Skip to main content

Table 3 Multivariable Negative Binomial Regression on Hospital Utilization from Year 2013 to 2015

From: Predictive ability of an expert-defined population segmentation framework for healthcare utilization and mortality - a retrospective cohort study

  IRR 95% Confidence Interval p-value
No. of ED visits from 2013 to 2015
 Mostly Healthy Reference
 Serious Acute 1.45 (1.42, 1.47) < 0.001
 Stable Chronic 2.15 (2.09, 2.21) < 0.001
 Complex Chronic without Frequent Hospital Admissions 3.36 (3.29, 3.44) < 0.001
 Complex Chronic with Frequent Hospital Admissions 14.52 (13.49, 15.64) < 0.001
 End of Life 9.56 (8.51, 10.75) < 0.001
No. of SOC visits from 2013 to 2015
 Mostly Healthy Reference
 Serious Acute 1.03 (1.02, 1.04) < 0.001
 Stable Chronic 1.91 (1.88, 1.94) < 0.001
 Complex Chronic without Frequent Hospital Admissions 2.67 (2.64, 2.71) < 0.001
 Complex Chronic with Frequent Hospital Admissions 7.71 (7.31, 8.13) < 0.001
 End of Life 11.50 (10.68, 12.39) < 0.001
No. of Hospital Admissions from 2013 to 2015
 Mostly Healthy Reference
 Serious Acute 1.33 (1.31, 1.36) < 0.001
 Stable Chronic 3.16 (3.08, 3.25) < 0.001
 Complex Chronic without Frequent Hospital Admissions 3.86 (3.78, 3.94) < 0.001
 Complex Chronic with Frequent Hospital Admissions 22.66 (21.07, 24.37) < 0.001
 End of Life 16.18 (14.49, 18.07) < 0.001
  1. Abbreviations: ED Emergency Department, SOC Specialist Outpatient Clinic, IRR Incidence Rate Ratio. Model is adjusted for age, gender, ethnicity, past utilization and use of survival time as exposure variables