Mary Joanne Verhoef

Surprise Question and performance status indicate urgency of palliative care needs 95 (ECOG) scale.37 Data abstraction was performed using a code book about which consensus was reached among members of the scientific team of the hospital’s PCCT. The code book contained definitions of all variables for data collection and data were imported into a digital database accordingly. Four chart abstractors were trained on how to use the code book. One expert of the PCCT monitored the work of the chart abstractors, was available in cases of doubt and performed random checks to assess interrater agreement. Electronic patient records (EPRs) were assessed to collect additional patient-, referral-, ED-visit- and follow-up characteristics and ECOG performance score of S-patients. The disease phase was evaluated and classified according to Lynn and Adamson: diseasemodifying phase (anti-cancer treatment to prolong life and/or symptom management); symptom-management phase (treatment directed to symptom-relief ); or terminal phase.38 The EPRs were searched for reports about PCCT-consultations 3 months before the current ED-visit. ‘Limitations on life-sustaining treatments (LSTs)’ included the codes: ‘do-not-resuscitate’; ‘do-not-ventilate’; and ‘no admission to the intensive care unit’ (ICU). Time of arrival at the ED was within office hours if the visit occurred Monday-Friday between 8am-6pm. The ‘main symptom’ was the referring symptom for the ED-visit according to the attending physician’s notes in the EPR. A ‘new symptom’ was a main symptom not described in the EPR <3 months before the ED-visit. Main symptoms were considered ‘acute symptoms’ if the onset was within one working day before the ED-visit. Using the Dutch Edmonton Symptom Assessment Scale (ESAS), symptoms were scored as present or absent based on documentation in the EPRs.39 The clinical diagnosis was defined as the conclusion of the attending physician at the ED. Statistical analysis Characteristics of NS- and S-patients were compared using the Chi-square test and Fisher’s exact test (nominal variables) and Mann-Whitney U test (ordinal variables and not-normally distributed continuous variables). Survival was estimated using KaplanMeier’s methodology. A log-rank test was used to compare overall survival between NS- and S-groups. Prognostic value of the SQ Two-by-two tables were used to calculate test characteristics (sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV)) of the SQ as screening tool for death <1 year. Harrell’s c-index was calculated to estimate the discriminative power of the SQ for death <1 year.40,41 Addition of other predictors of approaching death to the SQ To identify predictors of palliative care needs, univariable regression analyses using Cox proportional hazards models were performed to estimate the association between death 5

RkJQdWJsaXNoZXIy MTk4NDMw