Mortality rates increase in U.S. counties that vote for losing presidential candidates

An analysis of U.S. statistical data found that, after elections in which political leadership changed, mortality rates increased in counties that supported the losing presidential candidate compared to counties that supported the winning candidate. The increase was on average 7 deaths per 100,000 people. The research was published in PLoS One.

Recent years have seen an increase in political polarization in the United States. Hostility towards political opponents has increased substantially. A 2016 study reported that both Democrats and Republicans claim that the other faction generates “very unfavorable” feelings, such as frustration, fear, and anger.

One consequence of this polarization is that communities are becoming increasingly homogeneous politically. Political minorities living within these communities are found to be apprehensive about disclosing their political preferences and are more likely to be subject to discrimination and social isolation.

Political views also affect expectations about future economic outcomes. Individuals tend to be more optimistic about the economy when they are affiliated with the party that controls the government. Studies indicate that this tendency has increased during the last decade.

Study author Sris Chatterjee and his colleagues wanted to investigate whether a shift in political leadership and the lack of political representation can affect health outcomes. They note that holding political opinions that differ from those of the ruling party can increase stress, anxiety, and the feeling of social isolation. They reason that this might have adverse effects on physical and mental health.

The study authors used county-level mortality rates as an indicator of health. They combined these mortality rates with data on voting in Presidential elections.

For each county, the researchers calculated the share of voters who voted for the Democratic candidate and the share who voted for the Republican candidate. If a county primarily voted for a candidate who did not win the election, it was considered to have experienced an electoral loss.

The authors focused on elections in which Presidents Obama and Trump (the first time) were elected. These were elections in which political leadership changed, meaning that the elected president was not from the same party as his predecessor. President Obama was elected after a Republican president (Bush), while President Trump was elected after a Democratic president (Obama).

The researchers considered the total number of membership associations per 10,000 people in a county as a measure of individuals’ social involvement in the community. They also used data from two surveys that asked about health to produce additional measures of county-level health.

Results showed that mortality rate trends in counties supporting the winning candidate and those supporting the losing candidates diverged after the elections that resulted in political change. More specifically, losing counties experienced, on average, an increase in the age-adjusted mortality rate of 7 deaths per 100,000 population.

Previous studies cited by the authors found that a one percentage point increase in unemployment results in a decrease of the mortality rate by 4.6 deaths per 100,000, while the mortality rate decreased by 18 individuals per 100,000 during the Great Recession (2007-2009).

Thus, in terms of magnitude, counties supporting the losing candidate experienced an increase in mortality roughly equivalent to the absolute change found when unemployment increases by 1.5 percentage points. The increase was a bit less than half the magnitude of the change that occurred during the Great Recession.

“Evidence suggests that political sentiments and social isolation may be important factors underlying our findings. Indeed, we document after turnover Presidential elections a decrease in the degree of social interactions in the losing counties. Also, we identify a further increase in sudden causes of death around crucial political events and a worsening of the mental health of the individuals,” the study authors concluded.

The study contributes to the scientific understanding of the links between political processes and health. However, the authors note that they were not able to estimate how much an increase in sudden deaths contributed to the observed increases in mortality rates and could not exclude the possibility that mechanisms other than social isolation and political sentiments might play a role in the observed increases in mortality.

The paper, “The health costs of losing political representation: Evidence from U.S. Presidential Elections,” was authored by Sris Chatterjee, Iftekhar Hasan, and Stefano Manfredonia.