The central problem, in one number: 40.7%. In FY2016 — the only year with public service-category detail — Indiana paid $4.44B of its $10.90B Medicaid total out as managed-care capitation. That money leaves the state's books as a lump-sum premium. The services it actually bought appear in no public file.
By 2024, 81.6% of Indiana enrollees were in comprehensive managed care. The share of the dollar that is publicly traceable to a service has been shrinking the whole time.
$7.27B in FY2014 to $19.19B in FY2024 — 2.6x in a decade. The federal share peaked at 77.7% in FY2022 on the pandemic FMAP bump and fell back to 70.8% by FY2024, which means Indiana's own share of a much larger program grew faster than the headline. FY2025 holds three of four quarters and is drawn hollow.
Indiana ran above the national average from 2017 through 2022, then fell below it as unwinding cut enrollment. The 2024 jump on both lines is arithmetic, not care intensity: enrollment dropped while spending kept rising. 2020 is absent because Indiana has no enrollment row that year — the line is broken rather than interpolated.
Capitation is a premium, not a claim. Nothing public tells you which providers delivered what care for that $4.44B, at what charge. That detail exists only inside managed-care encounter records, which reach T-MSIS and stop there.
Long-term care dominates the part you can see: nursing facility regular and supplemental payments plus HCBS waivers are 51.6% of itemized fee-for-service spending. Acute hospital care barely registers, because most of it went out through capitation instead.
Federal share is the number that makes a state's program fragile. Indiana at 70.8% sits roughly ten points above California, New York, Florida and Massachusetts — a larger fraction of the program moves with federal policy rather than state appropriation.
Gross spend before rebates, 4,774 distinct drugs. This is the one public layer with per-unit prices, claim counts and a CMS outlier flag on the same record — the closest public proxy for payment-integrity signal.
Spend rose 47% from 2020 to 2023 and then flattened, while claims fell — price, not volume, carried the increase. 100 drugs out of 4,774 hold 55.4% of the money, so an outlier screen aimed at the top few hundred products covers most of the exposure.
Weighted average spend per dosage unit, compound annual rate. Filtered to drugs above $50M of 2024 spend.
The multi-manufacturer rows are the interesting ones. Sterile water for injection rising 42.9% a year per dosage unit across 11 manufacturers, propofol 33.5% across 12, triamcinolone 21.9% across 27 — these are commodity generics where competition should hold price down. CMS flagged 1,738 drugs as outliers in 2024, up from 1,085 in 2020.
data.medicaid.gov rejects any limit above 500 with a schema-validation message and an empty result set — HTTP 200, zero rows. A paging loop reads that as "end of data" and stops early without erroring.
18,511 rows, but each drug appears once per manufacturer and again as an Overall roll-up. Unfiltered, 2024 spend reads $222.6B against a true $111.3B. Filter Mftr_Name == 'Overall' and 4,774 real drugs remain.
The same column holds Prescribed Drugs and Total Net Expenditures. Summing it gives Indiana $37.42B against a true $10.90B. Pseudo-states — National Totals, TOTALS — sit in the state column of several files and double national figures the same way.