Effects of early palliative care intervention on medical resource use among end-of-life patients

Abstract

Background: In Taiwan, as the population ages, palliative care services (PCS) have expanded significantly to include comprehensive benefit plans for critically ill individuals, supported by reimbursements from the National Health Insurance program. However, incorporating palliative care into the medical management of these patients presents several challenges. We aim to evaluate the effects of palliative care interventions on medical resources in end-of-life scenarios, to promote earlier palliative care access and provide high-quality healthcare services for patients.

Methods: A total of 2202 patients were included in this study. Primary diagnosis and referral for PCS were assessed using ICD-10 and HNI code. All study subjects were divided into three groups: patients who did not receive PCS (no-PCS), patients who received PCS before their final hospital admission (PCS-before), and patients who received PCS after their final admission (PCS-after). We evaluated (i) the effects of PCS on eight medical resource utilization outcomes within the 30 days preceding death and (ii) the effects of early intervention on two major diseases.

Results: Initiating PCS before a patient's last hospital admission was associated with less aggressive medical interventions in the 30 days before death, including reduced length of intensive care unit (ICU) [odds ratio (OR) = 0.25], and rates of endotracheal intubation (OR = 0.12), respiratory ventilator support (OR = 0.20), cardiopulmonary resuscitation (OR = 0.18), and blood transfusion (OR = 0.65). Among patients with cancer and lung diseases, those who received PCS prior to their final hospitalization of over 14 days experienced reduced hospitalization duration (OR = 0.52 and 0.24, respectively). Patients with lung disease also had significantly lower odds of ICU stays (OR = 0.44) and respiratory ventilation (OR = 0.33).

Conclusion: The timing of palliative care intervention critically impacts on duration of hospitalization and ICU stay and the need for intubation procedures or cardiopulmonary resuscitation. The findings can help the government and medical providers in developing comprehensive palliative care policies and programs to improve care quality and patient rights.

Keywords: end-of-life; endotracheal intubation; intensive care unit; palliative care; respiratory ventilator; transfusion.

Figure 1

Flowchart of the patient enrollment process in the study.

Figure 2

Forest plot showing the ORs for healthcare resource utilization: (a) No-PCS vs. PCS started before their final admission, and (b) No-PCS vs. PCS started after the last admission. ORs were calculated following a univariate logistic regression analysis for eight independent variables related to healthcare resource utilization. Data are presented as ORs with 95% CIs.

Figure 3

Forest plot of OR for healthcare resource utilization. (a) Cancer patients had PCS intervention before their final hospitalization compared with PCS intervention after their final hospitalization. (b) Lung disease patients had PCS intervention before their final hospitalization compared with PCS intervention after their final hospitalization. (c) PCS interventions before vs. after last hospitalization in patients with cancer and lung disease. ORs after a univariate logistic regression analysis for eight independent variables related to healthcare resource utilization. Data presented as OR with 95% CI.