Explore the vast range of titles available.

A key task of emergency departments is to promptly identify patients who require hospital admission. Early identification ensures patient safety and aids organisational planning. Supervised machine learning algorithms can use data describing historical episodes to make ahead-of-time predictions of clinical outcomes. Despite this, clinical settings are dynamic environments and the underlying data distributions characterising episodes can change with time (data drift), and so can the relationship between episode characteristics and associated clinical outcomes (concept drift). Practically this means deployed algorithms must be monitored to ensure their safety. We demonstrate how explainable machine learning can be used to monitor data drift, using the COVID-19 pandemic as a severe example. We present a machine learning classifier trained using (pre-COVID-19) data, to identify patients at high risk of admission during an emergency department attendance. We then evaluate our model’s performance on attendances occurring pre-pandemic (AUROC of 0.856 with 95%CI [0.852, 0.859]) and during the COVID-19 pandemic (AUROC of 0.826 with 95%CI [0.814, 0.837]). We demonstrate two benefits of explainable machine learning (SHAP) for models deployed in healthcare settings: (1) By tracking the variation in a feature’s SHAP value relative to its global importance, a complimentary measure of data drift is found which highlights the need to retrain a predictive model. (2) By observing the relative changes in feature importance emergent health risks can be identified.

Early identification of patients who require onward referral to social care can prevent delays to discharge from hospital. We introduce an explainable machine learning (ML) model to identify potential social care needs at the first point of admission. This model was trained using routinely collected data on patient admissions, hospital spells and discharge at a large tertiary hospital in the UK between 2017 and 2023. The model performance (one-vs-rest AUROC = 0.915 [0.907 0.924] (95% confidence interval), is comparable to clinician’s predictions of discharge care needs, despite working with only a subset of the information available to the clinician. We find that ML and clinicians perform better for identifying different types of care needs, highlighting the added value of a potential system supporting decision making. We also demonstrate the ability for ML to provide automated initial discharge need assessments, in the instance where initial clinical assessment is delayed and provide reasoning for the decision. Finally, we demonstrate that combining clinician and machine predictions, in a hybrid model, provides even more accurate early predictions of onward social care requirements (OVR AUROC = 0.936 [0.928 0.943]) and demonstrates the potential for human-in-the-loop decision support systems in clinical practice.

ObjectiveTo investigate the association between initial discharge planning and transfers of inpatient care with discharge delay. To identify operational changes which could expedite discharge within the Discharge to Assess (D2A) model.DesignRetrospective cohort study.SettingUniversity Hospital Southampton National Health Service Foundation Trust (UHS).ParticipantsAll adults (≥18 years) who registered a hospital inpatient stay in UHS between 1 January 2021 and 31 December 2022 (n=258 051). After excluding inpatient stays without complete discharge planning data or key demographic/clinical information, 65 491 inpatient stays were included in the final analysis. Data included demographics, comorbidities, ward movements, care team handovers and discharge planning records.Primary and secondary outcome measuresThe primary outcome was discharge delay, defined as the number of days between the final estimated discharge date and the actual discharge date. For the purposes of OR analysis, discharge delay was modelled as a binary outcome: any delay (>0 days) versus no delay. Logistic regression models were used to examine associations between initial discharge planning accuracy, the number of ward moves and the number of in-specialty handovers and the likelihood of discharge delay, adjusting for demographic and patient complexity factors.ResultsOut of 65 491 inpatient stays, 10 619 had an initial planned discharge pathway that was different from the final discharge pathway. 7790 of these inpatient stays (75.1%) recorded a discharge delay. In contrast, among the 54 872 inpatient stays where the initial and final pathway matched, 10 216 (18.6%) recorded a delay. Using logistic regression modelling a binary outcome (any discharge delay vs no delay), an inaccurate initial pathway was associated with significantly increased odds of delay (adjusted OR (aOR) 2.72, 95% CI 2.55 to 2.91). Each additional ward move (aOR 1.25, 95% CI 1.23 to 1.28) and each in-specialty handover (aOR 1.17, 95% CI 1.14 to 1.20) were also associated with higher odds of discharge delay.ConclusionsThis study finds a strong association between inaccurate initial discharge plans and inpatient transfers of care with discharge delay, after controlling for patient complexity and acuity. This highlights the need to consider how initial plans and inpatient transfers affect discharge planning. Given the lead times for organising onward care, operational inefficiencies are most impactful for patients eventually discharged on pathways with higher planning complexity.

