Bridging the Gap: Use of Optical Character Recognition Software As a Process Improvement Solution to Public Health Data Collection

Tuesday, June 6, 2017: 2:40 PM
410B, Boise Centre
Angelique Harding , Houston Health Department, Houston, TX
Avi Raju , Houston Health Department, Houston, TX
Kavitha Gantu , Houston Health Department, Houston, TX
Biru Yang , Houston Health Department, Houston, TX

BACKGROUND: At the cornerstone of the core functions and essential services of public health, is data collection. The methods, types, and reasons for public health reporting vary widely. Nationally, local health departments (LHDs) are trying to balance the transition from traditional paper-based reporting to electronic reporting, hastened by Meaningful Use (MU). Traditional public health reporting involves notification by mail, phone, and/or facsimile. Annually, the Houston Health Department (HHD) receives thousands of paper-based reports for 80 reportable conditions within our jurisdiction. HHD deployed a process improvement initiative to identify an intermediary solution to electronic case reporting.

METHODS: Alternatives to those methods have been created as a result of the rapid pace of technology. HHD Informatics Group utilized the Define, Review, Identify, Verify, and Execute (DRIVE) approach to process improvement and the Inputs, Controls, Outputs, and Resources (ICOR) methodology for process mapping. Individual meetings were held with 4 surveillance staff members to collect information. Additionally 3 sessions of the informatics and surveillance workgroup were attended to throughout the project. These methods and tools helped HHD to identify a solution that would serve as a bridge between traditional reporting and electronic reporting.

RESULTS: HHD is employing the use of optical character recognition (OCR) to improve the data collection for our surveillance epidemiologists and reporters. The software has the capacity to process large volumes of paper, convert handwriting to text, and load data directly into our surveillance system. Following procurement and installation of the OCR application, informaticians and surveillance investigators were provided training. A total of three data collection forms were created. Forms were prioritized by collaborating with surveillance epidemiologists to assess staff burden, reporting frequency, and disease. Field types were selected to improve the quality and completeness for required data elements.

CONCLUSIONS: OCR technology was identified as a responsive alternative to the burdensome paper-based reporting at HHD. Process improvements that involve software solutions require buy-in from leadership and coordination with information technology (IT) staff. The DRIVE and ICOR methods of process improvement not only produce solutions but can serve as a communication tool to ensure stakeholders understand the process. Informatics plays an essential role in facilitating efficient and effective data collection.