Identification of hospital patients in need of palliative care--a predictive score

The need of patients living with cancer for palliative care

BACKGROUND

A comprehensive assessment of patients' problems and needs is essential for all patients with chronic diseases, including cancer.

AIM

This study assesses the problems, unmet needs and requirement for palliative care (PC) among patients with cancer.

METHOD

A descriptive cross-sectional design was employed using a valid self-reported questionnaire.

RESULTS

On average, 62% of patients had problems that were unresolved. The need for patients to have more information about their health was identified (75.1%), followed by financial problems because of the illness and ability to afford healthcare (72.9%), and psychological issues, such as depression, anxiety and stress (67.1%). Patients stated that their spiritual needs were not being met (78.8%), and that they were experiencing psychological distress and problems with daily living that needed to be addressed through PC (78%, 75.1%, respectively). A chi-square test revealed that all problems are significantly associated with the need for PC (P<.001).

CONCLUSION

Patients needed more assistance in psychological, spiritual, financial and physical domains, and this can be provided by palliative care. Palliative care in low-income countries is a human right for patients with cancer.

Preliminary results regarding automated identification of patients with a limited six-month survival prognosis using nursing assessment in uro-oncology patients

INTRODUCTION

Contrary to current recommendations, palliative co-management of tumor patients often occurs late in daily clinical practice. Palliative care specialist (PCS) co-management should be considered at the latest after a 6-month prognosis has been presumed. Therefore, identifying patients with a limited prognosis is a reasonable measure.

METHODS

Patients were identified using a screening tool for limited prognosis, which combined their tumor stage and data from the nursing anamnesis. In this retrospective study, a monocentric cohort of patients with urological malignancies-UICC (Union for International Cancer Control) stages III and IV - were enrolled from March to December 2019, with a 6-month follow-up period ending in May 2020.

RESULTS

Most patients were male and suffered from prostate cancer. Patients with uro-oncological tumors dying within 6 months correlated significantly with the presence of repeated hospitalizations within three months, pain on admission, malnutrition, impaired breathing and reduced mobility (P < 0.001). The test was fair in quality (AUC 0.727) at a cut-point of five; a sensitivity of 97% and a specificity of 25% were obtained. The PPV was 0.64 and NPV was 0.82.

DISCUSSION/CONCLUSION

We specifically identified the predictors of limited prognosis in urological cancer patients across several entities using an automated scoring system based on tumor stage and data from the nursing anamnesis. Therefore, we recognized hospitalization as an important transition point and determined nurses to be valuable partners in identifying unmet palliative care needs without additional technical, personnel or financial effort.

Comparison of the JCAHO Scoring System and the ESAS Scoring System in Determining the Palliative Care Needs of Gynecological Cancer Patients Treated at Hasan Sadikin Hospital

OBJECTIVE

The purpose of this study was to analyze the relationship between quality of life with the JCAHO and the ESAS scoring system, and to compare the JCAHO and the ESAS scoring system in determining the palliative care needs of gynecological cancer patients treated at RSHS.

METHOD

The subjects of this study were all gynecological cancer patients who were treated at RSHS in May-August 2020. This study was an analytic study with a cross sectional design. The data of this study were obtained from interviews, questionnaires and patient medical records, the study was analyzed bivariate using chi square with α = 0.05.

RESULTS

The results showed that the quality of life of patients with gynecological cancer was associated with the JCAHO palliative score (p <0.05), the better the patient's quality of life, the better the JCAHO palliative score. The quality of life of gynecological cancer patients was related to ESAS (p <0.05), the better the patient's quality of life, the better the ESAS. There was difference between the JCAHO palliative score and the ESAS in determining the palliative care needs of gynecological cancer patients (p< 0.05).

CONCLUSION

Quality of life has correlation with palliative scores, the lower the palliative score, the better the quality of life. This study showed significant difference between the JCAHO palliative score and the ESAS in determining the palliative care needs of gynecological cancer patients. The JCAHO palliative score measures objectively how the patient is on admission for treatment, this score not only measures the intensity of symptoms but measures the underlying disease, comorbid disease, functional status of the patient and other criteria for the patient.  ESAS assesses the intensity of symptoms, the assessment of palliative care needed can change rapidly if the intensity of symptoms in patients changes.

