| Popis: |
Johanna MC Broese,1,2 Rianne MJJ van der Kleij,1 Els ML Verschuur,2 Huib AM Kerstjens,3 Ewald M Bronkhorst,4 Niels H Chavannes,1 Yvonne Engels5 1Public Health & Primary Care, Leiden University Medical Center, Leiden, the Netherlands; 2Lung Alliance Netherlands, Amersfoort, the Netherlands; 3Respiratory Medicine & Tuberculosis, University of Groningen and University Medical Center Groningen, Groningen, the Netherlands; 4Health Evidence, Radboud University Medical Center, Nijmegen, the Netherlands; 5Anesthesiology, Pain & Palliative Medicine, Radboud University Medical Center, Nijmegen, the NetherlandsCorrespondence: Johanna MC Broese, Department of Public Health and Primary Care, Leiden University Medical Centre, Post Zone V0-P, Postbox 9600, Leiden, 2300 RC, the Netherlands, Email j.m.c.broese@lumc.nlBackground: Difficulty predicting prognosis is a major barrier to timely palliative care provision for patients with COPD. The ProPal-COPD tool, combining six clinical indicators and the Surprise Question (SQ), aims to predict 1-year mortality as a proxy for palliative care needs. It appeared to be a promising tool for healthcare providers to identify patients with COPD who could benefit from palliative care.Objective: To externally validate the ProPal-COPD tool and to assess user experiences.Methods: Patients admitted with an acute exacerbation COPD were recruited across 10 hospitals. Demographics, clinical characteristics and survival status were collected. Sensitivity, specificity, positive and negative predictive values of the tool using two cut-off values were calculated. Also, predictive properties of the SQ were calculated. In monitoring meetings and interviews, healthcare providers shared their experiences with the tool. Transcripts were deductively coded using six user experience domains: Acceptability, Satisfaction, Credibility, Usability, User-reported adherence and Perceived impact.Results: A total of 523 patients with COPD were included between May 2019 and August 2020, of whom 100 (19.1%) died within 12 months. The ProPal-COPD tool had an AUC of 0.68 and a low sensitivity (55%) and moderate specificity (74%) for predicting 1-year all-cause mortality. Using a lower cut-off value, sensitivity was higher (74%), but specificity lower (46%). Sensitivity and specificity of the SQ were 56% and 73%, respectively (AUC 0.65). However, healthcare providers generally appreciated using the tool because it increased awareness of the palliative phase and provided a shared understanding of prognosis, although they considered its outcome not always correct.Conclusion: The accuracy of the ProPal-COPD tool to predict 1-year mortality is limited, although screening patients with its indicators increases healthcare providers’ awareness of palliative care needs and encourages them to timely initiate appropriate care.Keywords: chronic obstructive pulmonary disease, prognostication, palliative care, advance care planning, Surprise Question |
