AdaptHealth
AI-enabled processing of incoming referral faxes
AdaptHealth's AI process read incoming faxes with 99.6% accuracy, against 89% for its legacy process
Problem
AdaptHealth, a US home medical equipment provider, received more than five million pages of faxes every month. Much of the content was unstructured referral data that staff had to sort and route by hand. The legacy process was accurate about 89% of the time.
Approach
The company built an automated process enabled by AI to handle the incoming unstructured documents. According to the CEO, it went live within weeks. The company described it as a first step toward automating more of the workflow.
Results
Takeaway
Document intake was a high-volume task where an accuracy gap against the old process could be measured directly. The figure comes from management's own account on an earnings call.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Transcript of AdaptHealth's third-quarter 2024 earnings call. CEO Suzanne Foster said the company received more than five million pages of faxes a month and that its new AI-enabled automated process was 99.6% accurate against 89% for the legacy process. She said it went live within weeks.
Full details are in the original sources.