HealthLeaders offers a look at issues and trends that are shaping the business of healthcare in the United States today. Free to download.
Artificial Intelligence in Healthcare
February 2019
Confidence is high among health care industry executives that AI technologies will drive more access and affordable care.
Across the health care industry, providers, payers and other stakeholders have been challenged to reduce costs while improving quality outcomes, the patient experience, and care. Increasingly, artificial intelligence technologies are seen as the solution that will help the industry achieve these goals and move toward a better future.
The healthcare community continues to see a rise in opioid misuse and drug-overdose deaths. In response, public health and healthcare entities have launched awareness campaigns and policy changes in an attempt to curb inappropriate use and prevent addiction. But has the effort to increase public and provider awareness decreased prescription rates? IBM® Watson Health™ conducted two studies to explore this issue.
While the CMS Hospital-Acquired Conditions (HACs) are believed to represent potentially avoidable complications of care, it is evident that many of these adverse outcomes occur each year.
In fiscal year 2016, nearly 28,000 patient safety adverse outcomes across 14 HACs were reported among 20,297,766 civilian inpatients.* The total cost of HACs in terms of days of stay, mortality, and costs of care are a function of the frequency of the adverse outcome and the incremental impact of the HAC on the measure of interest.
Researchers from IBM® Watson Health™ set out to evaluate the incremental consequences of selected inpatient adverse outcomes from selected CMS Hospital-Acquired Conditions (HAC) in terms of mortality, length of stay, and total hospital cost per case using all-payer data from acute care hospitals in the U.S.**
*Federal fiscal year 2016, IBM Projected Inpatient Data Base.
**Federal fiscal year 2016, IBM Projected Inpatient Data Base.
The Childhood Obesity Epidemic: Insights from claims plus EMR data
August 2018
Claims data alone may not accurately reflect the underlying prevalence of childhood obesity; healthcare providers may be less likely to add an obesity diagnosis to a claim because of the stigma associated with the condition.
Linking claims data and clinical data from electronic medical records (EMRs) can enhance the awareness and management of childhood obesity and the associated trajectories of care.
IBM Watson Health analyzed EMR data to differentiate between the prevalence of clinical evidence and the diagnosis of obesity in childhood, as well as describe how the combined data assets of claims and EMR data provide a more comprehensive understanding of childhood obesity and potential related comorbidities.
