Loss-of-function of ALDH3B2 transdifferentiates human pancreatic duct cells into beta-like cells.

Replenishment of pancreatic beta cells is a key to the cure for diabetes. Beta cells regeneration is achieved predominantly by self-replication especially in rodents, but it was also shown that pancreatic duct cells can transdifferentiate into beta cells. How pancreatic duct cells undergo transdifferentiated and whether we could manipulate the transdifferentiation to replenish beta cell mass is not well understood. Using a genome-wide CRISPR screen, we discovered that loss-of-function of ALDH3B2 is sufficient to transdifferentiate human pancreatic duct cells into functional beta-like cells. The transdifferentiated cells have significant increase in beta cell marker genes expression, secrete insulin in response to glucose, and reduce blood glucose when transplanted into diabetic mice. Our study identifies a novel gene that could potentially be targeted in human pancreatic duct cells to replenish beta cell mass for diabetes therapy.

Resident Impact of the Single Site Order Restricting Staff Mobility across Long-Term Care Homes in British Columbia, Canada.

The Single Site Order (SSO)-a policy restricting staff from working at multiple long-term care (LTC) homes-was mandated by the Public Health Agency of Canada to control the spread of COVID-19 in LTC homes, where nearly 70% of COVID-19-related deaths in Canada occurred. This mixed methods study assesses the impact of the SSO on LTC residents in British Columbia. Interviews were conducted (residents (n = 6), family members (n = 9), staff (n = 18), and leadership (n = 10) from long-term care homes (n = 4)) and analyzed using thematic analysis. Administrative data were collected between April 2019 and March 2020 and between April 2020 and March 2021 and analyzed using descriptive statistics and data visualization. Qualitative and quantitative data were triangulated and demonstrated that staffing challenges became worse during the implementation of the SSO, resulting in the mental and physical health deterioration of LTC residents. Qualitative data demonstrated decreased time for personalized and proactive care, increased communication challenges, and increased loneliness and isolation. Quantitative data showed a decline in activities of daily living, increased antipsychotic medication use, pressure ulcers, behavioural symptoms, and an increase in falls. Addressing staff workload and staffing shortages during SSO-related policy implementation is essential to avoid resident health deterioration.

A Mixed Methods Study to Implement the Synergy Tool and Evaluate Its Impact on Long-Term Care Residents.

BACKGROUND: There are ongoing workforce challenges with the delivery of long-term care (LTC), such as staffing decisions based on arbitrary standards. The Synergy tool, a resident-centered approach to staffing, provides objective, real-time acuity and dependency scores (Synergy scores) for residents. The purpose of this study was to implement and evaluate the impact of the Synergy tool on LTC delivery. METHODS: A longitudinal mixed methods study took place within two publicly funded LTC homes in British Columbia, Canada. Quantitative data included weekly Synergy scores for residents (24 weeks), monthly aggregated resident falls data (18 months) and a six-month economic evaluation. Qualitative data were gathered from family caregivers and thematically analyzed. RESULTS: Quantitative findings from Synergy scores revealed considerable variability for resident acuity/dependency needs within and across units; and falls decreased during implementation. The six-month economic evaluation demonstrated some cost savings by comparing Synergy tool training and implementation costs with savings from resident fall rate reductions. Qualitative analyses yielded three positive impact themes (improved care delivery, better communication, and improved resident-family-staff relationships), and two negative structural themes (language barrier and staff shortages). CONCLUSIONS: The Synergy tool provides useful data for enhancing a 'fit' between resident needs and available staff.

Impact of the single site order in LTC: exacerbation of an overburdened system.

