Quality & Stars

CAHPS (Consumer Assessment of Healthcare Providers and Systems)

~10 min readBy Pelica Health teamPublished August 28, 2026Updated August 28, 2026Reviewed by Catherine Zhao

Short answer

CAHPS is the patient experience survey CMS fields each spring on a sample of Medicare Advantage and Part D members. Nine Star Ratings measures come from it: six Part C measures and two Part D measures at weight 2 each, plus the Annual Flu Vaccine measure at weight 1. Since the 2026 Star Ratings, the patient experience, complaints, and access measures carry a weight of 2, down from 4. A survey fielded in spring 2026 lands in the 2027 Star Ratings, so the work that moves it happens more than a year before the rating publishes.

What CAHPS is

CAHPS is the survey CMS uses to measure what members experienced with their health plan. The Medicare version has been fielded annually since 1998, sponsored by CMS and developed with AHRQ, and it exists because Congress required it in the Balanced Budget Act of 1997 and the Medicare Modernization Act of 2003.

Three questionnaires go out: MA-Only, MA-PD, and PDP. They share a core set of questions and differ in the drug-plan sections. Members answer about their own experience over the previous six months, not about a clinical event, and the responses roll up to composite and rating measures at the contract level.

Nine of those measures feed the Part C and Part D Star Ratings, which for Medicare Advantage contracts determine quality bonus payments. That is why a survey about how easy it was to get an appointment ends up in a revenue forecast.

Which Star measures come from CAHPS

Nine Star Ratings measures are calculated directly from survey responses. Eight sit in the patients' experience and complaints category at weight 2. One, the flu vaccine measure, is classified as a process measure at weight 1.

Measure	Part	Weight	2026 Stars national average
Getting Needed Care	C	2	81
Getting Appointments and Care Quickly	C	2	84
Customer Service	C	2	90
Rating of Health Care Quality	C	2	87
Rating of Health Plan	C	2	87
Care Coordination	C	2	87
Rating of Drug Plan	D	2	87 (MA-PD)
Getting Needed Prescription Drugs	D	2	90 (MA-PD)
Annual Flu Vaccine	C	1	66%

The composite and rating measures are reported on a 0 to 100 scale. Annual Flu Vaccine is a straight percentage of sampled members who report getting the shot.

Two measures in the same weight category are commonly mistaken for CAHPS results and are not. Complaints about the Health Plan comes from the Complaints Tracking Module, and Members Choosing to Leave the Plan comes from disenrollment data. Both are administrative. Neither moves because a member filled out a survey.

The survey also produces measures reported back to contracts but never published: Doctors Who Communicate Well, Pneumonia Vaccine, and reminders to fill prescriptions and take medications. Those are useful diagnostics and carry no Star weight.

The weight change: 4 down to 2

CMS raised the weight of patient experience, complaints, and access measures from 2 to 4 for the 2023 Star Ratings, then reversed it. The 2026 Star Ratings Technical Notes list the change first in their summary of what moved: "The weight of the patient experience, complaints, and access measures was decreased from 4 to 2." That took effect with the 2026 Star Ratings and still holds for 2027.

Weight on patient experience, complaints, and access measures since the 2026 Star Ratings, down from 4.

Star Ratings measures calculated from CAHPS survey responses across Parts C and D.

~800

Target sample per Medicare Advantage contract. PDP contracts sample about 1,500.

Halving the weight did not make the survey unimportant. Eight measures at weight 2 is still a large block of a Part C summary rating. What changed is the ranking against everything else: intermediate outcome measures at weight 3 now sit above patient experience, so a plan buying experience gains at the price of clinical performance pays more for that choice than it used to.

How the survey is actually run

The timing is the part most teams get wrong, so it is worth walking the calendar. These are the 2026 administration dates from the CMS Quality Assurance Protocols.

Step	2026 date
CMS draws the sample from the Integrated Data Repository	January 2026
Vendors download the sample file	February 6
Pre-notification letter to every sampled member	February 25
Web survey invitation by email	March 2
First paper questionnaire mailed	March 10–11
Second questionnaire to non-respondents	March 30–31
Telephone follow-up begins	April 22 – May 2
Web cutoff, phone interviewing ends	May 30
Cutoff for returned mail surveys	June 1
Final data files submitted to CMS	June 9–11

Who gets sampled is equally fixed. Members must be at least 18, living in the United States, and continuously enrolled in the same contract for at least six months at the January sample draw. Contracts with 600 or more eligible enrollees as of July of the prior year must field the survey; those between 450 and 599 may field it optionally. Every contract surveys in English and Spanish, and may add Chinese, Korean, Tagalog, and Vietnamese.

