Farr Institute

This study integrates machine learning and wearable technology to detect nocturnal asthma attacks from daily questionnaire responses, patient information, and Mi-Band-3 smartwatch data. Four models-Logistic Regression, Naive Bayes, Random Forest, and XGBoost-were trained and tuned using grid search and five-fold cross-validation. XGBoost achieved the best performance on the test set, with an AUC of 0.87, an AUPRC of 0.71, and a 2.55-fold increase in precision. Models combining active- (patient-reported) and passive-(device-collected) monitoring features outperformed passive-monitoring only approaches, emphasizing the importance of multiple data sources. Key predictors included an indication of asthma trigger encounter, maximum expected PEF, age, obesity, and sleep quality metrics, revealing complex interactions. Although the Mi-Band-3 contributed valuable information, it could not fully replace active monitoring. Future work should incorporate a larger, more diverse participant pool, integrate additional asthma-related variables, and explore advanced time-series models to improve predictive accuracy and reduce patient burden.Clinical Relevance- Machine learning-driven detection modeling can serve as an early warning system for nocturnal asthma attacks, integrating diverse risk factors and wearable sensor data to improve accuracy. Such a tool could shift asthma management towards a more preventative, personalized approach-minimizing reliance on burdensome self-monitoring, enhancing patient quality of life, and enabling clinicians to intervene proactively when nocturnal symptoms are likely to worsen.

Abstract Interhospital transport of critically unwell children exacerbates physiological stress, increasing the risk of deterioration during transport. Due to the nature of illness and interventions occurring in this cohort, defining “normal” vital sign ranges is impossible, which can make identifying deterioration events difficult.A novel data-driven approach was developed to identify adverse respiratory and cardiovascular events in critically ill children during interhospital transport. In this retrospective cohort study of 1,519 transports (July 2016 to May 2021), vital signs were recorded at one-second intervals and then analysed using an adaptation of Bollinger Bands, a technique borrowed from financial market analysis. This method dynamically established each patient’s stable ranges for heart rate, blood pressure, oxygen saturation, and other respiratory parameters, and flagged adverse events when multiple parameters simultaneously fell outside their expected ranges.Adverse respiratory events were identified when oxygen saturation deviated below a dynamically defined threshold alongside at least one additional respiratory parameter. Cardiovascular events were defined by concurrent deviations in blood pressure and heart rate. Overall, 15.6 percent of transports had one or more adverse respiratory events, and 21.5 percent had at least one adverse cardiovascular event.To validate these labels, the number of adverse events and the cumulative duration of vital sign instability during transport were compared against clinical markers of deterioration. Each additional respiratory event was associated with increased odds of receiving respiratory support during transport and higher 30-day mortality, while each additional cardiovascular event was associated with increased odds of receiving vasoactive support during transport.Our method detects clinically meaningful respiratory and cardiovascular adverse events during transport. The approach is readily adaptable to other high-resolution intensive care datasets, for both retrospective labelling as well as automated, real-time identification of adverse events in the clinical setting, offering a foundation for improved monitoring and early intervention in critically ill patients.Author SummaryTransporting critically ill children between hospitals is challenging because their condition can worsen during the journey. Previously, studies have relied on fixed vital-sign cutoffs to identify “adverse events,” but these rigid thresholds may not account for individual differences in heart rate, blood pressure, and oxygen baseline levels.Here, we used a tool borrowed from financial market analysis to track minute-by-minute vital signs in over 1,500 transport episodes. Our approach identified sudden, patient-specific changes that signalled respiratory or cardiovascular problems. We found that around 15 percent of transfers had at least one breathing-related event and over 20 percent had a heart-related event. Moreover, these events were linked to worse outcomes, such as the need for extra breathing or blood-pressure support.By considering each child’s own “stable” vital sign range, this method can detect meaningful changes that standard fixed thresholds might miss. This personalised tracking of vital signs opens the door for further research on routinely collected vital-sign datasets. It could also be deployed in the clinical environment helping clinicians detect a child’s deterioration earlier and respond in real time. Although we focused on children in transport, our approach could be adapted to other intensive care settings to improve patient monitoring.

Background Most antibiotic prescribing occurs in primary care, largely in children under 5 years old, and often inappropriately. This study investigated knowledge, attitudes and behaviours (KABs) towards common childhood infections, antibiotic use and antimicrobial resistance (AMR), among parents of children under 5 years old. The concept of individual sacrifice (forgoing antibiotics—a selective pressure for AMR) to mitigate future societal risk of AMR and how the COVID-19 pandemic shaped views were explored. Methods This qualitative study included three, one-hour, virtual focus groups with mothers from parenting networks across inner-city London and semi-rural England, held mid-pandemic (2020). All had ≥1 child <5 years old. The Framework Method of analysis was used. Parents’ KABs towards antibiotic use/AMR formed the primary outcome, with emphases on their sense of personal agency towards mitigating the threat of AMR for society, plus how the pandemic influenced views on infection prevention and care. Results Fourteen mothers (groups of six, four, four) participated, with mixed ethnicities, education and employment status. Parent perceptions of their individual child’s immediate need for antibiotics outweighed concerns for any possible future threat of AMR to society. Four key themes were identified: uncertainty around symptoms; impact of socio-cultural background on KAB; poor understanding of how antibiotics/AMR work; and opportunities within the doctor–patient dialogue to shape mindset around AMR. The pandemic influenced views across themes. Conclusion Parents prioritising their child’s perceived, immediate, individual ‘need’ for antibiotics over any future impact of AMR on society highlights a continuing need to engage parents in how to mitigate AMR through appropriate antibiotic use, reducing threat to both their child and others. Framing point-of-care dialogue around antibiotic use/AMR in the present (versus future), drawing on pandemic insights and tailoring according to nuanced socio-cultural influences, may encourage a greater sense of personal agency towards taking action to mitigate antibiotic resistance.

