SKUPIE RESEARCH · WHITE PAPER No. 5
Field note
Measles in 2026: the numbers are public. The parents are not reading them.
Why a record year for a preventable disease is also a record year for the gap between what public health knows on Thursday and what a family hears, and what an amplification layer can and cannot do about it.
Summary
The United States has recorded more measles cases in 2026 than in any year since 1991. Nearly every case is in a person who was not vaccinated or whose status is unknown; well over half are children and teenagers; about one in nine has been hospitalized; children have died in three states for the first time in decades. None of this is secret. CDC posts it every Wednesday, state health departments post it daily, and the research community has described the cause (falling MMR coverage) with precision for years.
This paper is not a call to action. Skupie™ does not tell researchers what to study or parents what to decide. It makes one narrower argument: the measles data of 2026 is a case study in a last-mile failure. The surveillance works; the publishing works; the reading does not happen. We describe where the data sits, how it is written, who it reaches, and what an amplification layer (a product that summarizes accredited sources, dates them, and puts them where a parent already looks) adds and does not add. We close with what we have built, what we have not, and what we would ask of the people who produce the data.
The 2026 measles numbers did not fail to exist. They failed to arrive.
What the sources say (as of 13 November 2026)
Every figure below is a summary of a cited source. Where two accredited sources differ we show both; we do not reconcile them. Figures change weekly; the dated version on the Skupie website and in the app is the one to cite.
| Measure | Figure | Source and date |
|---|---|---|
| Confirmed cases, 2026 | 5,412 | CDC Measles Cases and Outbreaks, data as of 12 Nov 2026 |
| Outbreaks (3+ linked cases), 2026 | 43 | CDC, 12 Nov 2026 |
| Outbreak-associated share | 96% (5,216) | CDC, 12 Nov 2026 |
| Jurisdictions reporting cases | 47 (18 further cases among international visitors) | CDC, 12 Nov 2026 |
| Comparison: all of 2025 | 2,289 cases, 48 outbreaks | CDC, 2025 final |
| Last year with more cases | 1991 | CDC historical series, as reported 7 Nov 2026 |
| Hospitalized | 11% of cases | CDC via CIDRAP, Nov 2026 (7% in July; 11% in 2025) |
| Unvaccinated or unknown status | 96% of cases | CDC via CIDRAP, Nov 2026 (93% in July) |
| Age: under 5 · 5–19 | 18% · 41% (59% under 20) | CDC via Al Jazeera, 7 Nov 2026 |
| Deaths | CDC: 4 (7 Nov). States: 6 in Pennsylvania; 1 in New York and 1 in Wisconsin; press tallies 8–10 | CDC and state health departments; CDC and CSTE are writing a standard measles death definition, so the counts differ by design, not by error |
| Largest state outbreaks | PA 1,487 · WI 477 · OH 311 · NY 210 · KY 204 · MN 139 · IN 134 | State health departments, 13 Nov 2026 (CDC counts lag states by days to weeks) |
| Kindergarten MMR coverage | 92.4% (2025–26), down from 95.2% (2019–20) | CDC SchoolVaxView; 95% is the coverage generally cited as needed to stop sustained spread |
Three things stand out for a parent-facing reader, and we keep returning to them in the For Parents companion: the cases are concentrated (96% in outbreaks), they are concentrated among the unvaccinated (96%), and they are concentrated among children (59% under 20, 18% under five, the group most likely to be hospitalized and least able to be fully vaccinated).
Where the data lives, and who it is written for
CDC's measles page is excellent and is written for epidemiologists, clinicians and journalists. It is updated on Wednesdays. State pages are updated daily during outbreaks and vary in format from a dashboard (Pennsylvania, Wisconsin) to a press release (several states) to a PDF table. MMWR carries the analytic reports. None of these is wrong. All of them assume the reader will come to them.
A parent does not come to them. In our August 2026 measurements (White Papers 1 and 2) a parent with a sick child reached an accredited answer in a minority of attempts, and reached a dated one in fewer still. During an outbreak the problem inverts: the parent is not searching at all, because nothing has told them to. The school nurse learns of a local case from the health department; the parent learns of it from a group chat, a day later, with the numbers wrong.
- Latency. CDC's Wednesday count lags the states by days to weeks (Wisconsin: state 477, CDC 419 on the same day). A parent reading a national figure is reading last week.
- Register. "Outbreak-associated" and "importation" are precise and opaque. "Nearly every case was caught from someone nearby, not from a traveler" is the same fact.
- Fragmentation. Fifty state formats, no common feed. The researcher who wants a comparable table builds it by hand; so does the journalist; so would the parent, and does not.
- Silence by default. No accredited source pushes. A parent who wants to know whether measles is in their county must already know to ask.
