Challenge
A mental health provider was struggling with a lengthy intake process and growing waiting lists. Their initial assessment required 90-minute sessions with specialized staff, creating a bottleneck that delayed patient care, increased dropout rates, and prevented efficient practitioner matching.
Solution
An AI triage system to assess needs and optimize practitioner matching using Claude API and Make.com that:
- Guided patients through an intelligent assessment process
- Analyzed responses to identify urgency, condition indicators, and treatment needs
- Matched patients with the most appropriate specialists based on needs and expertise
- Prioritized cases based on clinical urgency while optimizing schedule utilization
- Provided preliminary insights to practitioners before the first appointment
- Tracked outcomes to continuously improve matching algorithms
Implementation
The system was implemented over 7 weeks:
- Clinical protocol digitization and assessment flow design
- Model development for condition identification and urgency assessment
- Practitioner matching algorithm creation based on expertise and outcomes data
- Integration with scheduling and electronic health record systems
- Clinical validation and refinement with practitioner feedback
Results
- Intake process reduced from 90 to 20 minutes, while gathering more comprehensive information
- 38% more patients accommodated with the same clinical staff
- 1.5 FTE saved in assessment work, allowing clinicians to focus on treatment
- Treatment outcomes improved by 23% through better practitioner matching
Privacy and Ethics
The system was designed with strict privacy controls and ethical guidelines, including transparent disclosure about AI usage and human clinical oversight of all recommendations.
