science

Some takeaways from the study:

One goal of the study was to separate external vs internal causes, that is things like car accidents and violence vs heart attacks and diabetes.

The gap appeared not just in those who identified long-term with either side but with those who flipped from one side to the other. Those who went right to left and left to right had better or worse outcomes respectively. The other factors that correlated with the gap the most were income, insurance, education then religiosity and marriage. Race, ethnicity and gender were the least correlated. Those who identified as moderates at some point correlated the least. Marriage and religiosity tend to correlate with longer lives (religion due to social support).

The negative coefficients in the first years of all-cause deaths show that liberals died at rates higher than or similar to conservatives in the early 2000s, although the coefficients do not statistically differ from zero. However, by 2020–2022, more conservative respondents (as measured in waves 3 and 5) were significantly more likely to die than more liberal ones.

Although ideology became more predictive of health during the 2010s, the relationship between politics and mortality became consistently significant only after 2020...A strong relationship appears: participants who identify as very conservative were about 1.144 percentage points...more likely to die in 2020–2022 than those who were very liberal. Unlike in the biomarker analyses, moderate, conservative and very conservative respondents showed worse health - not only the most conservative.

...even conservatives and liberals living in the same county have different health outcomes. These data cast doubt on entirely policy-driven explanations for differences in health outcomes.

Existing research shows that while confidence in medicine was once non-partisan, Republicans began to show less confidence than Democrats during the pandemic...

While both groups (left and right) were just as likely to have visited their PCP right aligned were less likely to trust or follow their recommendations. Right aligned were more likely to go to the ER but again, less likely to trust or follow their recommendations.

Taken together, these results suggest that right-leaning Americans - as measured by their vote choice, partisanship and to a lesser extent their ideology - are less willing to visit, trust and adhere to the advice of their PCPs. This is true even among people with chronic conditions that require regular contact with the medical system for management and appears across a variety of outcomes unrelated to care for COVID-19.