Outpatient palliative care referral system (PCRS) for patients with advanced cancer: an impact evaluation protocol

INTRODUCTION

Early palliative care (PC) in the clinical pathway of advanced cancer patients improves symptom control, quality of life and has a positive impact on overall quality of care. At present, standardised criteria for appropriate referral for early PC in oncology care are lacking. The aim of this project is to develop a set of standardised referral criteria and procedures to implement appropriate early PC for advanced cancer patients (the palliative care referral system, PCRS) and test its impact on user perception of quality of care received, on patient quality of life and on the use of healthcare resources.

SETTING

Selected oncology clinics and PC outpatient clinic.

METHODS AND ANALYSIS

A scoping literature review and an expert consultation through a nominal group technique will be used to revise existing referral tools and to develop a new one, the PCRS. 25 patients will be enrolled in a pilot study to assess feasibility of the implementation of PCRS; 10 interviews with patients and healthcare professionals will be carried out to evaluate applicability.A pretest-post-test quasiexperimental study involving 150 patients before implementation of the PCRS and 150 patients after implementation will be carried out.Patient satisfaction with care received, quality of life and use of resources, and caregiver satisfaction with care will also be assessed to explore the impact of the intervention.

ETHICS AND DISSEMINATION

Ethical approval for the study has been granted by the Institutional Review board of the Fondazione IRCCS Istituto Nazionale Tumori; approval reference INT201/19.Results will be disseminated through open access publications and through scientific communication presented at national and international conferences.

TRIAL REGISTRATION NUMBER

NCT04936568.

Improving time to palliative care review with predictive modeling in an inpatient adult population: study protocol for a stepped-wedge, pragmatic randomized controlled trial

Background

Palliative care is a medical specialty centered on improving the quality of life (QOL) of patients with complex or life-threatening illnesses. The need for palliative care is increasing and with that the rigorous testing of triage tools that can be used quickly and reliably to identify patients that may benefit from palliative care.

Methods

To that aim, we will conduct a two-armed stepped-wedge cluster randomized trial rolled out to two inpatient hospitals to evaluate whether a machine learning algorithm accurately identifies patients who may benefit from a comprehensive review by a palliative care specialist and decreases time to receiving a palliative care consult in hospital. This is a single-center study which will be conducted from August 2019 to November 2020 at Saint Mary’s Hospital & Methodist Hospital both within Mayo Clinic Rochester in Minnesota. Clusters will be nursing units which will be chosen to be a mix of complex patients from Cardiology, Critical Care, and Oncology and had previously established relationships with palliative medicine. The stepped wedge design will have 12 units allocated to a design matrix of 5 treatment wedges. Each wedge will last 75 days resulting in a study period of 12 months of recruitment unless otherwise specified. Data will be analyzed with Bayesian hierarchical models with credible intervals denoting statistical significance.

Discussion

This intervention offers a pragmatic approach to delivering specialty palliative care to hospital patients in need using machine learning, thereby leading to high value care and improved outcomes. It is not enough for AI to be utilized by simply publishing research showing predictive performance; clinical trials demonstrating better outcomes are critically needed. Furthermore, the deployment of an AI algorithm is a complex process that requires multiple teams with varying skill sets. To evaluate a deployed AI, a pragmatic clinical trial can accommodate the difficulties of clinical practice while retaining scientific rigor.

Trial registration

ClinicalTrials.gov NCT03976297. Registered on 6 June 2019, prior to trial start.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-021-05546-5.

Instruments for the identification of patients in need of palliative care: protocol for a systematic review of measurement properties

OBJECTIVE

The objective of this review is to provide a comprehensive overview of the psychometric properties of available clinician-reported instruments developed to identify patients in need of general and specialized palliative care in acute care settings.

INTRODUCTION

Identification of patients in need of palliative care has been recognized as an area where many health care professionals need guidance. Differentiating between patients who require general palliative care and patients with more complex conditions who need specialized palliative care is particularly challenging. To our knowledge, no dedicated instruments are available to date to assist health care professionals to make this identification.

INCLUSION CRITERIA

Included studies will report on i) instruments aiming to identify patients in need of palliative care, ii) adult patients in need of palliative care in acute-care settings, iii) clinician-reported outcome measures, or iv) the development process or one or more of its measurement properties. Studies conducted in intensive care units, emergency departments, or nursing homes will be excluded.

