Workflow P2 · From three patient experience studies
Patient experience listening and equity
What leaders see: What patients say between surveys, by topic, place and community, with gaps tracked to action.
- Who runs it
- Patient experience, quality, health equity office, communications
- Cadence
- Daily intake; monthly quality review
- Plugs into
- Social listening exportsComplaint systemSurvey vendor dataQuality dashboards
The workflow you run today, with 4 AI touchpoints
- 1Feedback arrivesPosts, complaints, comments
- 2Relevant or notMost mentions are not experienceSmall model
Classify which posts are patient experience; about 1 in 10 hospital mentions are
- 3Topics and sentimentWhat went well or wrongLLM
Group topics and tone; staff check a weekly sample
- 4Equity viewBy community and areaPredict
Break results down by community, with representativeness checks
- 5Quality reviewMonthly with leaders
- 6Respond and fixOwners and deadlines
- 7Track trendsAgainst formal surveysWeak signals
Flag rising issues early; compare with HCAHPS-style surveys
Highlighted steps are where AI plugs into the existing workflow. People keep every decision.
Patient voice, between surveys
- Public posts and reviews
- Complaints and comment cards
- Formal survey scores
- Census and demographic baselines
- Experience classifier
- Topic and sentiment models
- Equity breakdowns with bias checks
- Trend alerts
- Quality committee sets priorities
- Equity office owns gaps by group
- Fix the top issues with named owners
- Close the loop with patients and staff
- Results compared with formal surveys each quarter
- Labels refreshed as language changes
What changes
- Patient voice every week, not once a year
- Disparities by community visible early
- Every top issue has an owner and a date
How we measure it
| KPI | Today | With GTU | Type |
|---|---|---|---|
| Time from feedback to review | Months, with surveys | Days | Planning target |
| Feedback classified and routed | Read by hand, if at all | All incoming feedback | New measure |
| Experience gaps tracked by community | Hidden in averages | Reported monthly | New measure |
| Agreement with formal surveys | Unknown | Reviewed each quarter | New measure |
Targets are set with your team and tested in a pilot before anything scales.
Proof from our work: 2.76 million patient experience posts across the US over four years (JMIR, 2018); 404,065 posts to 2,349 hospitals and the CrowdClinical platform for 3,000+ hospitals (BMJ Quality & Safety); racial and ethnic gaps across 852,000 posts (JMIR, 2020). Longitudinal study · Hospital signals · Equity signals
Next step
Run this workflow on your data.
Start with a free 30-minute discovery call. Most teams pilot one touchpoint first.