BACKGROUND: The long-term care (LTC) sector has been at the epicentre of COVID-19 in Canada. This study aimed to understand the impact that the Single Site Order (SSO) had on staff and leadership in four LTC homes in the Lower Mainland of British Columbia, Canada. METHODS: A mixed method study was conducted by analyzing administrative staffing data. Overtime, turnover, and job vacancy data were extracted and analyzed from four quarters before (April 2019 - March 2020) and four quarters during the pandemic (April 2020 - March 2021) using scatterplots and two-part linear trendlines across total direct care nursing staff and by designation (i.e., registered nurses (RNs), licenced practical nurses (LPNs) and care aids (CAs)). Virtual interviews were conducted with a purposive sample of leadership (10) and staff (18) from each of the four partner care homes (n = 28). Transcripts were analyzed in NVivo 12 using thematic analysis. RESULTS: Quantitative data indicated that the total overtime rate increased from before to during the pandemic, with RNs demonstrating the steepest rate increase. Additionally, while rates of voluntary turnover showed an upward trend before the pandemic for all direct care nursing staff, the rate for LPNs and, most drastically, for RNs was higher during the pandemic, while this rate decreased for CAs. Qualitative analysis identified two main themes and sub-themes: (1) overtime (loss of staff, mental health, and sick leave) and (2) staff turnover (the need to train new staff, and gender/race) as the most notable impacts associated with the SSO. CONCLUSIONS: The results of this study indicate that the outcomes due to COVID-19 and the SSO are not equal across nursing designations, with the RN shortage in the LTC sector highly evident. Quantitative and qualitative data underscore the substantial impact the pandemic and associated policies have on the LTC sector, namely, that staff are over-worked and care homes are understaffed.

Effects of Robotics Education on Young Children's Cognitive Development: a Pilot Study with Eye-Tracking.

The emerging field of robotics education (RE) is a new and rapidly growing subject area worldwide. It may provide a playful and novel learning environment for children to engage with all aspects of science, technology, engineering, and mathematics (STEM) learning. The purpose of this research is to examine how robotics learning activities may affect the cognitive abilities and cognitive processes of 6-8 years old children. The study adopted the mixed methods approach with a repeated measures design; three waves of data collection over 6 months, including quantitative data obtained from cognitive assessments and eye-tracking, and qualitative data from the interviews. A total of 31 children were recruited from an afterschool robotics program. To the best of our knowledge, this study is the first RE research that used a combination of eye-tracking, cognitive assessments, and interviews for examining the effect of RE on children. Using linear growth models, the results of cognitive assessments showed that children's visuospatial working memory as well as logical and abstract reasoning skills improved over time. The interview data were analyzed by a thematic analysis. The results revealed that children perceived RE activities as game play, which made children more engaged in their study; parents found their children to be more focused on activities comparing to six months ago. Additionally, the visualization of the eye-tracking data suggested that children became more focused on RE activities and got faster to process the information across six months in general, which echoed the findings in assessments and interviews. Our findings may help educators and policymakers better understand the benefits of RE for young children.

Priority Nursing Populations for Mental Health Support Before and During COVID-19: A Survey Study of Individual and Workplace Characteristics.

BACKGROUND: Nursing is a high-risk profession and nurses' exposure to workplace risk factors such as heavy workloads and inadequate staffing is well documented. The COVID-19 pandemic has exacerbated nurses' exposure to workplace risk factors, further deteriorating their mental health. Therefore, it is both timely and important to determine nursing groups in greatest need of mental health interventions and supports. PURPOSE: The purpose of this study is to provide a granular examination of the differences in nurse mental health across nurse demographic and workplace characteristics before and after COVID-19 was declared a pandemic. METHODS: This secondary analysis used survey data from two cross-sectional studies with samples (Time 1 study, 5,512 nurses; Time 2, 4,523) recruited from the nursing membership (∼48,000) of the British Columbia nurses' union. Data was analyzed at each timepoint using descriptive statistics and ordinal logistic regression. RESULTS: Several demographic and workplace characteristics were found to predict significant differences in the number of positive screenings on measures of poor mental health. Most importantly, in both survey times younger age was a strong predictor of worse mental health, as was full-time employment. Nurse workplace health authority was also a significant predictor of worse mental health. CONCLUSIONS: Structural and psychological strategies must be in place, proactively and preventively, to buffer nurses against workplace challenges that are likely to increase during the COVID-19 crisis.

Learning From Synthetic CT Images via Test-Time Training for Liver Tumor Segmentation.