Since 2024 the protocol is web-mail-phone rather than mail-phone. CMS pilot-tested adding a web component in 2022 and found it lifted response rates. A vendor that misses the submission deadline hands its contract an automatic 1 star on every CAHPS measure, which makes vendor oversight a compliance item rather than a performance one.

How CAHPS measures get their stars

CAHPS does not use the clustering algorithm that sets cut points for clinical measures. It has its own method, relative distribution and significance testing, described in the Star Ratings Technical Notes.

To earn 5 stars, a contract's score has to rank at or above the 80th percentile of contracts and be statistically significantly higher than the national average, and it must either not be flagged as low reliability or sit more than one standard error above the 80th percentile. To land at 1 star, the score has to rank below the 15th percentile and be statistically significantly below the national average. The technical notes publish base group cut points for reference, but the significance test is what settles the final assignment.

The practical consequence is that CAHPS stars move with the field. A contract can hold its raw score flat and lose a star because other contracts improved. That behavior is different from a clinical measure cut point set by Tukey outlier deletion and mean resampling, and it is worth explaining to anyone who reads the two side by side.

Case-mix adjustment

Raw scores are not comparable across contracts, because who answers a survey predicts how they answer it. CMS adjusts every CAHPS measure for eight respondent characteristics outside the plan's control: administrative age, dual eligibility status, the low-income subsidy indicator, use of an Asian language survey, and self-reported education, general health status, mental health status, and proxy usage.

The coefficients are small per characteristic and they run in both directions. In the 2026 Technical Notes, respondents aged 75 to 79 rate their plan 0.0587 points higher than otherwise similar 70 to 74 year olds on Rating of Health Plan, so a contract skewed toward that age band gets adjusted down. Respondents who had proxy help score 0.0834 points lower on the same item, so a contract with more of them gets adjusted up. Coefficients are re-estimated every year from that year's data, and the patterns are not consistent across measures.

Why you cannot fix CAHPS in Q4

Every other Star measure has some version of a year-end push. CAHPS does not. The reason is structural, and it has five parts.

The sample is drawn in January. Who gets surveyed is decided from enrollment data at the start of the year and requires six months of continuous enrollment. A member who joined in October is not eligible.
The recall window has already closed. Questions ask about the last six months. A survey completed in April 2026 is scoring roughly October 2025 through April 2026. The flu question reaches back further still, to July 1 of the prior year.
Fielding ends June 1. After the mail cutoff there is nothing left to influence. The data goes to CMS in the second week of June.
Results arrive in August, for a year that is over. CMS sends contract-level CAHPS results to each Medicare Compliance Officer in August. By then the measured period ended three months ago.
The rating publishes a year later. Data collected in spring 2026 lands in the 2027 Star Ratings, which is what drives the corresponding quality bonus payment. The lag between the member experience and the money is roughly two years.

Work that changes a CAHPS score is work done in the second half of the prior year and the first quarter of the survey year: appointment availability, call center handling, formulary and pharmacy friction, and whether anyone followed up after a hospital stay. A campaign launched in October is aimed at a survey that will not be fielded for five months and will not be published for two years.

Common mistakes

The mistake	What is actually true
Counting Complaints about the Health Plan and Members Choosing to Leave as CAHPS results	Both are administrative measures in the same weight category, from complaint and disenrollment data
Treating the weight as still 4	It dropped to 2 with the 2026 Star Ratings and remains 2 for 2027
Assuming a flat score keeps a flat star	Stars are assigned by percentile rank plus significance testing, so the field moving is enough to move a star
Comparing raw scores across contracts	Published measure scores are case-mix adjusted for eight respondent characteristics
Reading spring survey results as current-year performance	The survey scores the previous six months and feeds the following year's Star Ratings
Contacting members about the survey during fielding	Survey administration is governed by CMS protocols and vendor rules; plan communication is constrained

How Pelica handles the work behind CAHPS

Nothing closes a CAHPS gap directly, because there is no gap to close. What the survey measures is whether the member's year went smoothly: whether the appointment happened, whether the prescription was ready, whether anyone called after the discharge, and whether the same question had to be asked three times.

