Table 3

Details of end-of-life discussion

ILDLung cancerP valueAdjusted p value*
Total (n)n (%)Total (n)n (%)
Holding EOL discussion8376 (91.6)8985 (95.5)0.360.06
Patient participation in EOL discussion7631 (40.8)8553 (62.4)0.0070.004
Topic discussed in EOL discussion6982
 Place of death10 (14.5)47 (57.3)<0.001<0.001
 Resuscitation46 (66.7)28 (34.1)<0.001<0.001
 Use of specialist palliative care service17 (24.6)30 (36.6)0.160.13
 Transfer to other facilities23 (33.3)25 (30.5)0.730.81
Provider of the initial EOL discussion75840.950.65†
 Pulmonologist65 (86.7)75 (89.3)
 Palliative care physician3 (4.0)3 (3.6)
 Primary care physician3 (4.0)2 (2.4)
 Other4 (5.3)4 (4.8)
Setting where EOL discussion initiated75840.010.01‡
 Inpatient62 (82.7)55 (65.5)
 Outpatient10 (13.3)27 (32.1)
 During home visit service3 (4.0)2 (2.4)
Timing of initiating EOL discussion74840.170.051§
 >3 months before the patient’s death20 (27.0)28 (33.3)
 1–3 months before the patient’s death21 (28.4)31 (36.9)
 <1 month before the patient’s death33 (44.6)25 (29.8)
  • Each parameter is expressed as number (percentage).

  • Parameters in each group were compared using Fisher’s exact test.

  • *To calculate adjusted p values, logistic regression models were used. All models were adjusted for patient’s age at death and patient’s sex.

  • †The dependent variable, provider of the initial EOL discussion, was handled as a dichotomous variable (ie, pulmonologist or others).

  • ‡The dependent variable, setting where EOL discussion initiated, was handled as a dichotomous variable (ie, pulmonologist or others).

  • §The dependent variable, timing of initiating EOL discussion, was handled as a dichotomous variable (ie, <1 month before the patient’s death or others).

  • EOL, end of life; ILD, interstitial lung disease.