Automatic liver tumor segmentation could offer assistance to radiologists in liver tumor diagnosis, and its performance has been significantly improved by recent deep learning based methods. These methods rely on large-scale well-annotated training datasets, but collecting such datasets is time-consuming and labor-intensive, which could hinder their performance in practical situations. Learning from synthetic data is an encouraging solution to address this problem. In our task, synthetic tumors can be injected to healthy images to form training pairs. However, directly applying the model trained using the synthetic tumor images on real test images performs poorly due to the domain shift problem. In this paper, we propose a novel approach, namely Synthetic-to-Real Test-Time Training (SR-TTT), to reduce the domain gap between synthetic training images and real test images. Specifically, we add a self-supervised auxiliary task, i.e., two-step reconstruction, which takes the output of the main segmentation task as its input to build an explicit connection between these two tasks. Moreover, we design a scheduled mixture strategy to avoid error accumulation and bias explosion in the training process. During test time, we adapt the segmentation model to each test image with self-supervision from the auxiliary task so as to improve the inference performance. The proposed method is extensively evaluated on two public datasets for liver tumor segmentation. The experimental results demonstrate that our proposed SR-TTT can effectively mitigate the synthetic-to-real domain shift problem in the liver tumor segmentation task, and is superior to existing state-of-the-art approaches.

Weakly Supervised Liver Tumor Segmentation Using Couinaud Segment Annotation.

Automatic liver tumor segmentation is of great importance for assisting doctors in liver cancer diagnosis and treatment planning. Recently, deep learning approaches trained with pixel-level annotations have contributed many breakthroughs in image segmentation. However, acquiring such accurate dense annotations is time-consuming and labor-intensive, which limits the performance of deep neural networks for medical image segmentation. We note that Couinaud segment is widely used by radiologists when recording liver cancer-related findings in the reports, since it is well-suited for describing the localization of tumors. In this paper, we propose a novel approach to train convolutional networks for liver tumor segmentation using Couinaud segment annotations. Couinaud segment annotations are image-level labels with values ranging from 1 to 8, indicating a specific region of the liver. Our proposed model, namely CouinaudNet, can estimate pseudo tumor masks from the Couinaud segment annotations as pixel-wise supervision for training a fully supervised tumor segmentation model, and it is composed of two components: 1) an inpainting network with Couinaud segment masks which can effectively remove tumors for pathological images by filling the tumor regions with plausible healthy-looking intensities; 2) a difference spotting network for segmenting the tumors, which is trained with healthy-pathological pairs generated by an effective tumor synthesis strategy. The proposed method is extensively evaluated on two liver tumor segmentation datasets. The experimental results demonstrate that our method can achieve competitive performance compared to the fully supervised counterpart and the state-of-the-art methods while requiring significantly less annotation effort.

Implementation of the Synergy Tool: A Potential Intervention to Relieve Health Care Worker Burnout.

(1) Background: Healthcare workers experienced rising burnout rates during and after the COVID-19 pandemic. A practice-academic collaboration between health services researchers and the surgical services program of a Canadian tertiary-care urban hospital was used to develop, implement and evaluate a potential burnout intervention, the Synergy tool. (2) Methods: Using participatory action research methods, this project involved four key phases: (I) an environmental scan and a baseline survey assessment, (II), a workshop, (III) Synergy tool implementation and (IV) a staffing plan workshop. A follow-up survey to evaluate the impact of Synergy tool use on healthcare worker burnout will be completed in 2023. (3) Results: A baseline survey assessment indicated high to severe levels of personal and work-related burnout prior to project initiation. During the project phases, there was high staff engagement with Synergy tool use to create patient care needs profiles and staffing recommendations. (4) Conclusions: As in previous research with the Synergy tool, this patient needs assessment approach is an efficient and effective way to engage direct care providers in identifying and scoring acuity and dependency needs for their specific patient populations. The Synergy tool approach to assessing patient needs holds promise as a means to engage direct care providers and to give them greater control over their practice-potentially serving as a buffer against burnout.

Investigation of 18F and 89Zr Isotopes Self-Absorption and Dose Rate Parameters for PET Imaging.

This work concerns study of self-absorption factor (SAF) and dose rate constants of zirconium-89 (89Zr) for the purpose of radiation protection in positron emission tomography (PET) and to compare them with those of 18F-deoxyglucose (18F-FDG). We analyzed the emitted energy spectra by 18F and 89Zr through anthropomorphic phantom and calculated the absorbed energy using Monte Carlo method. The dose rate constants for both radionuclides were estimated with 2 different fluence-to-effective dose conversion coefficients. Our estimated SAF value of 0.65 for 18F agreed with the recommendation of the American Association of Physicists in Medicine (AAPM). The SAF for 89Zr was in the range of 0.61-0.66 depending on the biodistribution. Using the fluence-to-effective dose conversion coefficients recommended jointly by the American National Standards Institute and the American Nuclear Society (ANSI/ANS), the dose rate at 1 m from the patient for 18F was 0.143 µSv·MBq-1·hr-1, which is consistent with the AAPM recommendation, while that for 89Zr was 0.154 µSv·MBq-1·hr-1. With the conversion coefficients currently recommended by the International Committee on Radiological Protection (ICRP), the dose rate estimates were lowered by 2.8% and 2.6% for 89Zr and 18F, respectively. Also, we observed that the AAPM derived dose is an overestimation near the patient, compared to our simulations, which can be explained by the biodistribution nature and the assumption of the point source. Thus, we proposed new radiation protection factors for 89Zr radionuclide.

