Skip to main content

Table 3 Regression results main model M1 (beta, lower and upper confidence interval)

From: Stroke units, certification, and outcomes in German hospitals: a longitudinal study of patient-based 30-day mortality for 2006–2014

  M1
Log dummy −0.059* (−0.112, −0.007)
Stroke unit (SU) − 0.056** (− 0.092, − 0.021)
SU certification − 0.005 (− 0.036, − 0.026)
THQ certification 0.016 (− 0.019, 0.051)
Log stroke case volume − 0.003 (− 0.010, 0.003)
Share stroke ICD/all ICD − 0.054 (− 0.743, 0.635)
Hospital beds − 0.000 (− 0.000, 0.000)
Teaching hospital status 0.007 (− 0.041, 0.056)
CMS ICD specialization index − 0.003 (− 0.039, 0.033)
Private, for-profit hospitalsa − 0.022 (− 0.144, 0.101)
Public hospitalsa − 0.011 (− 0.138, 0.117)
Log all OPS stroke procedures
Interaction SU and THQ certification
Constant 0.182* (0.030, 0.333)
R2-within 0.016
R2-between 0.032
R2-overalll 0.018
BIC 1084
Intraclass correlation 0.512
F-statistic 3.9
Number of observations 7376
  1. QSR stroke patient volume applied as analytical weights; *p < 0.05, **p < 0.01, ***p < 0.001; time-fixed effects not displayed separately (in M1, β2008 = − 0.02, β2010 = − 0.03, β2012 = − 0.03, β2013 = − 0.05***, β2014 = − 0.08***), test for joint significance of time effects in M1 with F-statistic of 5.59
  2. aprivate (non-profit) hospitals serve as reference category