Can AI-Driven Care Coordination Software Improve Workflows?

University Hospitals is leveraging AI to enhance care coordination across its network of 13 hospitals and numerous outpatient settings. This effort highlights the transformative potential of AI-driven platforms in streamlining workflows, improving patient outcomes, and addressing clinician burnout.

---

The Role of AI in Care Coordination

Care coordination ensures seamless collaboration between healthcare providers, aiming for safe, appropriate, and effective treatment. Effective information-sharing can:

• Minimize repeat lab tests.
• Reduce unnecessary visits.
• Address medication issues.

According to the U.S. Centers for Medicare & Medicaid Services (CMS), poor care coordination can lead to:

• Increased medical errors.
• Poor care transitions.
• Higher costs for patients.

The Agency for Healthcare Research and Quality (AHRQ) advocates for a mix of technology adoption and care-specific strategies, such as proactive care plans tailored to patient needs. While electronic health records (EHRs) aid in these efforts, AI’s ability to analyze vast data sets positions it as the next evolution in care coordination.

---

University Hospitals’ AI Initiative

University Hospitals has partnered with Aidoc to deploy its AI-powered platform, aiOS, to improve radiology and care coordination workflows. Chair of Radiology Donna Plecha shared insights on how AI is already assisting in their operations:

1. Early Adoption and Pilots:
   • The system has used AI tools for years, including portable chest X-ray algorithms to detect collapsed lungs and misplaced tubes.
   • The current partnership focuses on screening for conditions like pulmonary embolism using FDA-cleared AI algorithms.
2. Streamlining Care Coordination:
   • AI prioritizes urgent findings, allowing radiologists to address critical cases faster.
   • For example, pulmonary embolisms, which are subtle yet potentially fatal, are flagged more efficiently with AI.
3. Reducing Burnout:
   • With a shortage of radiologists and increasing workloads, AI algorithms reduce manual tasks and improve efficiency.
   • Plecha emphasizes that radiologists remain integral to the diagnostic process, using AI as a supportive tool.

---

Best Practices for Implementing AI

1. Identify High-Value Use Cases:

• University Hospitals performed an audit to avoid redundancies, eliminating existing use cases (e.g., pneumothorax screening) while focusing on gaps like pulmonary embolism detection.

2. Conduct Architectural Reviews:

• IT teams evaluate how AI integrates with existing systems, such as picture archiving and communication systems (PACS).
• This ensures that tools enhance, rather than hinder, workflows.

3. Monitor ROI and Metrics:

• Key performance indicators include:
  • Time to treatment for flagged cases.
  • Accuracy metrics like true positives and false negatives.
• AI tools can prioritize urgent cases, reducing treatment delays in critical scenarios.

4. Gain Clinician Buy-In:

• Engaging clinicians in the adoption process and addressing their concerns about AI’s role fosters trust and acceptance.
• Continuous feedback and shared learnings improve integration.

---

Looking Ahead

AI is proving to be a valuable tool in care coordination, but its adoption requires realistic expectations and a thoughtful approach. Plecha underscores that AI won’t replace radiologists but will empower those who embrace it. As healthcare faces increasing patient volumes and clinician shortages, leveraging AI to reduce workloads and enhance care quality is becoming a necessity.

With ongoing evaluations and phased implementations, University Hospitals is setting a precedent for how AI can drive innovation in care coordination while maintaining clinician oversight and patient trust.