Describing the Mental Health State of Nurses in British Columbia: A Province-Wide Survey Study.

A cross-sectional province-wide survey study of 3,978 British Columbia (BC) nurses was conducted to explore the mental health state of the nursing workforce in BC. About one third of nurses reported depression and anxiety; about half reported symptoms of post-traumatic stress disorder and at least one third reported high levels of one or more dimensions of burnout. Mental health problems were about 1.5 to 3 times more prevalent among BC nurses compared to their peers nationally. Improving nurses' mental health requires multi-factorial and multi-level efforts. Evidence-based and workplace-specific policies and interventions that better support nurses at risk are recommended.

Artificial intelligence in prediction of non-alcoholic fatty liver disease and fibrosis.

Artificial intelligence (AI) has become increasingly widespread in our daily lives, including healthcare applications. AI has brought many new insights into better ways we care for our patients with chronic liver disease, including non-alcoholic fatty liver disease and liver fibrosis. There are multiple ways to apply the AI technology on top of the conventional invasive (liver biopsy) and noninvasive (transient elastography, serum biomarkers, or clinical prediction models) approaches. In this review article, we discuss the principles of applying AI on electronic health records, liver biopsy, and liver images. A few common AI approaches include logistic regression, decision tree, random forest, and XGBoost for data at a single time stamp, recurrent neural networks for sequential data, and deep neural networks for histology and images.

Importance-aware personalized learning for early risk prediction using static and dynamic health data.

OBJECTIVE: Accurate risk prediction is important for evaluating early medical treatment effects and improving health care quality. Existing methods are usually designed for dynamic medical data, which require long-term observations. Meanwhile, important personalized static information is ignored due to the underlying uncertainty and unquantifiable ambiguity. It is urgent to develop an early risk prediction method that can adaptively integrate both static and dynamic health data. MATERIALS AND METHODS: Data were from 6367 patients with Peptic Ulcer Bleeding between 2007 and 2016. This article develops a novel End-to-end Importance-Aware Personalized Deep Learning Approach (eiPDLA) to achieve accurate early clinical risk prediction. Specifically, eiPDLA introduces a long short-term memory with temporal attention to learn sequential dependencies from time-stamped records and simultaneously incorporating a residual network with correlation attention to capture their influencing relationship with static medical data. Furthermore, a new multi-residual multi-scale network with the importance-aware mechanism is designed to adaptively fuse the learned multisource features, automatically assigning larger weights to important features while weakening the influence of less important features. RESULTS: Extensive experimental results on a real-world dataset illustrate that our method significantly outperforms the state-of-the-arts for early risk prediction under various settings (eg, achieving an AUC score of 0.944 at 1 year ahead of risk prediction). Case studies indicate that the achieved prediction results are highly interpretable. CONCLUSION: These results reflect the importance of combining static and dynamic health data, mining their influencing relationship, and incorporating the importance-aware mechanism to automatically identify important features. The achieved accurate early risk prediction results save precious time for doctors to timely design effective treatments and improve clinical outcomes.

Nurses' Workplace Conditions Impacting Their Mental Health during COVID-19: A Cross-Sectional Survey Study.

Among health workers, nurses are at the greatest risk of COVID-19 exposure and mortality due to their workplace conditions, including shortages of personal protective equipment (PPE), insufficient staffing, and inadequate safety precautions. The purpose of this study was to examine the impact of COVID-19 workplace conditions on nurses' mental health outcomes. A cross-sectional correlational design was used. An electronic survey was emailed to nurses in one Canadian province between June and July of 2020. A total of 3676 responses were included in this study. We found concerning prevalence rates for post-traumatic stress disorder (47%), anxiety (38%), depression (41%), and high emotional exhaustion (60%). Negative ratings of workplace relations, organizational support, organizational preparedness, workplace safety, and access to supplies and resources were associated with higher scores on all of the adverse mental health outcomes included in this study. Better workplace policies and practices are urgently required to prevent and mitigate nurses' suboptimal work conditions, given their concerning mental health self-reports during the COVID-19 pandemic.

