As healthcare delivery and payment systems are redesigned, providers are shouldering a greater share of risk. They also are presented with an opportunity to employ new models to deliver sustainable, cost-effective care and cost-effective access to imaging. By leveraging big data, imaging providers may meet these dual goals, according to the policy brief "Beyond Fee-For-Service: Emerging Payment Models in Radiology," issued by the Harvey L. Neiman Health Policy Institute.
The brief detailed two examples of employing big data to analyze and improve specialty services.
Inpatient hospital episodes, according to the brief, are relatively standardized and provide a foundation to examine the involvement of specialists in provision of patient care. Take, for example, acute ischemic stroke. By mining the Medicare Severity Diagnosis Related Groups (MS-DRG) system files, researchers observed “tremendous variability” in imaging costs per episode.
Reviewing these data, they noted the mean value is much closer to the 75 th percentile than the median. “This indicates that the average cost of care is heavily influenced by a relatively small number of very expensive encounters.”
Deeper analysis could allow providers to benchmark their services, identify opportunities to address increased risk under emerging models and locate providers who successfully balance lower costs with improved outcomes, whose practices may serve as a best practices template.
Screening mammography, which is nearly entirely managed by radiologists, provides another opportunity to analyze variability in delivery of patient care. “Analysis of that variability, with regard to utilization, costs, and outcomes would similarly help identify excessive, unjustifiable, or low utility care (as in our inpatient model) to preferentially drive value.”
The screening mammography analytical approach may serve as a model for other imaging indications, including low-dose CT screening and abdominal incidentalomas.
The paper listed other developments expected to solidify in new delivery models. This includes decision support with embedded appropriateness criteria applied in coordination with radiologists.
Additional tools that leverage the potential of big data to redesign specialist care also are likely to emerge, concluded the authors, Danny R. Hughs, PhD, research director and senior research fellow, and Richard Duszak, MD, CEO and senior research fellow.
To view the report click here.