Self-reporting digital apps provide a way of remotely monitoring and managing patients with chronic conditions in the community. Leveraging the data collected by these apps in prognostic models could provide increased personalization of care and reduce the burden of care for people who live with chronic conditions. This study evaluated the predictive ability of prognostic models for the prediction of acute exacerbation events in people with chronic obstructive pulmonary disease by using data self-reported to a digital health app. The aim of this study was to evaluate if data self-reported to a digital health app can be used to predict acute exacerbation events in the near future. This is a retrospective study evaluating the use of symptom and chronic obstructive pulmonary disease assessment test data self-reported to a digital health app (myCOPD) in predicting acute exacerbation events. We include data from 2374 patients who made 68,139 self-reports. We evaluated the degree to which the different variables self-reported to the app are predictive of exacerbation events and developed both heuristic and machine learning models to predict whether the patient will report an exacerbation event within 3 days of self-reporting to the app. The model's predictive ability was evaluated based on self-reports from an independent set of patients. Users self-reported symptoms, and standard chronic obstructive pulmonary disease assessment tests displayed correlation with future exacerbation events. Both a baseline model (area under the receiver operating characteristic curve [AUROC] 0.655, 95% CI 0.689-0.676) and a machine learning model (AUROC 0.727, 95% CI 0.720-0.735) showed moderate ability in predicting exacerbation events, occurring within 3 days of a given self-report. Although the baseline model obtained a fixed sensitivity and specificity of 0.551 (95% CI 0.508-0.596) and 0.759 (95% CI 0.752-0.767) respectively, the sensitivity and specificity of the machine learning model can be tuned by dichotomizing the continuous predictions it provides with different thresholds. Data self-reported to health care apps designed to remotely monitor patients with chronic obstructive pulmonary disease can be used to predict acute exacerbation events with moderate performance. This could increase personalization of care by allowing preemptive action to be taken to mitigate the risk of future exacerbation events.

Background Multimorbidity, the presence of multiple chronic health conditions, presents significant challenges in both health and social care settings. Addressing social care needs, such as assistance with daily activities and support for managing finances, is crucial in care management patients with multimorbidity. However, variability in the documentation and reporting of these needs remains poorly understood. This study aimed to quantify the variations in social care need (SCN) reporting across GP practices in England. Methods We conducted a population-based study using electronic health records from a national sample of 873,092 individuals with multimorbidity. Inclusion and exclusion criteria were applied to determine the final cohort, with demographic and clinical data extracted. We analysed SCN reporting rates at the practice level, using interquartile ranges (IQRs) and intra-class coefficients (ICCs) to assess variability. Factors influencing SCN reporting were examined, including long-term conditions, demographic variables, and socio-economic deprivation. Results Significant variability was observed in SCN reporting across GP practices. Outcomes related to mobility and residential needs showed the greatest differences in reporting rates. Moderate correlations were observed between certain SCN categories, such as mobility and activities of daily living, as well as disability and financial needs. Patients with long-term conditions, such as dementia and multiple sclerosis, were more likely to have their SCNs reported, while other multimorbidity conditions showed lower reporting rates. Demographic factors, including gender and socio-economic deprivation, were associated with higher reporting rates, particularly for females and patients in more deprived areas. Conclusions This study highlights the significant variability in the documentation of social care needs across healthcare practices, using electronic health records in a large population-based sample. The findings emphasise the need for standardised reporting practices to ensure comprehensive care for individuals with multimorbidity, particularly those from more deprived socio-economic backgrounds and with complex care needs. Improved reporting could enhance care coordination and reduce health inequalities.

Each year thousands of children are diagnosed with autism, a devastating neurological disorder that profoundly affects a person’s language and social development. Saving Ben is the story of one family coping with autism, told from the viewpoint of a father struggling to understand his son’s strange behavior and rescue him from a downward spiral. “Take him home, love him, and save your money for his institutionalization when he turns twenty-one.” That was the best advice his doctor could offer in 1990 when three-year-old Ben was diagnosed with autism. Saving Ben tells the story of Ben’s regression as an infant into the world of autism and his journey toward recovery as a young adult. His father, Dan Burns, puts the reader in the passenger’s seat as he struggles with medical service providers, the school system, extended family, and his own limitations in his efforts to pull Ben out of his darkening world. Ben, now 21 years old, is a work in progress. The full force and fury of the autism storm have passed. Using new biomedical treatments, repair work is underway. Saving Ben is a story of Ben’s journey toward recovery, and a family’s story of loss, grief, and healing. “Keep the faith, never give up.” These are the lessons of the author’s miraculous journey, saving Ben.

For Sachs, this type of CPD recasts teachers as learners and has the potential to change practice by connecting teachers to what they are teaching. Reimagining practice for the benefit of both teachers and students may seem utopian, particularly in the current political and economic climate. The National Curriculum states 'pupils need to be able to move fluently between representations of mathematical ideas' (DfE, 2013). [...]I believe group work allows students to communicate their ideas in a lower risk, more inclusive environment, and allows more children to speak than when only directly questioned.