METHODS

We will search for studies published in English and French in a variety of sources, including Embase, Medline Ovid SP, PubMed, CINAHL EBSCO, Google Scholar, government websites, and hospice websites. All citations will be screened and selected by two independent reviewers. Data extraction, quality assessment, and syntheses of included studies will be performed according to the COnsensus-based Standards for the selection of health Measurement Instruments (COSMIN) criteria.

SYSTEMATIC REVIEW REGISTRATION NUMBER

PROSPERO CRD42020150074.

Hospitalised patients with palliative care needs: Spain and Sweden compared

OBJECTIVES

This study aimed to describe and compare symptoms, care needs and types of diagnoses in hospitalised patients with palliative care needs in Spain and Sweden.

METHODS

A cross-sectional, population-based study was carried out at two hospitals in both Spain and Sweden. Using a questionnaire, we performed 154 one-day inventories (n=4213) in Spain and 139 in Sweden (n=3356) to register symptoms, care needs and diagnoses. Descriptive analyses were used.

RESULTS

The proportion of patients with care needs in the two countries differed (Spain 7.7% vs Sweden 12.4%, p<0.001); however, the percentage of patients with cancer and non-cancer patients was similar. The most prevalent symptoms in cancer and non-cancer patients in both countries were deterioration, pain, fatigue and infection. The most common cancer diagnosis in both countries was lung cancer, although it was more common in Spain (p<0.01), whereas prostate cancer was more common among Swedish men (p<0.001). Congestive heart failure (p<0.001) was a predominant non-cancer diagnosis in Sweden, whereas in Spain, the most frequent diagnosis was dementia (p<0.001). Chronic obstructive pulmonary disease was common in both countries, although its frequency was higher in Spain (p<0.05). In total, patients with cancer had higher frequencies of pain (p<0.001) and nausea (p<0.001), whereas non-cancer patients had higher frequencies of deterioration (p<0.001) and infections (p<0.01).

CONCLUSIONS

The similarities in symptoms among the patients indicate that the main focus in care should be on patient care needs rather than diagnoses. Integrating palliative care in hospitals and increasing healthcare professional competency can result in providing optimal palliative care.

Comparison of Artificial Neural Networks and Logistic Regression for 30-days Survival Prediction of Cancer Patients

Introduction

A machine learning technique that imitates neural system and brain can provide better than traditional methods like logistic regression for survival prediction and create an algorithm by determining influential factors.

Aim

To determine the influential factors on survival time of palliative care cancer patients and to compare two statistical methods for better prediction of survival.

Methods

One-year data is gathered from the patients that we followed in the palliative care clinic of our hospital (2017-2018) (n = 189). All data were retrospectively evaluated. After descriptive statistics, we used Pearson and Spearman correlations for parametric and non-parametric variables. The Artificial Neural Networks (ANN) and logistic regression model were applied to parameters which have a significant correlation with short survival.

Results

Significantly correlated variables with short survival were Palliative Performance Scale (PPS), Edmonton Symptom Assessment System (ESAS), Karnofsky Performance Scale (KPS), brain, liver, and distant metastasis, hemogram parameters, cero-reactive protein (CRP) and albumin (ALB). ANN model showed 89.3% prediction accuracy while the logistic regression model showed 73.0%. ANN model achieved a better AUC value of 0.86 than logistic regression model (0.76).

Discussion

There are several prognostic evaluation tools such as PPS, KPS, CRP, albumin, leukocytes, neutrophil were reported several studies as survival-related parameters in logistic regression models, also. Many studies compare ANN with logistic regression. When we evaluated these parameters totally, we observed the same relations with survival then we used the same parameters in the ANN model. The effectivity of the survival prediction models can be improved with the use of ANN.

Conclusion

ANN provides a more accurate estimation than logistic regression. ANN model is an important statistical method for survival prediction of cancer patients.

IDentification of patients in need of general and specialised PALLiative care (ID-PALL©): item generation, content and face validity of a new interprofessional screening instrument

BACKGROUND

Early identification of patients requiring palliative care is a major public health concern. A growing number of instruments exist to help professionals to identify these patients, however, thus far, none have been thoroughly assessed for criterion validity. In addition, no currently available instruments differentiate between patients in need of general vs. specialised palliative care, and most are primarily intended for use by physicians. This study aims to develop and rigorously validate a new interprofessional instrument allowing identification of patients in need of general vs specialised palliative care.

