There are a number of challenges facing integration of predictive analytics in health care.
1. Lack of standardized data and interoperability – Currently the penetration of EHR’s in hospital and clinics is pretty dismal. Furthermore, most EHR vendors try to lock in hospitals and clinics by using proprietary standards to prevent exchange of data. This limits the availability of usable data to perform predictive analytics, which in turn effects the reliability of information generated
2. Misaligned incentives for ROI – The projected cost savings from using EMR’s will most likely benefit payers such as insurance companies due to reduced utilization rates. Therefore, hospitals and clinics are hesitant to bear the costs of implementation due to uncertain ROI
3. Accountability – Information generated from predictive analytics has the potential to expose inefficiencies in the current system. This will hold leaders accountable and force change. This may require huge investments in time and money without substantial ROI
The ARRA and HITECH act have spurned adoption and implementation of EHR’s and HIE’s throughout the country. Although many challenges still remain in this area, this is a step in the right direction. The data collected from these EHR’s will eventually lead to better analytical tools and business intelligence. Furthermore, comparative effectiveness research done using real time data will help improve clinical standards and interventions.
Formation of ACO’s, bundled payment and elimination of fee for service are reforms designed to align incentives to improve health of the population. This will spurn adoption of tools and techniques including standards of care and predictive analytics to provide more efficient and effective care.
Accountability requires a culture change, and strong leadership can only achieve this. Furthermore, if financial incentives are aligned correctly, leaders will be more enthusiastic about using information gleaned from predictive analytics to champion implementation of standards of care.