Table 3.
HCRUa and costsb in PwPSP and matched MHS members without PSP during the year post-index date
| N with event | Crude estimated mean | Adjusted estimated meanc | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| PSP n = 88 |
No PSP n = 264 |
p value crude | p value adjusted | PSP | No PSP | p value | PSP | No PSP | p value | ||
| Hospitalization days | 32 (36.4%) | 45 (17.0%) | < 0.001 | 0.028 | HCRU | 16.3 (10.4–25.5) | 14.9 (7.5–29.7) | 0.840 | 11.6 (6.3–21.3) | 8.7 (4.3–17.7) | 0.402 |
| Costs | 10,085 (6621–15,361) | 8292 (6056–11,353) | 0.465 | 10,275 (5752–18,353) | 6787 (3404–13,534) | 0.239 | |||||
| ED visits | 37 (42.0%) | 50 (18.9%) | 0.385 | 0.006 | HCRU | 1.1 (0.7–1.7) | 0.6 (0.4–1) | 0.084 | 1.9 (1.4–2.4) | 1.3 (0.9–1.8) | 0.012 |
| Costs | 383 (328–447) | 304 (271–341) | 0.018 | 379 (314–457) | 308 (233–406) | 0.096 | |||||
| Outpatients visits | 87 (98.9%) | 245 (92.8%) | 0.065 | 0.163 | HCRU | 45.2 (39.6–51.5) | 27.7 (24.7–31.1) | < 0.001 | 44.6 (36.3–54.8) | 26.8 (21.3–33.7) | < 0.001 |
| Costs | 2982 (2658–3346) | 1968 (1800–2152) | < 0.001 | 3007 (2517–3591) | 1917 (1562–2353) | < 0.001 | |||||
| Treatment and procedures | 82 (93.2%) | 226 (85.6%) | 0.069 | 0.104 | HCRU | 4.8 (3.9–5.8) | 7.3 (5.9–8.9) | 0.003 | 5.6 (4.1–7.7) | 8 (5.3–12) | 0.017 |
| Costs | 758 (520–1105) | 1293 (1034–1615) | 0.017 | 750 (434–1298) | 1210 (756–1937) | 0.027 | |||||
| Medications | 85 (96.6%) | 244 (92.4%) | 0.182 | 0.602 | HCRU | 126.6 (111–144.3) | 83.5 (74.6–93.4) | < 0.001 | 108.4 (87.8–133.8) | 67.4 (52.4–86.6) | < 0.001 |
| Costs | 1451 (1242–1695) | 963 (832–1115) | < 0.001 | 1215 (953–1548) | 794 (602–1049) | < 0.001 | |||||
| Overall direct costs | Costs | 11,198 (9245–13,563) | 6801 (5825–7941) | < 0.001 | 12,020 (8643–16,716) | 7258 (5041–10,449) | < 0.001 | ||||
Estimated means, 95% CI, and p value were extracted from GEE with negative binomial distribution with log link for HCRU analyses and gamma distribution with log-link function for costs’ analyses
PwPSP patients with progressive supranuclear palsy, HCRU healthcare resource utilization, ED emergency department, CI confidence interval, GEE generalized estimating equations
aPresented data were calculated for patients with ≥ 1 HCRU
bTop 1% was excluded
cAnalyses were adjusted for age, sex, socioeconomic status, chronic kidney disease, dementia, and mild cognitive impairment