Table 4. Adjusted outcomes for sepsis discharges by biomarker use categorya (N = 922,594).
| Outcome | Sepsis biomarker use category | |||
|---|---|---|---|---|
| >1 PCT | 1 PCT | 0 PCT, ≥1 CRP, and/or lactate | No sepsis biomarkers | |
| Number of discharges | 33,930 | 77,511 | 610,386 | 200,767 |
| Total hospital costs, 2016 US$ (95% CI) | ||||
| Adjusted mean | $17,478 (17,350–17,607) |
$13,915 (13,847–13,983) |
$13,546 (13,522–13,569) |
$11,639 (11,602–11,676) |
| Mean differenceb | $5839 (5668–6006) |
$2276 (2170–2381) |
$1907 (1849–1966) |
– |
| Hospital costs per day, 2016 US$ (95% CI) | ||||
| Adjusted mean | $1904 (1896–1911) |
$1889 (1884–1894) |
$1836 (1835–1838) |
$1661 (1658–1664) |
| Mean differenceb | $243 (231–256) | $228 (220–236) | $175 (171–180) | – |
| Overall length of hospital stay, days (95% CI) | ||||
| Adjusted mean | 9.23 (9.17–9.29) | 7.40 (7.36–7.43) | 7.35 (7.34–7.36) | 6.93 (6.91–6.95) |
| Mean ratio | 1.33 (1.32–1.34) | 1.07 (1.06–1.07) | 1.06 (1.06–1.06) | – |
| Duration of sepsis antimicrobial use, days (95% CI) | ||||
| Adjusted mean | 7.96 (7.91–8.01) |
6.29 (6.26–6.32) |
6.32 (6.31–6.33) |
5.80 (5.78–5.82) |
| Mean ratio | 1.37 (1.36–1.38) |
1.08 (1.08–1.09) |
1.09 (1.08–1.09) |
– |
| Total antimicrobial exposure, days (95% CI) | ||||
| Adjusted mean | 15.07 (14.96–15.19) |
11.92 (11.86–11.98) |
11.68 (11.66–11.70) |
10.12 (10.09–10.16) |
| Mean ratio | 1.49 (1.48–1.50) |
1.18 (1.17–1.18) |
1.15 (1.15–1.16) |
– |
| Discharge status, OR (95% CI) | ||||
| Died in hospital | 0.64 (0.61–0.67) |
0.88 (0.85–0.91) |
0.98 (0.96–1.00)* |
– |
| Home | Ref | Ref | Ref | – |
| Hospice | 0.88 (0.84–0.93) |
0.94 (0.90–0.98) |
1.07 (1.04–1.10) |
– |
| Other HC facility or unknown | 1.07 (1.04–1.10) |
0.99 (0.97–1.02) |
1.12 (1.11–1.14) |
– |
| Patients alive at discharge readmitted to same hospital within 30 days, OR (95% CI)c | 1.06 (1.02–1.10) |
1.04 (1.01–1.07)* |
1.03 (1.01–1.04) |
– |
CI, confidence interval; CRP, C-reactive protein; HC, healthcare; OR, odds ratio; PCT, procalcitonin.
aPatients with missing values or 0 for cost variables were excluded from the outcomes analysis.
bVersus no sepsis biomarkers (reference): 95% CIs for mean differences were calculated using bootstrapping approach (repeated 500 times).
cThe most recent readmission within 30 days after sepsis discharge was considered. Patients with readmission on the same day as prior discharge were considered as planned readmissions and were excluded from the readmission analysis. Thus, N = 790,679.
*p >0.05; not statistically significant.
Differences between groups were statistically significant (p <0.001) for all variables, except where noted.