METHODS

The instrument development involved four steps: i) literature review to determine the concept to measure; ii) generation of a set of items; iii) review of the initial set of items by experts to establish the content validity; iv) administration of the items to a sample of the target population to establish face validity. We conducted a Delphi process with experts in palliative care to accomplish step 3 and sent a questionnaire to nurses and physicians non-specialised in palliative care to complete step 4. The study was conducted in the French and Italian-speaking regions of Switzerland. An interdisciplinary committee of clinical experts supervised all steps.

RESULTS

The literature review confirmed the necessity of distinguishing between general and specialised palliative care needs and of adapting clinical recommendations to these different needs. Thirty-six nurses and physicians participated in the Delphi process and 28 were involved in the face validity assessment. The Delphi process resulted in two lists: a 7-item list to identify patients in need of general PC and an 8-item list to identify specialised PC needs. The content and face validity were deemed to be acceptable by both the expert and target populations.

CONCLUSION

This instrument makes a significant contribution to the identification of patients with palliative care needs as it has been designed to differentiate between general and specialised palliative care needs. Moreover, diagnostic data is not fundamental to the use of the instrument, thus facilitating its use by healthcare professionals other than physicians, in particular nurses. Internal and criterion validity assessments are ongoing and essential before wider dissemination of the instrument.

Improving palliative care with deep learning

BACKGROUND

Access to palliative care is a key quality metric which most healthcare organizations strive to improve. The primary challenges to increasing palliative care access are a combination of physicians over-estimating patient prognoses, and a shortage of palliative staff in general. This, in combination with treatment inertia can result in a mismatch between patient wishes, and their actual care towards the end of life.

METHODS

In this work, we address this problem, with Institutional Review Board approval, using machine learning and Electronic Health Record (EHR) data of patients. We train a Deep Neural Network model on the EHR data of patients from previous years, to predict mortality of patients within the next 3-12 month period. This prediction is used as a proxy decision for identifying patients who could benefit from palliative care.

RESULTS

The EHR data of all admitted patients are evaluated every night by this algorithm, and the palliative care team is automatically notified of the list of patients with a positive prediction. In addition, we present a novel technique for decision interpretation, using which we provide explanations for the model's predictions.

CONCLUSION

The automatic screening and notification saves the palliative care team the burden of time consuming chart reviews of all patients, and allows them to take a proactive approach in reaching out to such patients rather then relying on referrals from the treating physicians.

Palliative care needs in hospitalized cancer patients: a 5-year follow-up study

PURPOSE

The aims of this study were to describe and compare diagnoses, symptoms, and care needs in palliative cancer patients in two medium-sized hospitals in a county council with no specialized palliative care available 24/7; to analyze the relationships between diagnosis and symptoms/care needs; and to compare results and trends from two datasets (from 2007 and 2012).

METHODS

The study was population-based with a cross-sectional design and was conducted at two acute care hospitals. We performed 142 one-day inventories (n = 2972) in 2007 and 139 in 2012 (n = 2843) to register symptoms, care needs, and diagnosis based on a questionnaire. Multiple logistic regression models were used in the analysis.

RESULTS

During 2007 and 2012 combined, 10% (n = 589) of hospitalized patients were assessed as having cancer in a palliative phase. Prostate (12%) and colorectal (12%) cancers were most common. Pain (42%) and deterioration (42%) were the most prevalent symptoms and were associated with pancreas cancer in our regression models (p = 0.003 and p = 0.019, respectively). Other cancers had different associations: hematologic malignancies were associated with infections and blood transfusions (p < 0.001), breast cancer with pleurocentesis (p = 0.002), and stomach/esophagus cancer with nausea (p < 0.001). Nausea was more common in women than in men (p < 0.01). The mean number of symptoms/care needs was 2.9; patients with stomach/esophagus cancer had the highest number of symptoms/care needs (3.5).

CONCLUSIONS

Acute care hospitals still play an important role for patients requiring palliative care. Symptoms and care needs were not strongly associated with specific diagnoses. Therefore, symptoms, rather than the specific cancer diagnoses, should be the focus of care.