In low- and middle-income countries, heart failure (HF) is the leading cause of death and disability. A feasibility study was conducted to assess the fidelity, reach, and adoption of an educational program led by non-medical staff to improve outpatient care for patients hospitalized with HF in the local public health system. Thirty patients were included, with a mean age of 55.3 years (63.3% male). A total of 97.3% of planned home visits and 90% of scheduled phone calls were completed. Counselling modules were delivered during 90.4% of home visits, with no significant challenges reported during implementation. At the end of follow-up, there was a trend towards improved lifestyle habits, a reduction in mean heart rate (78.0 to 68.3 beats per minute; p = 0.016), a decrease in the proportion of patients in NYHA functional class III (20% to 7.4%; p = 0.041), and a slight reduction in mean body mass index (29.5 vs. 28.9; p = 0.042). A home-based educational program, designed to optimize outpatient management of heart failure and led by non-medical healthcare personnel, was well-received and demonstrated feasibility for implementation in low-income patients relying solely on the Argentine public health system.

We investigated the association between atrial fibrillation (AF) and dementia, and its subtypes (vascular-VaD, Alzheimer, mixed and rare dementia), and identified predictors for dementia in AF patients. The analysis was based on 183,610 patients with new-onset AF and 367,220 non-AF controls in the United Kingdom between 1998 and 2016, identified in three prospectively collected, linked electronic health records sources. Time-to-event (dementia or subtypes) analyses were performed using Cox proportional hazards and weighted Cox. Sub-analyses performed: including & censoring stroke and age (median used as cut-off). Over a median follow-up of 2.67 years (IQR .65-6.02) for AF patients and 5.84 years for non-AF patients (IQR 2.26-11.80), incidence of dementia in the AF cohort was 2.65 per 100 person-years, compared to 2.02 in the non-AF cohort. After adjustment, a significant association was observed between AF and all-cause dementia (HR = 1.38, 95% CI: 1.31-1.45), driven by a strong association with VaD (HR = 1.55, 95% CI: 1.41-1.70). AF was also associated with mixed dementia (HR = 1.26, 95% CI: 1.01-1.56), but we could not confirm an association with Alzheimer (HR = 1.05, 95% CI: .94-1.16) and rare dementia forms (HR = 1.19, 95% CI: .90-1.56). Ischemic stroke (HR = 1.40, 95% CI: 1.26-1.56), subarachnoid haemorrhage (HR = 2.08, 95% CI: 1.47-2.96), intracerebral haemorrhage (HR = 1.95, 95% CI: 1.54-2.48) and diabetes (HR = 1.32, 95% CI: 1.24-1.41) were identified as the strongest predictors of dementia in AF patients. AF patients have an increased risk of dementia, independent of stroke, with highest risk of VaD. Management and prevention of the identified risk factors could be crucial to reduce the increasing burden of dementia.

Male pattern hair loss (MPHL), also referred to as male androgenetic alopecia (AGA) is the most common type of non‐scarring progressive hair loss, with 80% of men suffering from this condition in their lifetime. In MPHL, the hair line recedes to a specific part of the scalp which cannot be accurately predicted. Hair is lost from the front, vertex, and the crown, yet temporal and occipital follicles remain. The visual effect of hair loss is due to hair follicle miniaturisation, where terminal hair follicles become dimensionally smaller. Miniaturisation is also characterised by a shortening of the growth phase of the hair cycle (anagen), and a prolongation of the dormant phase (kenogen). Together, these changes result in the production of thinner and shorter hair fibres, referred to as miniaturised or vellus hairs. It remains unclear why miniaturisation occurs in this specific pattern, with frontal follicles being susceptible while occipital follicles remain in a terminal state. One main factor we believe to be at play, which will be discussed in this viewpoint, is the developmental origin of the skin and hair follicle dermis on different regions of the scalp.

Free-of-charge (FoC) medicine schemes are increasingly available and allow access to investigational treatments outside clinical trials or in advance of licensing or NHS commissioning. We retrospectively reviewed FoC medicine schemes evaluated between 2013 and 2019 by a single NHS trust and a regional drug and therapeutics committee (DTC). The details of each locally reviewed FoC scheme, and any nationally available Medicines and Healthcare products Regulatory Agency Early Access to Medicines Scheme (MHRA EAMS) in the same period, were recorded and categorised. Most FoC schemes (95%) allowed access to medicines intended to address an unmet clinical need. Over 7years, 90% were company-FoC schemes and 10% were MHRA EAMS that were locally reviewed. Phase 3 clinical trial data were available for 44% of FoC schemes, 37% had phase 2 data and 19% were supported only by phase 1 data, retrospective observational studies or preclinical data. Utilisation of company-FoC schemes increased on average by 50% per year, while MHRA EAMS schemes showed little growth. Company-FoC medicine schemes are increasingly common. This may indicate a preference for pharmaceutical companies to independently co-ordinate schemes. Motivations for company-FoC schemes remain unclear and many provide access to treatments that are yet to be evaluated in appropriately conducted clinical trials, and whose efficacy and risk of harm remain uncertain. There is no standardisation of this practice and there is no regulatory oversight. Moreover, no standardised data collection framework is in place that could demonstrate the utility of such programmes in addressing unmet clinical need or to allow generation of further evidence.