Abstract WP321: Smartphone Application Iodine

    loading  Checking for direct PDF access through Ovid


Background and Purpose: Stroke studies and trials frequently require identification and screening of specific stroke patients based on set selection criteria, missed by traditional methods that rely on physician reporting and discharge diagnosis codes. Our aim was to determine the utility of a smartphone application, Iodine, in screening and identifying cervical artery dissection (CAD) patients.

Methods: Iodine is used to notify healthcare providers regarding specific patient populations for the purpose of process and quality improvement. Real-time patient data is sent to Iodine via feeds secured by encrypted virtual private network tunnels, and continuously evaluated against Iodine’s library of Alert rules. All communication between Iodine and users are encrypted, all data accesses are audited, and data is not stored on mobile devices, thereby ensuring Health Insurance Portability and Accountability Act compliance. CAD is an uncommon cause for stroke or TIA that can affect young patients without traditional risk factors. Recent advances in imaging, including MRI and CT angiogram, have led to more frequent identification of CAD. Based on the identification of the keyword “dissection” in all radiological reporting, we setup a notification to alert study investigators. These notifications were reviewed daily for enrollment in an ongoing prospective study COMPASS. These patients were compared with those identified by ICD-9 discharge diagnosis codes.

Results: In a six month period between January 01, 2015 and June 30, 2015, 18 patients with CAD were identified through the Iodine out of 289 total dissection alerts. (Mean age ± SD = 57 ± 17, 50% male, 61% Caucasian, and 39% African American). Of these 18 patients, 3 died in-hospital and thus could not be identified using traditional methods. Of these, only 13 patients were identified through ICD-9 discharge diagnosis codes, out of 4683 total discharges. Iodine was found to have a significantly higher proportion of detecting CAD (χ2=146.1; p<0.0001).

Conclusions: We report real time screening and identification of CAD using Iodine, which appears to be superior to traditional methods. This method may be successfully applied for other disease conditions for consecutive study patient identification.

Related Topics

    loading  Loading Related Articles