last time you mindfully took a day or a couple hours for yourself? If its been a while, maybe it’s time to dust off that book you’ve been meaning to read, or buy a bottle of good wine and rent a good movie on Amazon.

Sometimes a day off isn’t enough. No shame in that. I’ve been there, and so have millions of others. Luckily there are (free) resources for you to take advantage of if you feel like you need to: 
  • The National Alliance on Mental Illness (NAMI) is a fantastic organization that has volunteers that offer support and answer questions you may have on next steps. They also hold local support groups (that are now virtual due to COVID). 
  • If you feel like your substance use has risen beyond a level you’re comfortable with since the quarantine, SMART recovery is a great evidence-based organization that utilizes cognitive behavioral therapy (CBT) to reduce or eliminate substance use. All meetings are virtual, and you don’t have to talk if you don’t want to. 
In truth, I’ve been struggling with my mental health myself. I’ve been watching a lot of Avatar: The Last Airbender in my time between jobs (read: I’m unemployed). One of my favorite characters is Uncle Iroh, the “wise old man” of the show. He’s got a lot of nuggets of wisdom, but one that I found the most inspiring is one that I think could speak to others who might be similarly struggling:
“In the darkest times, hope is something you give yourself. 
That is the meaning of inner strength.”
And now, some upbeat Chance the Rapper to play us out:


* indicates required

Leave a Reply

Your email address will not be published. Required fields are marked *