Retrospective Analysis of the Integration of Palliative Care Into the Care of Stroke Patients Admitted to a Regional Stroke Center.

Background: Palliative care (PC) aims to enhance the quality of life for patients when confronted with serious illness. As stroke inflicts high morbidity and mortality, the integration of PC within acute stroke care remains an important aspect of quality inpatient care. However, there is a tendency to offer PC to stroke patients only when death appears imminent. We aim to understand why this may be by examining stroke patients admitted to a regional stroke centre who subsequently died and their provision of PC. Methods: We conducted a retrospective single-centre cohort study of patients who died during admission to the regional stroke centre at Sunnybrook Health Sciences Centre (SHSC) in Toronto, Ontario, Canada. Baseline demographics were assessed using means, standard deviations (SD), medians, interquartile ranges (IQR), and proportions. Descriptive statistics, univariate, and multivariate analyses were performed to ascertain relationships between collected variables. Results: Univariate modeling demonstrated that older age, being female, no stroke diagnosis at admission to hospital, ischemic stroke, and comorbidities of cancer or dementia were associated with a higher incidence of palliative medicine consultation (PMC), while admission from an acute care hospital and a Glasgow Coma Scale (GCS) coma classification were associated with a lower incidence of PMC. The multivariate model identified the GCS coma-related category as the only significant factor associated with a higher incidence of death but was non-significantly related to a lower incidence of PMC. Conclusion: These results highlight continued missed opportunities for PC in stroke patients and underscore the need to better optimize PMC.

Evaluating the Effectiveness of the SPIKES Model to Break Bad News - A Systematic Review.

Introduction: Breaking bad news to patients and families can be challenging for healthcare providers. The present study conducted a systematic review of the literature to determine if formal communication training using the SPIKES protocol improves learner satisfaction, knowledge, performance, or system outcomes. Method: MEDLINE, Embase, CINAHL Plus (Nursing & Allied Health Sciences), and PsycINFO Databases were searched with keywords BAD NEWS and SPIKES. Studies were required to have an intervention using the SPIKES model and an outcome that addressed at least one of the four domains of the Kirkpatrick model for evaluating training effectiveness. The Cochrane Risk of Bias Tool was used to conduct a risk of bias assessment. Due to heterogeneity in the interventions and outcomes, meta-analysis was not undertaken and instead, a narrative synthesis was used with the information provided in the tables to summarise the main findings of the included studies. Results: Of 622 studies screened, 37 publications met the inclusion criteria. Interventions ranged from the use of didactic lecture, role play with standardised patients (SPs), video use, debriefing sessions, and computer simulations. Evaluation tools ranged from pre and post intervention questionnaires, OSCE performance with rating by independent raters and SPs, and reflective essay writing. Conclusions: Our systematic review demonstrated that the SPIKES protocol is associated with improved learner satisfaction, knowledge and performance. None of the studies in our review examined system outcomes. As such, further educational development and research is needed to evaluate the impact of patient outcomes, including the optimal components and length of intervention.

Attitudes of Canadian stroke physicians regarding palliative care for patients with acute severe stroke: A national survey.

BACKGROUND: Palliative care (PC) aims to enhance the quality of life for patients and their families when confronted with serious illness.  As stroke continues to inflict high morbidity and mortality, the integration of palliative care within acute stroke care remains an important aspect of quality inpatient care. AIM: This study aims to investigate the experiences and perceived barriers of PC integration for patients with acute severe stroke in Canadian stroke physicians. METHODS: We conducted an anonymous, descriptive, cross-sectional web-based self-administered survey of stroke physicians in Canada who engage in acute severe stroke care. The questionnaire contained three sections related to stroke physician characteristics, practice attributes, and opinions about palliative care.  Descriptive statistics, univariate, and regression analysis were performed to ascertain relations between collected variables. RESULTS: Of the 132 physician associate members, 120 were surveyed with a response rate of 69 (58%). Stroke physicians reported that PC services were consulted "sometimes" and that PC services were consulted rarely for prognostication and more often for end-of-life care which they agreed was better delivered off the stroke unit. Several barriers for early integration of palliative care services were identified including uncertainty in prognosis. Stroke physicians endorsed education of both families and physicians would be beneficial. CONCLUSIONS: There remain perceived barriers for integration of palliative care within the acute stroke population. Challenges include consultation of PC services, uncertainty around patient prognosis, engagement, and educational barriers. There are opportunities for further integration and collaboration between palliative care physicians and stroke physicians.

My first time.

A medical student's recount and personal reflection of her first patient death.

Machine learning vs. conventional statistical models for predicting heart failure readmission and mortality.

AIMS: This study aimed to review the performance of machine learning (ML) methods compared with conventional statistical models (CSMs) for predicting readmission and mortality in patients with heart failure (HF) and to present an approach to formally evaluate the quality of studies using ML algorithms for prediction modelling. METHODS AND RESULTS: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we performed a systematic literature search using MEDLINE, EPUB, Cochrane CENTRAL, EMBASE, INSPEC, ACM Library, and Web of Science. Eligible studies included primary research articles published between January 2000 and July 2020 comparing ML and CSMs in mortality and readmission prognosis of initially hospitalized HF patients. Data were extracted and analysed by two independent reviewers. A modified CHARMS checklist was developed in consultation with ML and biostatistics experts for quality assessment and was utilized to evaluate studies for risk of bias. Of 4322 articles identified and screened by two independent reviewers, 172 were deemed eligible for a full-text review. The final set comprised 20 articles and 686 842 patients. ML methods included random forests (n = 11), decision trees (n = 5), regression trees (n = 3), support vector machines (n = 9), neural networks (n = 12), and Bayesian techniques (n = 3). CSMs included logistic regression (n = 16), Cox regression (n = 3), or Poisson regression (n = 3). In 15 studies, readmission was examined at multiple time points ranging from 30 to 180 day readmission, with the majority of studies (n = 12) presenting prediction models for 30 day readmission outcomes. Of a total of 21 time-point comparisons, ML-derived c-indices were higher than CSM-derived c-indices in 16 of the 21 comparisons. In seven studies, mortality was examined at 9 time points ranging from in-hospital mortality to 1 year survival; of these nine, seven reported higher c-indices using ML. Two of these seven studies reported survival analyses utilizing random survival forests in their ML prediction models. Both reported higher c-indices when using ML compared with CSMs. A limitation of studies using ML techniques was that the majority were not externally validated, and calibration was rarely assessed. In the only study that was externally validated in a separate dataset, ML was superior to CSMs (c-indices 0.913 vs. 0.835). CONCLUSIONS: ML algorithms had better discrimination than CSMs in most studies aiming to predict risk of readmission and mortality in HF patients. Based on our review, there is a need for external validation of ML-based studies of prediction modelling. We suggest that ML-based studies should also be evaluated using clinical quality standards for prognosis research. Registration: PROSPERO CRD42020134867.