VARDR Program Review
Missouri

Synthetic data. Practice names are invented. Several practices carry planted billing patterns, and several benign look-alikes are planted to show what the analysis declines to flag. Cases, people and outcomes are invented too.

Program integrity · Medicaid · Missouri

Billing Pattern Watch

Paid claim lines compared with same-specialty peers: codes per visit, units, office-visit levels, modifiers, spend per patient and growth. A flag is a statistical anomaly for analyst review under 42 CFR 455.13–.14. It is not a finding of improper payment, and never a conclusion about intent.

Practices compared
42
2026Q3, against 6 quarters
Priority review
5
two or more independent signals, persistent
Scheduled review
1
one persistent signal
Expected false leads
0.00
among this quarter's flags, by the method's own estimate
Period: 2026Q3, with 6 quarters of history for persistence and growth. The synthetic extract is complete, so no quarter is held back; an agency's extract holds quarters inside a 90-day window.

Where each practice sits against its peers

Choose a measure. Select a mark to open the practice.

Priority reviewScheduled reviewRoutine95% limits99.8% limits
0.0×0.5×1.0×1.5×2.0×2.5×5102050100200500peer rateExpected count at peer rates (volume, log scale)Observed ÷ expectedMotion Physical TherapyClearwater Counseling

Each mark is a practice in the latest quarter. The funnel narrows as volume grows, so a small practice needs a larger ratio to sit outside it. Limits allow for overdispersion (φ = 1.0). The tier reflects every measure, persistence and the false-discovery control, not this chart alone.

Review list

Ordered by tier, then by the number of independent signal families, then by paid amount. Correlated measures count once per family.

Show only practices on review
PracticeTierSignalsLeading observationPaid in window
Maple Avenue Clinic
Family medicine · Boone County
Routine
Within the range of peers$117,025
Core Motion Rehab
Physical therapy · Butler County
Routine
Within the range of peers$116,374
Compass Counseling
Mental health counseling · Jackson County
Routine
Within the range of peers$112,542
Northside Family Care
Family medicine · Jasper County
Routine
Within the range of peers$109,278
Willow Counseling Group
Mental health counseling · Jasper County
Routine
Within the range of peers$109,177
Harbor Primary Care
Family medicine · Jackson County
Routine
Within the range of peers$104,770
Summit Family Medicine
Family medicine · Greene County
Routine
Within the range of peers$104,588
Stillwater Wellness
Mental health counseling · Butler County
Routine
Within the range of peers$94,245
Pine Hollow Family Practice
Family medicine · Butler County
Routine
Within the range of peers$93,824
Southfield Family Health
Family medicine · Jasper County
Routine
Within the range of peers$93,785
Cedar Grove Primary Care
Family medicine · Jackson County
Routine
Within the range of peers$89,447
Fresh Start Counseling
Mental health counseling · Boone County
Routine
Within the range of peers$88,215
Greenleaf Therapy
Mental health counseling · Jackson County
Routine
Within the range of peers$86,387
Elm Park Medical
Family medicine · Boone County
Routine
Within the range of peers$79,655
Lakeview Medical
Family medicine · Butler County
Routine
Within the range of peers$79,116
Westgate Health
Family medicine · Jasper County
Routine
Within the range of peers$77,960
Bridgeway Therapy
Mental health counseling · Boone County
Routine
Within the range of peers$26,023

Dollars by practice

The same paid dollars, divided different ways. A practice that leads on the raw total and falls away per patient is large; one that leads per patient is worth a look.

How it is computedDollars paid to the practice, 2026Q3. Not divided by anything.

Across the selection: half of practices are at or below $22,777, and nine in ten are at or below $61,687.

