The Obama administration came to similar conclusions. In 2016, the Centers for Medicare and Medicaid Services revisited the question of whether sex reassignment surgery would have to be covered by Medicare plans. Despite receiving a request that its coverage be mandated, it refused, on the ground that we lack evidence that it benefits patients.

Here’s how the June 2016 “Proposed Decision Memo for Gender Dysphoria and Gender Reassignment Surgery” put it:


Based on a thorough review of the clinical evidence available at this time, there is not enough evidence to determine whether gender reassignment surgery improves health outcomes for Medicare beneficiaries with gender dysphoria. There were conflicting (inconsistent) study results—of the best designed studies, some reported benefits while others reported harms. The quality and strength of evidence were low due to the mostly observational study designs with no comparison groups, potential confounding, and small sample sizes. Many studies that reported positive outcomes were exploratory type studies (case-series and case-control) with no confirmatory follow-up.
The final August 2016 memo was even more blunt. It pointed out:

Overall, the quality and strength of evidence were low due to mostly observational study designs with no comparison groups, subjective endpoints, potential confounding (a situation where the association between the intervention and outcome is influenced by another factor such as a co-intervention), small sample sizes, lack of validated assessment tools, and considerable lost to follow-up.
That “lost to follow-up,” remember, could be pointing to people who committed suicide.

And when it comes to the best studies, there is no evidence of “clinically significant changes” after sex reassignment:

....

The study identified increased mortality and psychiatric hospitalization compared to the matched controls. The mortality was primarily due to completed suicides (19.1-fold greater than in control Swedes), but death due to neoplasm and cardiovascular disease was increased 2 to 2.5 times as well. We note, mortality from this patient population did not become apparent until after 10 years. The risk for psychiatric hospitalization was 2.8 times greater than in controls even after adjustment for prior psychiatric disease (18 percent). The risk for attempted suicide was greater in male-to-female patients regardless of the gender of the control. Further, we cannot exclude therapeutic interventions as a cause of the observed excess morbidity and mortality. The study, however, was not constructed to assess the impact of gender reassignment surgery per se.
These results are tragic. And they directly contradict the most popular media narratives, as well as many of the snapshot studies that do not track people over time. As the Obama Centers for Medicare and Medicaid pointed out, “mortality from this patient population did not become apparent until after 10 years.”

So when the media tout studies that only track outcomes for a few years, and claim that reassignment is a stunning success, there are good grounds for skepticism.