AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium
-
Increasing regulatory incentives to computerize provider order entry (CPOE) and connect stores of unvalidated allergy information with the electronic health record (EHR) has created a perfect storm to overwhelm clinicians with high volumes of low or no value drug allergy alerts. Data sources include the patient and family, non-clinical staff, nurses, physicians and medical record sources. There has been little written on how to collect hypersensitivity information suited for drug allergy alerting. Opiates in particular are a frequently ordered class of drugs that have one of the highest rates of allergy alert override and are often a component of pre-populated Computerized Provider Order Entry (CPOE) order sets. Targeted research is needed to reduce alert volume, increase clinician acceptance, and improve patient safety and comfort. ⋯ With an increasingly complex, information dependent healthcare culture, clinicians do not have unlimited time and cognitive capacity to interpret and effectively act on high volumes of low value alerts. Drug allergy alerting was one of the earliest and supposedly simplest forms of CPOE clinical decision support (CDS), yet still has unacceptably high override rates. Targeted strategies to exclude GI non-allergic type hypersensitivities, mild overdose, or adverse effects could yield large reductions in overall drug overrides rates. Explicit allergy and severity definitions, staff training, and improved clinical decision support at the point of allergy data input are needed to inform how we process new and re-process historical allergy data.
-
AMIA Annu Symp Proc · Jan 2013
Comparative StudyMobile app versus Web app: a comparison using 2008-2012 "PubMed for Handhelds" server data.
Recent surveys show that mobile apps are more popular than Web apps. Apple's iTunes Store, now has about 800,000 apps and reported to have about 40 billion downloads. Android apps, although fewer, is available to the most number of smartphones today. ⋯ Month-by-month comparison showed a 3 to 5-fold increase in queries. The six-month total accesses comparison increased 280% from the previous four-year average. A review of 500 randomly selected queries revealed that the majority of queries were clinical questions ((97.8%) and 61% of these queries are searches related to therapy.
-
The Tele Intensive Care Unit (tele-ICU) supports a high volume, high acuity population of patients. There is a high-volume of incoming and outgoing calls, especially during the evening and night hours, through the tele-ICU hubs. The tele-ICU clinicians must be able to communicate effectively to team members in order to support the care of complex and critically ill patients while supporting and maintaining a standard to improve time to intervention. ⋯ The software provides a multi-relational database of message instances to mine information for evaluation and quality improvement for all entities that touch the tele-ICU. The software design incorporates years of critical care and software design experience combined with new skills acquired in an applied Health Informatics program. This software tool will function in the tele-ICU environment and perform as a front-end application that gathers, routes, and displays internal communication messages for intervention by priority and provider.
-
AMIA Annu Symp Proc · Jan 2013
A modified real AdaBoost algorithm to discover intensive care unit subgroups with a poor outcome.
The Intensive Care Unit (ICU) population is heterogeneous. At individual ICUs, the quality of care may vary within subgroups. We investigate whether poor outcomes of an ICU can be traced back to excess deaths in specific patient subgroups, by discovering candidate subgroups, with a modified adaptive decision tree boosting algorithm applied to 80 Dutch ICUs. ⋯ Variables Glasgow Coma Scale and age were used most. There were 29 ICUs with overall poor outcomes, and for 22 our algorithm found all excess deaths. A new method based on adaptive decision tree boosting discovered many subgroups of ICU patients for which there is potentially room for outcomes improvement.
-
AMIA Annu Symp Proc · Jan 2013
Developing predictive models using electronic medical records: challenges and pitfalls.
While Electronic Medical Records (EMR) contain detailed records of the patient-clinician encounter - vital signs, laboratory tests, symptoms, caregivers' notes, interventions prescribed and outcomes - developing predictive models from this data is not straightforward. These data contain systematic biases that violate assumptions made by off-the-shelf machine learning algorithms, commonly used in the literature to train predictive models. ⋯ We highlight the importance of carefully considering both the special characteristics of EMR as well as the intended clinical use of the predictive model and show that failure to do so could lead to developing models that are less useful in practice. Finally, we describe approaches for training and evaluating models on EMR using early prediction of septic shock as our example application.