Pelica works the operations underneath that. One live record per member across claims, EHR, pharmacy, lab, and ADT means the care team is not reconstructing history on the phone. Copilots handle refill follow-up, transitions after a discharge, and quality outreach so the member gets one coordinated contact instead of four from four departments. Customers hold 96%+ adherence, close 70%+ of transitions-of-care gaps within 30 days, and improve gap closure by 41%.

Related terms

HOS is the other Medicare member survey, run on a two-year cohort design and scoring health outcomes rather than experience. Cut points explains how clinical measures get their thresholds, which is a different method from the one CAHPS uses. Triple-weighted measures covers the weight-3 measures that now outrank CAHPS, and D-SNP explains why dual status is a case-mix adjuster in the first place.

Sources

Frequently asked questions

Common questions about the Medicare CAHPS survey, the measures it produces, its weight in the Star Ratings, and its calendar.

What is the CAHPS survey?

CAHPS stands for Consumer Assessment of Healthcare Providers and Systems. The Medicare version is a survey CMS has fielded annually since 1998 to measure what members experienced with their health or drug plan over the previous six months. It exists because Congress required it in the Balanced Budget Act of 1997 and the Medicare Modernization Act of 2003. Three versions go out: MA-Only, MA-PD, and PDP.

Which Star Ratings measures come from CAHPS?

Nine. Six Part C measures at weight 2: Getting Needed Care, Getting Appointments and Care Quickly, Customer Service, Rating of Health Care Quality, Rating of Health Plan, and Care Coordination. Two Part D measures at weight 2: Rating of Drug Plan and Getting Needed Prescription Drugs. And the Annual Flu Vaccine measure, a process measure at weight 1. Complaints about the Health Plan and Members Choosing to Leave the Plan sit in the same weight category but are administrative measures, not survey results.

What weight do CAHPS measures carry in the Star Ratings?

Weight 2. CMS raised patient experience, complaints, and access measures from 2 to 4 for the 2023 Star Ratings, then reversed the increase. The 2026 Star Ratings Technical Notes list the change as: the weight of the patient experience, complaints, and access measures was decreased from 4 to 2. That took effect with the 2026 Star Ratings and remains weight 2 for the 2027 Star Ratings. The Annual Flu Vaccine measure is separate, classified as a process measure at weight 1.

When is the Medicare CAHPS survey conducted?

Each spring. For the 2026 administration, CMS drew the sample in January, vendors mailed the pre-notification letter on February 25, the web invitation went out March 2, the first paper questionnaire went out March 10 and 11, telephone follow-up ran from April 22, and the cutoff for returned mail surveys was June 1. Final data files were due to CMS between June 9 and 11.

Who gets the CAHPS survey?

A random sample of members at least 18 years old, living in the United States, and continuously enrolled in the same contract for at least six months at the January sample draw. Medicare Advantage contracts sample about 800 members and standalone Part D contracts about 1,500. Contracts with 600 or more eligible enrollees as of July of the prior year must field the survey; contracts with 450 to 599 may field it optionally.

How are CAHPS star ratings calculated?

By relative distribution and significance testing, not by the clustering method used for clinical measures. For 5 stars a contract's score must rank at or above the 80th percentile of contracts and be statistically significantly above the national average, and must either not be flagged as low reliability or sit more than one standard error above the 80th percentile. For 1 star the score must rank below the 15th percentile and be statistically significantly below the national average. Because the assignment is relative, a flat score can still lose a star if other contracts improve.

Why can't a plan improve CAHPS at the end of the year?

The sample is drawn in January and requires six months of continuous enrollment, so who answers is already decided. The questions ask about the previous six months, so the experience being scored has already happened. Fielding closes June 1, results reach each contract's Medicare Compliance Officer in August, and the data feeds the following year's Star Ratings. Work done in Q4 is aimed at a survey that will not be fielded for another five months.

Is CAHPS the same as HOS?

No. Both are Medicare member surveys, and both feed Star Ratings measures, but they measure different things on different schedules. CAHPS is fielded every spring on a fresh cross-sectional sample and measures experience with the plan and with care. HOS follows the same cohort of members over two years and measures self-reported health outcomes, including whether physical and mental health held up better or worse than expected.

Nothing closes a CAHPS gap, because there is no gap to close.

What the survey scores is whether the appointment happened, the prescription was ready, and someone called after the discharge. Pelica runs those operations off one live record per member, with copilots that handle refill follow-up, transitions, and quality outreach as a single coordinated contact. Customers hold 96%+ adherence, close 70%+ of transitions-of-care gaps within 30 days, and improve gap closure by 41%.

Book a Demo → See the Quality & Pharmacy copilot