Search AMIA.org

To estimate the risk of hospital admission and mortality from COVID-19 to patients and measure the association of race and area-level social vulnerability with those outcomes.

COVID-19 mortality forecasting models provide critical information about the trajectory of the pandemic, which is used by policymakers and public health officials to guide decision-making. However, thousands of published COVID-19 mortality forecasts now exist, many with their own unique methods, assumptions, format, and visualization. As a result, it is difficult to compare models and understand under which circumstances a model performs best. Here, we describe the construction and usability of [...]

Our aim is to develop a patient engagement technology that makes it easy for patients to access their own medical information and share it with others.

Patient-generated health data (PGHD) have not achieved widespread clinical adoption. However, the COVID-induced shift to telemedicine may have created opportunities for PGHD as surrogates for vital signs collected in person. We assessed whether this shift was associated with greater ambulatory care PGHD use. We conducted an interrupted time series analysis of physician enrollment in, and patient-initiated vital sign transmission of non-COVID-associated PGHD through, a national PGHD platform (Validic). Ten health [...]

Clinical workflows require the ability to synthesize and act on existing and emerging patient information. While offering multiple benefits, in many circumstances electronic health records (EHRs) do not adequately support these needs.

De-identification is a fundamental task in electronic health records to remove protected health information entities. Deep learning models have proven to be promising tools to automate de-identification processes. However, when the target domain (where the model is applied) is different from the source domain (where the model is trained), the model often suffers a significant performance drop, commonly referred to as domain adaptation issue. In de-identification, domain adaptation issues can [...]

Using a risk stratification model to guide clinical practice often requires the choice of a cutoff-called the decision threshold-on the model's output to trigger a subsequent action such as an electronic alert. Choosing this cutoff is not always straightforward. We propose a flexible approach that leverages the collective information in treatment decisions made in real life to learn reference decision thresholds from physician practice. Using the example of prescribing a [...]