Background:Many patients show deterioration in functioning and increased care needs in the last year of life. End-of-life care needs and health care utilization might differ between groups of acutely hospitalized older patients.Aim:To investigate differences in geriatric conditions, advance care planning, and health care utilization in patients with cancer, organ failure, or frailty, who died within 1 year after acute hospitalization.Design:Prospective cohort study conducted between 2002 and 2008, with 1-year follow-up.Setting:University teaching hospital in the Netherlands.Participants:Aged ⩾65 years, acutely hospitalized for ⩾48 h, and died within 1 year after hospitalization. At admission, all patients received a systematic comprehensive geriatric assessment. Hospital records were searched for advance care planning information and health care utilization. Differences between patient groups were calculated.Results:In total, 306 patients died within 1 year after acute admission (35%) and were included; 151 with cancer, 98 with end-stage organ failure, and 57 frail older persons. At hospital admission, 72% of the frail group had delirium and/or severe pre-existing cognitive impairment. The frail and organ failure group had many pre-existing disabilities. Three months post-discharge, 75% of the frail and organ failure group had died, 45% of these patients had an advance care plan in their hospital records.Conclusion:Patients with frailty and organ failure had highest rates of geriatric conditions at hospital admission and often had missing information on advance care planning in the hospital records. There is a need to better identify end-of-life needs for these groups.
Background: A transitional care pathway (TCP) could improve care for older patients in the last months of life. However, barriers exist such as unidentified palliative care needs and suboptimal collaboration between care settings. The aim of this study was to determine the feasibility of a TCP, named PalliSupport, for older patients at the end of life, prior to a stepped-wedge randomized controlled trial. Methods: A mixed-method feasibility study was conducted at one hospital with affiliated primary care. Patients were ≥ 60 years and acutely hospitalized. The intervention consisted of (1) training on early identification of the palliative phase and end of life conversations, (2) involvement of a transitional palliative care team during admission and post-discharge and (3) intensified collaboration between care settings. Outcomes were feasibility of recruitment, data collection, patient burden and protocol adherence. Experiences of 14 professionals were assessed through qualitative interviews. Results: Only 16% of anticipated participants were included which resulted in difficulty assessing other feasibility criteria. The qualitative analysis identified misunderstandings about palliative care, uncertainty about professionals' roles and difficulties in initiating end of life conversations as barriers. The training program was well received and professionals found the intensified collaboration beneficial for patient care. The patients that participated experienced low burden and data collection on primary outcomes and protocol adherence seems feasible. Discussion: This study highlights the importance of performing a feasibility study prior to embarking on effectiveness studies. Moving forward, the PalliSupport care pathway will be adjusted to incorporate a more active recruitment approach, additional training on identification and palliative care, and further improvement on data collection.
Estimating the remaining useful life (RUL) of an asset lies at the heart of prognostics and health management (PHM) of many operations-critical industries such as aviation. Mod- ern methods of RUL estimation adopt techniques from deep learning (DL). However, most of these contemporary tech- niques deliver only single-point estimates for the RUL without reporting on the confidence of the prediction. This practice usually provides overly confident predictions that can have severe consequences in operational disruptions or even safety. To address this issue, we propose a technique for uncertainty quantification (UQ) based on Bayesian deep learning (BDL). The hyperparameters of the framework are tuned using a novel bi-objective Bayesian optimization method with objectives the predictive performance and predictive uncertainty. The method also integrates the data pre-processing steps into the hyperparameter optimization (HPO) stage, models the RUL as a Weibull distribution, and returns the survival curves of the monitored assets to allow informed decision-making. We vali- date this method on the widely used C-MAPSS dataset against a single-objective HPO baseline that aggregates the two ob- jectives through the harmonic mean (HM). We demonstrate the existence of trade-offs between the predictive performance and the predictive uncertainty and observe that the bi-objective HPO returns a larger number of hyperparameter configurations compared to the single-objective baseline. Furthermore, we see that with the proposed approach, it is possible to configure models for RUL estimation that exhibit better or comparable performance to the single-objective baseline when validated on the test sets.
