Accuracy of a Simple Screening Tool (PALI-SP) to Predict Palliative Care Needing of In-Hospital Adult Patients in Five Tertiary Brazilian Hospitals

Abstract

Background: To identify near of end-of-life in-hospital patients to provide them adequate support is a complex task. A simple and easily applicable tool is still needed. The aim of this study was to measure the accuracy of a simple tool with four questions, including two surprise questions, to predict one-year mortality of adult patients. Methods: This was a multicenter prospective observational cohort accomplishing five Brazilian public tertiary hospitals. A four questions tool (PALI-SP), able to classify patients in five categories was previously developed to help the coordination of care based on disease progression. This questionnaire was applied to the caring physician of each consecutively hospital admitted patient, during seven consecutive days. Data were collected between 2014 and 2015. A proportional hazard Cox model was used to the analysis. Results: A total of 1810 patients were studied. Patients’ age was 55±19 years-old, 833(46%) were females, 798(44%) were in a medical ward, and 1231(68%) have no advanced diseases. 975(54%) of analyses were done for physicians who had ≤ 3 years since graduation. The one-year mortality was 439(24%), death occurred within 27[9,111] days after the initial evaluation, non-survivors were younger, and have no-advanced diseases in 40% (174/439), followed by metastatic cancer [91/439(21%)]. The tool had the unadjusted and adjusted accuracies of 0.716(CI-95%=0.687-0.746) and 0.769(CI-95%=0.742-0.795) respectively. The five prognostic classification groups had incremental risk-ratios of one-year mortality: Group I = reference, Group II =1.484(0.885-2.487), Group IIIA=3.114(1.634–5.936), Group IIIB = 3.728(2.962–4.692), and Group IV = 6.339(4.618–8.702). Interpretation: This simple screening tool (PALI-SP) has a good overall accuracy to predict one-year mortality of in-hospital patients. Funding Statement: Study funded by the governmental agency Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP). Declaration of Interests: The authors have no conflict of interests to declare. Ethics Approval Statement: The study was approved by the local ethics committee of the Hospital das Clinicas de Sao Paulo under the process number 507007/07/2014. Informed consent was obtained from all patients and physicians prior to data collection. Trial Registration: The authors state that there is no Brazilian or US national clinical trials registration.