The primary weakness of this paper is its complete reliance on two extremely small and poorly-designed studies. The first was performed on Reddit with n = 194 Redditors who self-reported how healthy they were on a 9 point scale, how liberal/conservative they felt on a 9 point scale, and answered a series of questions to establish a personal responsibility score (PRS). The second was performed with n = 204 local students, mostly teenagers, recruited based on political party affiliation, whose healthiness was established only by how often they claimed to take the stairs.

You should be able to identify at least 6 major design flaws in the studies above, but in short the researcher not only failed to prevent but seemingly employed predictable biases, especially in regards to his measures of health, which were entirely self-reported. It should go without saying but just because some group of people tend to consider themselves better than others in some respect does not actually make it so, especially when it comes to popular vanity metrics. Yet that is precisely what this paper claims.

As to contrary evidence: you typically won’t find a paper published in any serious journal whose thesis is so close to “ideology A is better.” Eschewing scholarly impartiality on politically charged topics is generally frowned upon. Doing so in exchange for publishing and/or favor with wealthy patrons has always been possible, and while increasingly prevalent in recent years, it is primarily the realm of conservative academics if only because it causes a greater stir (shares, citations, impact factor). Even using the loaded phrase “personal responsibility” in a political context, and equating it to the term used in health literature, marks this as a rather obvious insider piece not subject to the typical quality controls. So, it’s unlikely you will easily find an equivalently obnoxious antithesis like “conservatives are less fit,” “liberals have better dental hygiene,” or what have you. But does that mean conservatives are healthier?

No. We can confirm this a variety of ways, since exposure to any social science will routinely surround you with high quality evidence to the contrary, but here is where I would start:

1. With high statistical significance, positive health outcomes are strongly correlated with education.
2. Yet, education is inversely correlated with political conservatism in nearly every developed nation in the world with high magnitude and significance.

If I could live forever on the premise of I never, ever, stop talking, I would spend all my time nitpicking every conservative point that is skewed to make it look like their opinion is right. However, I have a life to live, so I cannot go over it point by point. All I have to say here is that when I was looking over the study and its sources when I felt like posting, I could not take the data presented in good faith because it seems overtly clear that it is a biased study.

So as much as I would like to, I would just have to point you to the other comments here regarding Elsevier's history of "scientific journals" and other peoples qualms about the study.