Congratulations, Jose! This is well deserved.

One final note for @gregsargent.bsky.social and others asking, this evidence, on policies that save or cost lives for people living in every state, is at risk with cuts to federal research funds.

3. Small choices, like the choice to let dual Medicare-Medicaid beneficiaries keep the LIS for some months after Medicaid loss, save many lives.

2. Disruptions in insurance coverage have many harms. Yet, our patchwork health insurance is full of such “unforced errors”. Gaps in coverage rise after reaching key age, income, or other eligibility cutoffs, or because people struggle to show they are still eligible.

There are many lessons. 1. Lower drug costs save lives. Efforts to lower drug costs to patients while supporting high-value innovation are needed.

17 months after losing Medicaid, deaths were 3 per 1000 HIGHER in the group losing the LIS sooner. Note that these people still had Medicare coverage.

Depending on the month they leave Medicaid, some lose the LIS “early” within 6-12 months and some lose LIS “late”, 12-18 months later.

Eric Roberts of @pennmedicine.bsky.social and @joefigs.bsky.social of @hsph.harvard.edu led this clever and careful study of dual Medicare-Medicaid beneficiaries. Each year 7-8% of these folks lose Medicaid & with it, help with drug costs, the low income subsidy (LIS).

I’d say that 1. VBC has a role, not that it is “the” driver 2. 1 reason it’s hard to measure effect is that comparison groups have lots of VBC too and 3. The effect has more to do with direction of systems away from (v. Toward) FFS (that is the zeitgeist).