Borgess Health: A Community Health System

Borgess Health, founded in 1889 and based in Kalamazoo, Michigan, is a community-focused health system operating hospitals, urgent care centers, and physician practices. As part of the larger Ascension network, it provides a comprehensive range of medical services to southwestern Michigan. With 1001-5000 employees, it represents a mid-sized player in the hospital sector, large enough to have complex operational challenges but often without the vast R&D budgets of national health giants. Its mission centers on personalized, compassionate care within its community.

Why AI Matters at This Scale

For a health system of Borgess's size, AI is not a futuristic luxury but a practical tool for survival and improvement. The organization faces universal healthcare pressures: tightening margins, clinician burnout, staffing shortages, and a shift toward value-based care that rewards quality and efficiency over volume. At this scale—large enough to generate significant data but agile enough to implement focused changes—AI offers a path to do more with existing resources. It can automate administrative burdens, provide clinical decision support to augment staff, and optimize complex systems like patient flow and supply chains. Ignoring AI risks falling behind in care quality and financial performance, while strategic adoption can strengthen community trust and operational sustainability.

Concrete AI Opportunities with ROI Framing

1. Predictive Analytics for Patient Flow

Implementing machine learning models to forecast admissions and predict patient length-of-stay can dramatically improve bed management and reduce emergency department bottlenecks. For Borgess, this translates directly to increased capacity without physical expansion, higher patient satisfaction, and better revenue capture from available beds. ROI comes from reduced need for costly temporary staff and avoided penalties for care delays.

2. Clinical Decision Support for Sepsis Detection

AI algorithms that continuously analyze electronic health record (EHR) data in real-time can identify early, subtle signs of sepsis—a leading cause of hospital mortality and cost. Early detection allows for faster intervention, improving survival rates and reducing average length of stay and associated treatment costs. The ROI is measured in saved lives, improved quality metrics, and significant avoidance of high-cost complications.

3. Robotic Process Automation (RPA) for Revenue Cycle

Deploying AI-driven RPA to handle repetitive back-office tasks like claims processing, prior authorization, and patient billing follow-up can reduce administrative costs by 30-50%. For a mid-sized system, this can free up millions in annual operational expense and reduce revenue cycle time. The ROI is clear, quantifiable, and rapid, often within a single fiscal year, while also improving accuracy and staff satisfaction by removing tedious work.

Deployment Risks Specific to This Size Band

Borgess's size presents unique implementation risks. First, resource constraints: while large enough to need AI, it may lack a dedicated data science team, requiring reliance on vendors or parent-network support, which can create dependency and integration challenges. Second, legacy system integration: mid-sized hospitals often have a patchwork of older IT systems; connecting them for unified AI data feeds is costly and complex. Third, change management at scale: rolling out new AI tools to a workforce of thousands requires extensive training and can meet resistance if not coupled with strong clinician leadership and clear communication on benefits. A failed pilot can poison the well for future initiatives. Finally, data governance: ensuring high-quality, standardized, and secure data for AI models is a significant undertaking that requires cross-departmental coordination often difficult in organizations not designed for agile tech projects. A focused, phased approach starting with high-ROI, low-complexity use cases is critical to mitigate these risks.