How patient satisfaction surveys work
CARE sends a survey the day after the visit, scores it, and emails you within moments when someone rates you badly. It also produces the MIPS quality reporting and net promoter numbers most practices are assembling by hand.
How a survey travels
Five things happen in order. When something is not working, the fastest way to find it is to ask which step it stopped at.
-
The patient has an appointment
Nothing is sent during or immediately after the visit.
-
The next day, CARE picks a survey
It reviews yesterday's appointments and checks your delivery rules to decide which survey each patient receives. Rules can key off provider, location or appointment type, so a post-op patient and a routine check-up get different questions.
-
The patient gets an email or text
You write both. Merge fields drop in the patient's name, the provider and the visit date.
-
They answer and submit
Star ratings, a nought to ten recommendation score, yes or no, multiple choice, or free text. Skipping is allowed and a skipped question is not counted against you.
-
You get scores, alerts and reports
Results land in the reports immediately. If the score was at or below your threshold, an alert email goes out within moments of submission.
What the patient answers, and what you get back
Short and mobile-first on their side. Filterable, exportable analysis on yours.
- Five question types: star ratings with an N/A option, a nought to ten recommendation scale, yes or no, multiple choice, and free-text comments
- One survey per appointment, so nobody is asked twice for the same visit
- A confirmation screen you word yourself, optionally steering happy patients to a public review site
- Static links for kiosks, QR codes or a website, separate from the after-visit send
- Follow-up reminders for patients who did not answer the first invitation
Questions we get about surveys
When exactly is a survey sent?
The day after the appointment. CARE reviews yesterday's visits and sends then, so if you switch surveys on today the first ones go out tomorrow rather than immediately.
What triggers a low-score alert?
Three things. The overall rating at or below your warning threshold, any single rating question at or below it if you have enabled that, or offensive language in a written comment even when the score was high.
What does NOT trigger an alert?
A skipped question is not counted at all. An N/A on a star question is stored as zero and ignored, unless you deliberately switch on counting zero as a low score.
Does the alert email contain patient details?
No. It carries the survey name, the date, the overall score, any individual questions that came in low, and a link to the full response. You follow the link to see who it was, which keeps patient identity out of an email inbox.
Can we alert on a specific answer instead of a score?
Yes, that is a question alert. Pick a question, an operator and a value, then set who is emailed. It catches what a score threshold cannot, such as a No on a yes-or-no question about whether the patient understood their discharge instructions.
Can different patients get different surveys?
Yes. Delivery rules select the survey by provider, location or appointment type, so a surgical follow-up and a routine visit can ask entirely different questions.
Does this help with MIPS reporting?
Yes. The MIPS quality measure report follows CAHPS for MIPS methodology on a one to ten scale, stratifies by age band, gender, location and provider, tracks the data-completeness and case-minimum requirements, and exports in a format built for CMS submission as a QCDR or registry.
Do we get a net promoter score?
Yes, calculated the standard way from the recommendation question: promoters score nine or ten, passives seven or eight, detractors nought to six. It is reported alongside the other measures rather than needing a separate survey.
MIPS reporting, without the spreadsheet
Patient experience is a MIPS quality category, and most practices assemble it by hand. CARE produces it as a report.
- Follows CAHPS for MIPS methodology on the standard one to ten rating scale
- Performance bands: exceptional at nine or above, above average at seven to eight, average at five to six
- Primary performance rate calculated as the percentage of responses rating seven or higher
- Stratified by age band (0-17, 18-64, 65+), gender, location and provider, as CMS requires
- Tracks data completeness against the 70 percent threshold and warns on case minimums
- Exports in a CMS-submission format for QCDR or registry reporting, with timestamps for the audit trail
Six reports, plus an AI layer
Everything filters by survey, location, provider and date range, and everything exports.
- Performance report: per-question breakdowns with counts and percentages
- Survey response tracker: who was invited, who answered, and who did not
- Survey details: individual responses, including the free-text comments
- Historical trends: whether scores are moving, and in which direction
- Net promoter and MIPS quality measure reporting
- AI analysis offering trend summary, negative watchlist, provider comparison, operational improvement, sentiment analysis and an experience scorecard
- The AI layer transmits no PHI, and every report exports to CSV or PDF
See it with your own survey questions
We will build one of your surveys live and show you the alert that fires when a score comes in low.
Talk to our team