Longitudinal K-means approaches to clustering and analyzing EHR opioid use trajectories for clinical subtypes.

Identification of patient subtypes from retrospective Electronic Health Record (EHR) data is fraught with inherent modeling issues, such as missing data and variable length time intervals, and the results obtained are highly dependent on data pre-processing strategies. As we move towards personalized medicine, assessing accurate patient subtypes will be a key factor in creating patient specific treatment plans. Partitioning longitudinal trajectories from irregularly spaced and variable length time intervals is a well-established, but open problem. In this work, we present and compare k-means approaches for subtyping opioid use trajectories from EHR data. We then interpret the resulting subtypes using decision trees, examining how each subtype is influenced by opioid medication features and patient diagnoses, procedures, and demographics. Finally, we discuss how the subtypes can be incorporated in static machine learning models as features in predicting opioid overdose and adverse events. The proposed methods are general, and can be extended to other EHR prescription dosage trajectories.

Defining Data Migration Across Multidisciplinary Ambulatory Clinics Using Participatory Design.

OBJECTIVE: This study aimed to develop an institutional approach for defining data migration based on participatory design principles. METHODS: We outline a collaborative approach to define data migration as part of an electronic health record (EHR) transition at an urban hospital with 20 ambulatory clinics, based on participatory design. We developed an institution-specific list of data for migration based on physician end-user feedback. In this paper, we review the project planning phases, multidisciplinary governance, and methods used. RESULTS: Detailed data migration feedback was obtained from 90% of participants. Depending on the specialty, requests for historical laboratory values ranged from 2 to as many as 145 unique laboratory types. Lookback periods requested by physicians varied and were ultimately assigned to provide the most clinical data. This clinical information was then combined to synthesize an overall proposed data migration request on behalf of the institution. CONCLUSION: Institutions undergoing an EHR transition should actively involve physician end-users and key stakeholders. Physician feedback is vital for developing a clinically relevant EHR environment but is often difficult to obtain. Challenges include physician time constraints and overall knowledge about health information technology. This study demonstrates how a participatory design can serve to improve the clinical end-user's understanding of the technical aspects of an EHR implementation, as well as enhance the outcomes of such projects.

Pancreaticopleural fistula: an uncommon cause of amylase-rich pleural effusion.

Pancreaticopleural fistula (PPF) causing pleural effusion as a complication of chronic pancreatitis is a rare finding. We present this finding in a 52-year-old man with a medical history significant for alcohol abuse, acute on chronic pancreatitis and severe chronic obstructive pulmonary disease, who presented with worsening dyspnoea for 3 days. CT scan of the chest showed a new large right-sided pleural effusion. Thoracentesis was performed and pleural fluid analysis showed an amylase-rich, exudative pleural effusion. The effusion reaccumulated within 3 days necessitating repeat thoracentesis. Endoscopic retrograde chloangiopancreatography showed contrast leak through a single disruption in the dorsal pancreatic duct, suspicious for an underlying PPF. The patient underwent stenting of the pancreatic duct with subsequent resolution of right-sided pleural effusion.

Flip flop fungus sign: an FDG PET sign of benign pulmonary nodules.

A 71-year-old man was referred to pulmonary clinic for incidental findings of hypermetabolic lung nodule and mediastinal adenopathy on CT FDG PET performed for evaluation of cough. The patient underwent bronchoscopy with endobronchial ultrasound that was non-diagnostic. The patient was subsequently sent for video-assisted thoracoscopic lymph node biopsy notable for confluent caseating granulomas due to chronic infection by Histoplasma capsulatum Review of previous PDG PET was notable for the flip flop fungus sign-a PDG PET finding that could have altered the patients' clinical course by potentially avoiding the need for invasive surgical tissue diagnosis.