Skip to main content
. 2022 Feb 22;5(2):e220158. doi: 10.1001/jamanetworkopen.2022.0158

Table 3. Rate Ratios Preimplementation and Postimplementation by Order Set Use.

Characteristic Hospitalizations, No. (%)a Difference in differences
With use of order set Without use of order set Unadjusted
RR (95% CI)
P value Adjusted
RR (95% CI)b
P value
Before implementation (n = 8237) After implementation (n = 4652) P value Before implementation (n = 6301) After implementation (n = 3709) P value
Inpatient mortality 107 (1.3) 93 (2.0) .002 126 (2.0) 166 (4.5) <.001 0.69 (0.48-0.98) .04 0.86 (0.37-1.97) .72
Hospital LOS, median (IQR), d 2.9 (1.8-5.0) 2.9 (1.8-4.8) .44 2.8 (1.4-5.0) 3.2 (1.9-5.9) <.001 0.84 (0.78-0.91) <.001 0.71 (0.58-0.86) <.001
ICU admission 2310 (28.0) 1229 (26.4) .047 1284 (20.4) 841 (22.7) .007 0.85 (0.78-0.94) .001 0.71 (0.56-0.89) .003
Readmission 1433 (17.6) 718 (15.7) .007 1191 (19.3) 766 (21.6) .006 0.93 (0.79-1.08) .34 0.80 (0.59-1.08) .14

Abbreviations: COPS2, Comorbidity Point Score, version 236; ICU, intensive care unit; LAPS2, Laboratory Acute Physiology Score, version 236; LOS, length of stay; RR, rate ratio.

a

Preimplementation and postimplementation period data are presented as unadjusted number (percentage) of encounters for binary outcomes and as mean (SD) for hospital LOS.

b

Adjusted for age, sex, comorbidity (COPS2) and acuity (LAPS2) scores, prior alcohol withdrawal syndrome hospitalization in the prior 6 months, observation vs inpatient admission, weekend admission, urine toxicology, and month of year and preintervention vs postimplementation time trends.