  1. Caring Hands Home Care$174,133
  2. Motion Physical Therapy$70,348
  3. HomeFirst Personal Care$65,076
  4. Kindred Spirit Home Care$64,806
  5. Hearthside Home Services$62,195
  6. Daily Living Supports$57,115
  7. Comfort Keepers of the Valley$50,426
  8. Golden Years Aides$48,968
  9. New Path Behavioral$48,762
  10. Trusted Companions$43,740
  11. Clearwater Counseling$39,908
  12. Prairie Health Partners$35,201
  13. Stride Physical Therapy$30,520
  14. Kinetic Rehab Center$29,720
  15. Riverbend Family Practice$29,297
  16. Balance Point Therapy$28,248
  17. Peak Rehab$24,868
  18. Haven Counseling Center$23,816
  19. Restore PT$23,514
  20. Frontier Rehab$23,401
  21. Lighthouse Therapy$22,969
  22. ActiveCare PT$22,584
  23. Anchor Behavioral Health$21,210
  24. Meadowbrook Clinic$20,176
  25. Maple Avenue Clinic$19,575

Checked, not flagged

Practices that resemble a pattern and were left alone, with the reason. A reviewer should be able to see what the analysis declined to raise.

  • Hearthside Home Services
    In-home personal care · Butler County
    • Paid per patient: not flagged — the only in-home personal care in Butler County; high per-patient spend there can reflect access, not billing.
    • Paid per visit: not flagged — the only in-home personal care in Butler County; high per-patient spend there can reflect access, not billing.
  • Golden Years Aides
    In-home personal care · Jasper County
    • Paid per patient: not flagged — the only in-home personal care in Jasper County; high per-patient spend there can reflect access, not billing.
    • Paid per visit: not flagged — the only in-home personal care in Jasper County; high per-patient spend there can reflect access, not billing.
  • Stride Physical Therapy
    Physical therapy · Jasper County
    • Paid per patient: not flagged — the only physical therapy in Jasper County; high per-patient spend there can reflect access, not billing.
    • Paid per visit: not flagged — the only physical therapy in Jasper County; high per-patient spend there can reflect access, not billing.
  • Prairie Health Partners
    Family medicine · Boone County
    • Growth comes from more patients (1.8×) at normal intensity — a practice still building its panel. Not flagged.
  • Frontier Rehab
    Physical therapy · Shannon County
    • Paid per patient: not flagged — the only physical therapy in Shannon County; high per-patient spend there can reflect access, not billing.
    • Paid per visit: not flagged — the only physical therapy in Shannon County; high per-patient spend there can reflect access, not billing.
  • Core Motion Rehab
    Physical therapy · Butler County
    • Paid per patient: not flagged — the only physical therapy in Butler County; high per-patient spend there can reflect access, not billing.
    • Paid per visit: not flagged — the only physical therapy in Butler County; high per-patient spend there can reflect access, not billing.
  • Bridgeway Therapy
    Mental health counseling · Boone County
    • Paid per patient: elevated this quarter only — persistence needs 2 of the last 3 quarters.

Unusually low

Low billing is a data-quality or access question (incomplete claims, under-service, a closed office) and is never routed to investigation.

No practice in this selection is unusually low this quarter.

How this is computed

  • No extract date was supplied, so every quarter is treated as complete. With one, quarters still inside their 90-day maturity window are held back.
  • Peers are the same specialty statewide; each practice is left out of its own reference.
  • Counts are shrunk toward the peer group (empirical Bayes) — a small practice needs more evidence to stand out.
  • A metric counts only if it exceeds 1.5× peers with 95% posterior probability, survives a 5% false-discovery control, and persists 2 of the last 3 quarters.
  • Practices under 30 visits or 11 patients a quarter are not compared.
  • E/M levels are age-standardised: an older panel is expected to code higher.
  • The only provider of a specialty in a county is not flagged on spend — high per-patient spend there can reflect access.
  • Growth from more patients at normal intensity is labelled expansion and never flagged.
  • Unusually LOW results go to a separate data-quality / access list, never to investigation.
  • Correlated signals are not added: each family counts once, by its strongest member.

Empirical-Bayes Gamma-Poisson against a leave-one-out peer reference; Benjamini-Hochberg across all metrics and specialties in the quarter; spend on a robust log-ratio scale; growth = patients × visits per patient × codes per visit × paid per code, compared with peer growth and county eligible growth.

Every number on this screen comes from run RUN-e13e368456792cc8, computed 2026-11-16 by engine billing-pattern-5 on input 0004c6ed954b… with parameters be6f8bfe67db…. The same input, engine and parameters always give this run.