A closer look at geographic data finds no correlation between generally happy locales and rates of suicide, according to new research by CU Boulder and U of California Irvine

Several years ago, a research paper made headlines by finding that the happiest states in the United States also had the highest suicide rates.

This seemingly paradoxical conclusion caught the attention of a trio of researchers from the University of Colorado Boulder and University of California Irvine for a different reason: could there be more to the story than the eye-catching findings suggested?

By looking at counties instead of states and making a few other experimental tweaks, researchers put the newsy findings to the test. The results of their study, published last fall in the Journal of Happiness Studies, contradict the earlier study, adding another layer to a growing body of research about how our geographic location affects our happiness and well-being.

“Our basic finding was that we don’t see any relationship between the general level of happiness in a place and people’s likelihood of committing suicide,” said Tim Wadsworth, an associate sociology professor at CU Boulder and one of the study’s co-authors.

Designing the experiment

The initial headline-grabbing study, conducted by a team of economists and published in 2011, concluded that the happiest U.S. states also had the highest suicide rates. The economists’ explanation for this paradox was that people tended to compare themselves to those around them—if you’re an unhappy person in a happy place, your negative feelings might be exacerbated by your positive surroundings, which could lead to suicide.

Wadsworth, along with then-CU Boulder graduate student Philip Pendergast and UC Irvine criminology professor Charis Kubrin, were skeptical of the study’s focus on people living in the same state. 

Since states can span hundreds of thousands of square miles, there are major regional differences to consider—Boulder, for example, is very different from Colorado Springs. Positing that people were more likely to compare themselves to others in their immediate vicinity, the researchers narrowed in on data at the county level instead.

“We found it to be problematic, unless you’re going to make the argument that people in one part of California are comparing themselves to people in other parts of California,” Wadsworth said. “But I will buy the idea that I’m influenced by what I see living in Boulder among other Boulderites.”

They deviated from the initial study in other ways, too, such as by controlling for regional variations in suicide rates and well-being. The researchers also accounted for the availability of firearms and included a measure of overall health in their analysis, since those two factors are also correlated with suicide. In essence, they created a more rigorous experiment to see if they could reproduce the initial findings.

In the end, their study failed to find evidence of a relationship, either negative or positive, between suicide rates and the average subjective well-being, or happiness, of U.S. counties. Instead, the findings reinforced the traditional predictors of suicide, such as the accessibility of firearms and health and regional variations. 

Noting the limitations of their work and the low number of studies on this topic, the researchers also pointed out the need for future research.

“The question is then, ‘Well, who’s right?’” Kubrin said. “No study is ever the full explanation of anything. What awaits is additional studies to weigh in on this—that’s the way science works. More than anything, our paper raises questions about those initial findings.”

Understanding the findings

Practically speaking, the study’s findings suggest that you shouldn’t necessarily be concerned about living in a happy place if you’re unhappy, nor that you should rush out and move to a happy place in an attempt to become happier.

“The idea that anybody could move to a place to change their level of happiness, there’s no evidence that that’s actually happening,” Wadsworth said. “Let’s not all of a sudden jump on the bandwagon and say we need to be extra concerned about unhappy people in happy places, not to say that it’s impossible, but (this study was) really just sort of holding up our hands and saying, ‘Whoa.’”

Another analysis made headlines recently, too, this one from the U.S. Centers for Disease Control and Prevention. In May, the federal agency reported that suicide rates across the country increased 25 percent between 1999 and 2016.

While that statistic alone is alarming, sociologists like Wadsworth and Kubrin say it’s important for public health officials and policymakers to drill down into the data to look at state, regional and local trends, as well as other factors such as race, age, education level, marital status and the availability of firearms.

“One of the things we know about suicide, as well as homicide, is that it’s not equally distributed,” Kubrin said. “One of the things you have to do is disaggregate the suicide rate. The overall suicide rate is helpful, but not nearly as helpful as looking at different groups of people and the areas where the increase has occurred in particular.”

On the other side of the spectrum, when our friends, family members and even celebrities such as Kate Spade and Anthony Bourdain die by suicide, we tend to focus on their individual circumstances, such as their mental health. But if we really want to prevent more suicides across the country, it’s equally as important to look at widescale trends that go beyond individual factors such as depression, Kubrin said.

“Individual-level explanations are 100 percent absolutely important,” she said. “But if we just stay at the individual level, we might be missing bigger, broader factors that impact suicide above and beyond any one individual’s characteristics. There are other questions about suicide that need answering. Why do suicide rates cluster in certain regions of the U.S.? It’s not just a function of people with depression moving to certain areas. There’s something beyond the individual that plays into this distribution.”

Suicide research continues

Overall, the researchers say that further study on suicide is needed. Though suicide is much more common than homicide or violent crime, it receives a fraction of the attention.

“We’re obsessed with crime and crime trends and crime data,” Kubrin said. “The reality is that suicide, especially for certain groups, is much more likely to occur than homicide.”

For his part, Pendergast said he hopes to continue his contributions to suicide research in his new role as administrator of the U.S. Census Bureau’s Rocky Mountain Federal Statistical Research Data Center, which is housed on campus.

After completing his doctorate at CU in May, Pendergast now reviews proposals from people who want to use the center’s restricted population data in their research. With such a rich dataset, Pendergast said he hopes researchers will explore topics such as why the American West has a higher suicide rate than other parts of the country. More broadly, Pendergast said he hopes this and other studies help spark much-needed conversations about suicide.

“I hope that any papers about suicide would have the implication of causing people to engage in more of a cultural dialogue about suicide and making it less stigmatized by simply talking about it,” Pendergast said.