Clinical prediction models typically make point estimates of risk. However, values of key variables are often missing during model development or at prediction time, meaning that the point estimates mask significant uncertainty and can lead to over-confident decision making. We present a model of mortality risk in emergency laparotomy which instead presents a distribution of predicted risks, highlighting the uncertainty over the risk of death with an intuitive visualisation. We developed and validated our model using data from 127134 emergency laparotomies from patients in England and Wales during 2013–2019. We captured the uncertainty arising from missing data using multiple imputation, allowing prospective, patient-specific imputation for variables that were frequently missing. Prospective imputation allows early prognostication in patients where these variables are not yet measured, accounting for the additional uncertainty this induces. Our model showed good discrimination and calibration (95% confidence intervals: Brier score 0.071–0.078, C statistic 0.859–0.873, calibration error 0.031–0.059) on unseen data from 37 hospitals, consistently improving upon the current gold-standard model. The dispersion of the predicted risks varied significantly between patients and increased where prospective imputation occurred. We present a case study that illustrates the potential impact of uncertainty quantification on clinical decision making. Our model improves mortality risk prediction in emergency laparotomy and has the potential to inform decision-makers and assist discussions with patients and their families. Our analysis code was robustly developed and is publicly available for easy replication of our study and adaptation to predicting other outcomes.

There has existed a severe ventilator deficit in much of the world for many years, due in part to the high cost and complexity of traditional ICU ventilators. This was highlighted and exacerbated by the emergence of the COVID-19 pandemic, during which the increase in ventilator production rapidly overran the global supply chains for components. In response, we propose a new approach to ventilator design that meets the performance requirements for COVID-19 patients, while using components that minimise interference with the existing ventilator supply chains. The majority of current ventilator designs use proportional valves and flow sensors, which remain in short supply over a year into the pandemic. In the proposed design, the core components are on-off valves. Unlike proportional valves, on-off valves are widely available, but accurate control of ventilation using on-off valves is not straightforward. Our proposed solution combines four on-off valves, a two-litre reservoir, an oxygen sensor and two pressure sensors. Benchtop testing of a prototype was performed with a commercially available flow analyser and test lungs. We investigated the accuracy and precision of the prototype using both compressed gas supplies and a portable oxygen concentrator, and demonstrated the long-term durability over 15 days. The precision and accuracy of ventilation parameters were within the ranges specified in international guidelines in all tests. A numerical model of the system was developed and validated against experimental data. The model was used to determine usable ranges of valve flow coefficients to increase supply chain flexibility. This new design provides the performance necessary for the majority of patients that require ventilation. Applications include COVID-19 as well as pneumonia, influenza, and tuberculosis, which remain major causes of mortality in low and middle income countries. The robustness, energy efficiency, ease of maintenance, price and availability of on-off valves are all advantageous over proportional valves. As a result, the proposed ventilator design will cost significantly less to manufacture and maintain than current market designs and has the potential to increase global ventilator availability.