Build Journal · THC CDSS

Mapping AI assessments to the programs patients actually enroll in

By Ganesh R Datta. I build healthcare AI end to end: clinical decision support, clinic intake, and the marketing and operations around them.

An assessment that ends in a generic list of recommendations creates more work. Someone still has to turn it into a plan the clinic actually offers.

The clinic's structure

Total Health Centers delivers care through 12-week programs focused on heart, brain and longevity. Patients enroll in a program, not a list of tips.

Designing the output

THC CDSS maps the assessment directly into those programs. The physician sees a proposed plan in the shape the clinic already delivers, ready to review and adjust.

Why this is the important design choice

The value of decision support is measured at the moment of action. If the output matches how the clinic delivers care, the gap between "assessment done" and "patient started" shrinks.

For anyone building a CDSS

Ask the clinic what a finished plan looks like before you design the assessment. Build backwards from it.

More from the Build Journal

• What building clinical decision support across streams of medicine taught me
• Decision support, not decision making: where I draw the line in a CDSS
• Structuring an integrative assessment around root causes
• Building one assessment across modern medicine, Ayurveda and diagnostics

The full Build Journal · The THC CDSS work · Book a 15-minute call