Explainable Uncertainty-Aware Convolutional Recurrent Neural Network for Irregular Medical Time Series.

Influenced by the dynamic changes in the severity of illness, patients usually take examinations in hospitals irregularly, producing a large volume of irregular medical time-series data. Performing diagnosis prediction from the irregular medical time series is challenging because the intervals between consecutive records significantly vary along time. Existing methods often handle this problem by generating regular time series from the irregular medical records without considering the uncertainty in the generated data, induced by the varying intervals. Thus, a novel Uncertainty-Aware Convolutional Recurrent Neural Network (UA-CRNN) is proposed in this article, which introduces the uncertainty information in the generated data to boost the risk prediction. To tackle the complex medical time series with subseries of different frequencies, the uncertainty information is further incorporated into the subseries level rather than the whole sequence to seamlessly adjust different time intervals. Specifically, a hierarchical uncertainty-aware decomposition layer (UADL) is designed to adaptively decompose time series into different subseries and assign them proper weights in accordance with their reliabilities. Meanwhile, an Explainable UA-CRNN (eUA-CRNN) is proposed to exploit filters with different passbands to ensure the unity of components in each subseries and the diversity of components in different subseries. Furthermore, eUA-CRNN incorporates with an uncertainty-aware attention module to learn attention weights from the uncertainty information, providing the explainable prediction results. The extensive experimental results on three real-world medical data sets illustrate the superiority of the proposed method compared with the state-of-the-art methods.

Temporal matrix completion with locally linear latent factors for medical applications.

Regular medical records are useful for medical practitioners to analyze and monitor patient's health status especially for those with chronic disease. However, such records are usually incomplete due to unpunctuality and absence of patients. In order to resolve the missing data problem over time, tensor-based models have been developed for missing data imputation in recent papers. This approach makes use of the low-rank tensor assumption for highly correlated data in a short-time interval. Nevertheless, when the time intervals are long, data correlation may not be high between consecutive time stamps so that such assumption is not valid. To address this problem, we propose to decompose matrices with missing data over time into their latent factors. Then, the locally linear constraint is imposed on the latent factors for temporal matrix completion. By using three publicly available medical datasets and two medical datasets collected from Prince of Wales Hospital in Hong Kong, experimental results show that the proposed algorithm achieves the best performance compared with state-of-the-art methods.

Workplace Violence among British Columbia Nurses Across Different Roles and Contexts.

Workplace violence in healthcare settings is on the rise, particularly against nurses. Most healthcare violence research is in acute care settings. The purpose of this paper is to present descriptive findings on the prevalence of types and sources of workplace violence among nurses in different roles (i.e., direct care, leader, educator), specialties, care sectors (i.e., acute, community, long-term care) and geographic contexts (i.e., urban, suburban, rural) within the province of British Columbia (BC), Canada. This is a province-wide survey study using a cross-sectional descriptive, correlational design. An electronic survey was emailed by the provincial union to members across the province in Fall 2019. A total of 4462 responses were analyzed using descriptive and chi-square statistics. The most common types of workplace violence were emotional abuse, threats of assault and physical assault for all nursing roles and contexts. Findings were similar to previous BC research from two decades ago except for two to ten times higher proportions of all types of violence, including verbal and physical sexual assault. Patients were the most common source of violence towards nurses. Nurses should be involved in developing workplace violence interventions that are tailored to work environment contexts and populations.

Machine learning model to predict recurrent ulcer bleeding in patients with history of idiopathic gastroduodenal ulcer bleeding.

