Americans' confidence that their own doctor, nurse, or primary clinician will give them trustworthy public-health information sat at 87 percent in August, statistically unchanged from 88 percent in September 2024. [1] Confidence in the federal agencies those clinicians are supposed to work with did not hold. Trust in the CDC fell from 72 percent to 61 percent over the same span; the FDA from 73 to 59; the NIH from 74 to 61. [1]

That doctor-versus-institution split is the same one playing out in Lancaster County, where a coroner's forensic finding last weekend contradicted the state's measles-death classification and the CDC has now excluded both Pennsylvania deaths from its national dashboard. The Annenberg Public Policy Center poll puts a national number on a local fight this paper has treated as a documentary-chain problem, not a mood.

The drop in agency trust is dated. Annenberg has run the same question since April 2021. The collapse it reports is concentrated after February 2025, when Robert F. Kennedy Jr. became Health and Human Services secretary and began saying, on numerous occasions, that his job was to "restore faith" in federal health agencies. [1] Laura A. Gibson, an Annenberg research analyst, answered that claim with the survey itself: "We don't see evidence in our data that the Trump administration is restoring trust in the federal health agencies." [1]

The sample is 1,904 adults, with a margin of sampling error of plus or minus 3.2 percentage points at 95 percent confidence. [1] The doctor number is inside that band of the prior reading. The agency numbers are not. FDA confidence is now the lowest of the three, 14 points below its September 2024 mark. CDC and NIH each lost 11 and 13 points. Those are not rounding errors, and they are not a general collapse of medical authority. People still believe the person in the exam room. They have stopped believing the buildings in Atlanta, Silver Spring, and Bethesda.

Kennedy's restoration pitch treated distrust as a communications problem the new leadership would fix. The Annenberg series measures the opposite: the clinician relationship is stable; the federal brands are not. A reader who only sees X commentary about captured agencies misses that the public never withdrew trust from doctors. A reader who only sees CIDRAP's doctor-trust headline misses that the restoration project has now been scored, in the same survey, as a miss.

The Pennsylvania measles dispute is the operating illustration. Local forensic authority — a county coroner, a pathologist, a death certificate — is the layer the public still credits. The CDC dashboard is the layer it no longer does. The poll does not adjudicate which Lancaster death was measles-associated. It does show why a dashboard exclusion and a coroner's dissent land as the same story in two registers: one clinical, one federal, and only one of them still commands 87 percent.

-- NORA WHITFIELD, Chicago