Health insurance is such a complicated product and it's so hard to be an educated consumer. A story about Rivka*, who could be on the hook for almost $7,000 this year due to a change buried on page 128 of her plan benefits document.
* Fake name enthusiastically chosen by subject Rivka gets an infusion of a prescription drug at her doctor's office every 1 to 2 months for a chronic condition. The drug really helps! And it's been affordable - the copay was $25.
Then in January she got a bill in the mail of $1,200 (20% coinsurance) for her latest infusion.

with the explosion in #COVID19 cases in NYC and around the country, you may be wondering what hospitals will be paid to treat these patients. in this thread i'll walk through how that's going to work. i'll assume the hospital bills medicare: half of #COVID19 patients are 65+, other insurers often use the same or similar payment approaches, & hospitals might get medicare rates for uninsured patients under the CARES act. i'll account for the 20% payment bump in the CARES act too

i have updated my CMS hospital cost report (HCRIS) data and processing code. it's now much more up to date, and it includes data on ICU and other critical care beds potentially relevant for #COVID19 work. 2000-2019 data here: github.com/asacarny/hospi… the update adds variables for beds in:
- intensive care unit
- coronary care unit
- burn intensive care unit
- surgical intensive care unit
- other special care unit
adult+pediatric+neonatal beds are not disaggregated, but pediatric+neonatal beds are mostly in special care unit.

some thoughts & data inspired by the thought-provoking and carefully conducted @rkwadhera @rwyeh @kejoynt @Changyushen312 et al study on CMS hospital readmissions reduction program (HRRP). to get things started, here's some info on pneumonia index admissions in FFS medicare. @rkwadhera @rwyeh @kejoynt @Changyushen312 the question is: did HRRP cause an increase in mortality? while medical journals don't allow causal language outside RCTs, i'm talking about this research because i think it's informative about the causal effect! otherwise it's just 'association of 2012-2015 with mortality'

a lil thread about the magic of randomization inference and blinded data analysis.

@ml_barnett @andrewolenski & i are doing a study where we already know our confidence intervals but have no idea what our point estimates are! and here’s why i think that’s so cool. @ml_barnett @andrewolenski in 2015 CMS sent strong letters to prescribers of antipsychotic Seroquel. we showed they reduced Seroquel Rx ~15%.

now we're studying if the letters had peer effects, i.e. did they affect doctors who worked with the ones who received a letter?

https://twitter.com/asacarny/status/1024714768916066305