BACKGROUND: Patients with a history of Helicobacter pylori-negative idiopathic bleeding ulcers have an increased risk of recurring ulcer complications. AIM: To build a machine learning model to identify patients at high risk for recurrent ulcer bleeding. METHODS: Data from a retrospective cohort of 22 854 patients (training cohort) diagnosed with peptic ulcer disease in 2007-2016 were analysed to build a model (IPU-ML) to predict recurrent ulcer bleeding. We tested the IPU-ML in all patients with a diagnosis of gastrointestinal bleeding (n = 1265) in 2008-2015 from a different catchment population (independent validation cohort). Any co-morbid conditions which had occurred in >1% of study population were eligible as predictors. RESULTS: Recurrent ulcer bleeding developed in 4772 patients (19.5%) in the training cohort, during a median follow-up period of 2.7 years. IPU-ML model built on six parameters (age, baseline haemoglobin, and presence of gastric ulcer, gastrointestinal diseases, malignancies, and infections) identified patients with bleeding recurrence within 1 year with an area under the receiver operating characteristic curve (AUROC) of 0.648. When we set the IPU-ML cutoff value at 0.20, 27.5% of patients were classified as high risk for rebleeding with a sensitivity of 41.4%, specificity of 74.6%, and a negative predictive value of 91.1%. In the validation cohort, the IPU-ML identified patients with a recurrence ulcer bleeding within 1 year with an AUROC of 0.775, and 84.3% of overall accuracy. CONCLUSION: We developed a machine-learning model to identify those patients with a history of idiopathic gastroduodenal ulcer bleeding who are not at high risk for recurrent ulcer bleeding.

Dynamic Graph Co-Matching for Unsupervised Video-based Person Re-Identification.

Cross-camera label estimation from a set of unlabelled training data is an extremely important component in unsupervised person re-identification (re-ID) systems. With the estimated labels, existing advanced supervised learning methods can be leveraged to learn discriminative re-ID models. In this paper, we utilize the graph matching technique for accurate label estimation due to its advantages in optimal global matching and intra-camera relationship mining. However, the graph structure constructed with non-learnt similarity measurement cannot handle the large cross-camera variations, which leads to noisy and inaccurate label outputs. This paper designs a Dynamic Graph Matching (DGM) framework, which improves the label estimation process by iteratively refining the graph structure with better similarity measurement learnt from intermediate estimated labels. In addition, we design a positive re-weighting strategy to refine the intermediate labels, which enhances the robustness against inaccurate matching output and noisy initial training data. To fully utilize the abundant video information and reduce false matchings, a co-matching strategy is further incorporated into the framework. Comprehensive experiments conducted on three video benchmarks demonstrate that DGM outperforms state-of-the-art unsupervised re-ID methods and yields competitive performance to fully supervised upper bounds.

Deprescribing benzodiazepine receptor agonists: Evidence-based clinical practice guideline.

OBJECTIVE: To develop an evidence-based guideline to help clinicians make decisions about when and how to safely taper and stop benzodiazepine receptor agonists (BZRAs); to focus on the highest level of evidence available and seek input from primary care professionals in the guideline development, review, and endorsement processes. METHODS: The overall team comprised 8 clinicians (1 family physician, 2 psychiatrists, 1 clinical psychologist, 1 clinical pharmacologist, 2 clinical pharmacists, and 1 geriatrician) and a methodologist; members disclosed conflicts of interest. For guideline development, a systematic process was used, including the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach. Evidence was generated by conducting a systematic review of BZRA deprescribing trials for insomnia, as well as performing a review of reviews of the harms of continued BZRA use and narrative syntheses of patient preferences and resource implications. This evidence and GRADE quality of evidence ratings were used to generate recommendations. The team refined guideline content and recommendations through consensus and synthesized clinical considerations to address front-line clinician questions. The draft guideline was reviewed by clinicians and stakeholders. RECOMMENDATIONS: We recommend that deprescribing (tapering slowly) of BZRAs be offered to elderly adults (≥ 65 years) who take BZRAs, regardless of duration of use, and suggest that deprescribing (tapering slowly) be offered to adults aged 18 to 64 who have used BZRAs for more than 4 weeks. These recommendations apply to patients who use BZRAs to treat insomnia on its own (primary insomnia) or comorbid insomnia where potential underlying comorbidities are effectively managed. This guideline does not apply to those with other sleep disorders or untreated anxiety, depression, or other physical or mental health conditions that might be causing or aggravating insomnia. CONCLUSION: Benzodiazepine receptor agonists are associated with harms, and therapeutic effects might be short term. Tapering BZRAs improves cessation rates compared with usual care without serious harms. Patients might be more amenable to deprescribing conversations if they understand the rationale (potential for harm), are involved in developing the tapering plan, and are offered behavioural advice. This guideline provides recommendations for making decisions about when and how to reduce and stop BZRAs. Recommendations are meant to assist with, not dictate, decision making in conjunction with patients.