Over the last couple years, The University of Kansas Health System (”UKHS”) has leveraged Maize Analytics’ machine learning auditing system to help automate the manual processes surrounding access and authorization in its process to ensure its compliance efforts and to best protect patient data. The auditing system allows privacy officials to focus on high-risk behavior, while reducing false positive alerts. The system learns to recognize when access is necessary based on clinical context (e.g., an appointment, medication order, etc.) in order to identify and rank suspicious record entries which may be lacking a clinical or operational justification and flags these particular record entries for review.
Once a potential unauthorized record entry has been identified, the privacy team investigates the access. Instead of completing the review in isolation, the privacy official uses Maize’s collaborative reviewer system to help streamline the process. For each suspicious access, the privacy official assigns the user’s manager (or other relevant personnel) to the investigation. The manager then provides input on the employee’s involvement with the patient’s care (e.g., was the employee floating on a floor to provide clinical support). To date, over 150 managers have participated as a reviewer of an investigation, allowing the privacy office to more efficiently work through cases and attain relevant information more quickly than before.
The deployment of the auditing system and the collaborative privacy process is helping UKHS to ensure its culture of compliance. UKHS employees, like most healthcare institution employees, are continuously trained and educated regarding HIPAA compliance and UKHS’s policies and procedures related to HIPAA. A part of UKHS’s thorough compliance training includes making employees aware that their accesses are being monitored, which UKHS believes is helping to deter non-compliant behavior. Since the system has been deployed, UKHS has been more efficient in monitoring and investigating possible unauthorized medical record access and has been able to achieve and confirm its goals related to HIPAA compliance. Moreover, because privacy responsibilities are now shared visibly across the organization, privacy processes are increasingly becoming a visible component of day-to-day operations in addition to scheduled mandatory and annual compliance training.
Ensuring the privacy of patient data is one of UKHS’s paramount responsibilities. In collaboration with Maize Analytics, the University of Kansas Health System is working to deploy effective tools and successful processes to protect the privacy of patients entrusted to its care.