JAMA network open
-
Accurate prediction of outcomes among patients in intensive care units (ICUs) is important for clinical research and monitoring care quality. Most existing prediction models do not take full advantage of the electronic health record, using only the single worst value of laboratory tests and vital signs and largely ignoring information present in free-text notes. Whether capturing more of the available data and applying machine learning and natural language processing (NLP) can improve and automate the prediction of outcomes among patients in the ICU remains unknown. ⋯ Intensive care unit mortality prediction models incorporating measures of clinical trajectory and NLP-derived terms yielded excellent predictive performance and generalized well in this sample of hospitals. The role of these automated algorithms, particularly those using unstructured data from notes and other sources, in clinical research and quality improvement seems to merit additional investigation.
-
Pragmatic clinical trials that seek informed consent after randomization (ie, postrandomization consent) are increasingly used, but debate on ethics persists because control arm patients are not specifically informed about the trials and randomization occurs before consent for the trials. The public's attitude toward postrandomization consent trials is unknown, but the way the trials are described could bias people's views. ⋯ The public's generally high rate of approval of the ethics of postrandomization informed consent for pragmatic trial designs does not appear to be affected by whether postrandomization consent design is framed using traditional randomized clinical trial terminology, regardless of the stakes of the trial. Promoting better understanding of the design may increase its acceptance by the public.
-
Although readmission rates are declining under Medicare's Hospital Readmissions Reduction Program (HRRP), concerns remain that the HRRP will harm quality at safety-net hospitals because they are penalized more often. Disparities between white and black patients might widen because more black patients receive care at safety-net hospitals. Disparities may be particularly worse for clinical conditions not targeted by the HRRP because hospitals might reallocate resources toward targeted conditions (acute myocardial infarction, pneumonia, and heart failure) at the expense of nontargeted conditions. ⋯ Findings from this study suggest that disparities are widening within safety-net hospitals, specifically for non-HRRP-targeted conditions. Although increases in racial disparities for nontargeted conditions were modest, they represent 6 times more discharges in our cohort than targeted conditions.
-
Navigating health insurance and health care choices requires considerable health insurance literacy. Although recommended preventive services are exempt from out-of-pocket costs under the Affordable Care Act, many people may remain unaware of this provision and its effect on their required payment. Little is known about the association between individuals' health insurance literacy and their use of preventive or nonpreventive health care services. ⋯ This study's findings suggest that lower health insurance literacy may be associated with greater avoidance of both preventive and nonpreventive services. It appears that to improve appropriate use of recommended health care services, including preventive health services, clinicians, health plans, and policymakers may need to communicate health insurance concepts in accessible ways regardless of individuals' health insurance literacy. Plain language communication may be able to improve patients' understanding of services exempt from out-of-pocket costs.