Skip to main content

Table 4 Results of the regression analyses on health care utilization 6 months before death for all people dying from colorectal cancer, divided between those not married, married and previously married

From: Till death do us part: the effect of marital status on health care utilization and costs at end-of-life. A register study on all colorectal cancer decedents in Norway between 2009 and 2013

 Never marriedMarriedPreviously marriedModel
Reference value aMeanp-valuesSE95% CIMeanp-valuesSE95% CI
Living situation
 Home 104.6830.42***(2.66)25.1935.647.06*(2.78)1.6112.51Two part model (logistic regression/OLS)
 Long-term institution 43.56−29.16***(2.82)−34.69−23.62−8.42**(2.95)−14.20−2.64Two part model (logistic regression/OLS)
 Short-term institution 23.04−10.12***(1.90)−13.85−6.40−3.65 (1.97)−7.510.20Negative Binomial 1
 Hospital19.232.90***(0.77)1.394.411.21 (0.81)−0.372.79Negative Binomial 1
Health care utilization
 Inpatient stay 3.320.26 (0.21)−0.150.67−0.02 (0.20)−0.410.37Negative Binomial 2
 Outpatient treatments 5.772.57***(0.27)2.043.100.89**(0.27)0.361.43Negative Binomial 1
 GP consultations 10.232.49***(0.40)1.703.270.84*(0.41)0.031.64Negative Binomial 1
 ER visits 1.52−0.06 (0.07)−0.200.090.09 (0.08)−0.050.24Negative Binomial 1
 Hours home nursing 56.36−6.59 (3.38)−13.210.03−0.52 (3.51)−7.406.36Negative Binomial 1
 Hours practical assistance 13.17−9.36***(1.21)−11.73−7.00−1.39 (1.07)−3.490.72Negative Binomial 1
Costs
 Specialist health care204,16441,535***(8,746)24,39258,6779,818 (8,339)− 6,52526,161GLM (Identity/gamma)
 Primary health care9,7272,051***(540)9923,109303 (529)− 7331,340GLM (Identity/gamma)
 Home- and community-based care241,005− 137,604***(13,256)− 163,586−111,622−50,943***(13,944)−78,271−23,611GLM (Log/gamma)
 Total costs434,964−65,621***(10,922)−87,026−44,215−20,321 (11,279)−42,4301,788GLM (Identity/gaussian)
  1. The covariates age, gender, education, income, comorbidities and time since diagnosis was included in all analyses
  2. aThe predicted absolute values on their original scale (i.e. the model predicts that never married patients spend 104 days at home). These can be used to estimate the average value for the married and previously married. E.g. the predicted number of days Married patients spent at home was 135.09 (104.68 + 30.42)
  3. * = p < 0.05, ** = p < 0.01, *** = p < 0.001
\