Acute kidney injury and early fluid load in a retrospective cohort of neonatal sepsis.

BACKGROUND: Sepsis and acute kidney injury (AKI) are associated with mortality in the newborn intensive care unit (NICU). There is a paucity of studies that describe AKI and fluid overload in neonatal sepsis and their association with mortality. METHODS: Retrospective study of neonates with culture positive sepsis admitted to the NICU between June 2020 and June 2021 was conducted. Primary outcome was in-hospital mortality according to AKI as defined by the neonatal modified Kidney Diseases Improving Outcomes criteria. Secondary outcomes were early fluid overload and vasopressor use. RESULTS: Thirty-three percent of neonates had AKI with sepsis, and 57% of cases were severe AKI. AKI was associated with mortality after adjusting for variables that were different between survivors and non-survivors (aOR 5.7 [95% CI 1.1-36], p = 0.04). Early fluid overload occurred in 27% of neonates who were at higher risk of having AKI with sepsis (OR 7.4 [95% CI 1.6-26.0], p = 0.01) and higher risk of mortality (aOR 17.8 [95% CI 2-7545], p = 0.02). CONCLUSIONS: AKI and early fluid overload are associated with mortality in sepsis in our retrospective cohort. Mitigating AKI and early fluid overload in sepsis might be a fruitful strategy in reducing mortality with sepsis. A higher resolution version of the Graphical abstract is available as Supplementary information.

Ethical Redress of Racial Inequities in AI: Lessons from Decoupling Machine Learning from Optimization in Medical Appointment Scheduling.

An Artificial Intelligence algorithm trained on data that reflect racial biases may yield racially biased outputs, even if the algorithm on its own is unbiased. For example, algorithms used to schedule medical appointments in the USA predict that Black patients are at a higher risk of no-show than non-Black patients, though technically accurate given existing data that prediction results in Black patients being overwhelmingly scheduled in appointment slots that cause longer wait times than non-Black patients. This perpetuates racial inequity, in this case lesser access to medical care. This gives rise to one type of Accuracy-Fairness trade-off: preserve the efficiency offered by using AI to schedule appointments or discard that efficiency in order to avoid perpetuating ethno-racial disparities. Similar trade-offs arise in a range of AI applications including others in medicine, as well as in education, judicial systems, and public security, among others. This article presents a framework for addressing such trade-offs where Machine Learning and Optimization components of the algorithm are decoupled. Applied to medical appointment scheduling, our framework articulates four approaches intervening in different ways on different components of the algorithm. Each yields specific results, in one case preserving accuracy comparable to the current state-of-the-art while eliminating the disparity.

Pharmaceutical policy in China.

Contradictory goals plague China's pharmaceutical policy. The government wants to develop the domestic pharmaceutical industry and has used drug pricing to cross-subsidize public hospitals. Yet the government also aims to control drug spending through price caps and profit-margin regulations to guarantee access even for poor patients. The resulting system has distorted market incentives, increased consumers' costs, and financially rewarded inappropriate prescribing, thus undermining public health. Pharmaceuticals account for about half of total health spending in China, representing 43 percent of spending per inpatient episode and 51 percent of spending per outpatient visit. Yet some essential medicines are unavailable or of questionable quality.