Because humans provide a much-needed dimension to drive decision-making, d2i stands firmly in using intelligence that leverages practical experience combined with reliable data sets that can help improve clinical operations.
Intelligence that leverages both practical experience and trusted data is key to improving both ED efficiency and clinical outcomes.
Advanced analytics purpose-built for optimizing clinical operations is critical for driving ED efficiency and improving outcomes. Even if you know the questions to ask, based on high-level metrics, it’s often difficult to identify high-impact actionable insights. d2i’s hundreds of context-sensitive analytics (with millions of variations) anticipates your questions, bringing you the right answers just a click away.
A computer’s proficiency in identifying patterns in big data gives Artificial Intelligence (AI) programs the ability to learn. d2i’s Advanced Analytic Application learns from reliable data as well as from the information we’ve gleaned from millions of ED visits, thousands of providers, and 20 years of working in the field of emergency medicine.
AI is driving a paradigm shift in the health care industry, but it’s important to remember that the most effective use of AI is one that enhances human capabilities instead of replacing humans. The latest research suggests that it wouldn’t be in our best interest to do so anyway, because humans provide a much-needed dimension to drive decision-making that AI lacks — at least for now.
Can AI perform better than human doctors?
A recent MIT study indicated that AI can’t replace human doctors’ “gut instinct” when it comes to analyzing medical data to diagnose and treat health problems. The study showed the decision-making process is driven by doctors’ sentiments about their patients, coupled with intuition, the part not directly related to data analysis.
The researchers performed a sentiment analysis of doctors’ written notes in the medical records of 60,000 ICU patients admitted to Beth Israel Deaconess Medical Center in Boston over a 10-year period. Using computer algorithms, sentiment analysis examined positive and negative sentiments associated with written language.
Research uncovered that doctors’ diagnostic and treatment decisions were based on factors including family history, severity of the medical problem, and lifestyle habits. That data was supplemented by the doctors’ “gut feelings.
Interestingly, the researchers found that intuition played a more important role during the first couple of days of a patient’s hospital stay, before much data is collected. Also, if doctors felt pessimistic about a patient’s condition, they ordered more testing but only up to a certain point, after which fewer tests were ordered.
What AI Can Do for Health Care
Integrating aspects of AI into health care can be hugely beneficial in driving meaningful change across the continuum of care. The consulting firm Frost & Sullivan reported that the health care AI market is projected to grow 40 percent through 2021, potentially improving outcomes 30-40 percent and cutting treatment costs by as much as half.
Some of the clear implementation benefits to date include:
- Helping doctors’ deliver accurate disease diagnosis and treatment.
- Addressing the physician shortage in HCOs of all sizes.
- Facilitating medical diagnostics. Diagnostic errors are estimated to affect about 10 percent of patient deaths and 6-17 percent of complications. AI can introduce less time-consuming, more accurate methods.
- Improving health care marketing, in the pharma sector in particular, by employing machine learning-based predictive analytics tools to find optimal treatment and refine patient targeting.
- Reducing medication non-adherence. Americans are spending $309.5 billion annually on prescription drugs, and using algorithmic tools can help reduce medication non-adherence-related waste by identifying which patients are likely to deviate from their prescribed treatment regimen, and therefore need more support and targeted communication.
Perhaps the ideal approach to improving both operational efficiency and clinical outcomes is one that embodies the best of both worlds — human and machine — while putting the control and the knowledge management squarely in the hands of domain experts and those responsible for patient care. Analytics that are organically developed by combining evidence-based data enhanced with human experience can then be directly integrated into an Advanced Analytic Application. Analytics can help identify opportunities with the potential for improvement, demonstrate the impact adjustments have on performance, and then continuously reinforce behaviors that support new processes.