The sources are built to be consulted. Families need something that is built to be noticed.
What an amplification layer is, and is not
Skupie's position has not changed since White Paper 3: we do not correct accredited data. We summarize it, and when sources differ, we note the difference. Nothing in this paper argues for new science, new guidance or new authority. It argues for a last mile.
| An amplification layer does | It does not |
|---|---|
| Reads the accredited sources on their schedule (CDC Wednesdays, states daily) and records the date on every number it shows. | Produce its own counts, models or forecasts. |
| Rewrites at a parent's reading level with the source's meaning intact, and links back to the source. | Advise. It does not say "vaccinate", "keep your child home" or "see a doctor"; it shows what CDC and the state say and who to ask. |
| Shows the difference when two accredited sources disagree (a state and CDC count; a state and CDC death definition). | Pick a winner. |
| Puts the summary where the parent already is: the school-year app on the phone, What's Going Around for the county, the Measles Fact Sheet one tap away. | Replace the health department's notice, the school's letter or the pediatrician's call. |
| Tells the parent what is unchanged, so a record year does not read as an emergency in every household. | Create urgency it cannot source. |
The last row matters most. The most useful sentence in a parent-facing outbreak summary is often the one about what has not changed: the MMR schedule CDC publishes is the same schedule it published last year; a child with two documented doses is in the 97%-protected group CDC describes; a family with no cases in the county has nothing to do today except know where to look. Amplification without that sentence is alarm.
For researchers and public health: what would make the last mile shorter
We are not in a position to tell the people who produce this data what to do, and we do not. These are the things that, from the reading end, would make a faithful summary easier to produce and easier to trust. Each is a request, not a finding.
- A dated, machine-readable national count. CDC's table is an HTML page. A CSV or JSON with the same "as of" date, one row per jurisdiction, would let every amplifier show the same number on the same day. (CDC's data.cdc.gov already does this for other conditions.)
- State counts with a stated reporting lag. "477 cases, reported through 12 Nov, CDC will reflect by 19 Nov" removes the appearance of contradiction that a parent reads as someone being wrong.
- One death definition, published. CDC and CSTE are writing it. Publishing the definition with the first count under it will end the 4-versus-10 problem, which is a definitional difference being read in public as a dispute.
- Age and vaccination breakdowns at the state level. The national 96% and 59% are the two numbers a parent most needs; most states publish neither.
- Plain-language one-liners alongside the technical table. The health departments that already do this (several state outbreak pages carry a "what this means for you" box) are the ones whose text an amplifier can carry verbatim, which is the safest case for everyone.
Skupie is approaching federal and state sources this month to ask for exactly these things, in writing, as part of its source-access program; this paper is the reasoning behind that letter.
Where we stand, honestly
- Built
- 41 Fact Sheets including measles, each summarized from a cited accredited source and dated; What's Going Around (county-level signal); the Research series; the source registry behind this paper.
- Built, not yet launched
- The app itself is in pre-launch testing on Android; iOS follows Apple enrollment. Nothing in this paper is in a parent's hand today.
- Not built
- Daily automated re-reading of state outbreak pages (the watcher is designed, scheduled for December); a push notice when a county signal changes; Spanish for the For Parents companion (planned).
- Not ours to build
- The data. Every number here is CDC's or a state's, and will be wrong the week after this is published unless re-read. The dated version on skupie.com/research is the one that will be kept current.
Sources
- CDC. Measles Cases and Outbreaks. cdc.gov/measles/data-research. Data as of 12 Nov 2026 (5,412 cases; 43 outbreaks; 96% outbreak-associated; 47 jurisdictions). 2025: 2,289 cases, 48 outbreaks.
- CIDRAP, University of Minnesota. "US measles total for 2026 tops 5,100 cases" (Nov 2026): 11% hospitalized; 96% unvaccinated or unknown; 59% under 20; state counts PA, WI, OH, NY, KY, MN, IN. And "US adds 53 more measles cases to its 2026 total" (July 2026): 7% hospitalized; 93% unvaccinated or unknown; 20% under 5, 49% 5–19.
- Al Jazeera, 7 Nov 2026, citing CDC: 5,182 cases; 4 CDC-counted deaths; Pennsylvania 1,487 cases and 6 deaths; under-5 18%, 5–19 41%; kindergarten MMR 92.4% (2025–26) vs 95.2% (2019–20); highest since 1991.
- Healio, 9 Nov 2026: New York and Wisconsin infant deaths (first in 30 and 36 years); Pennsylvania's sixth death.
- 2 Minute Medicine, Nov 2026, citing CDC as of 12 Nov: 5,412 cases; CDC/CSTE standard measles death definition in progress.
- Skupie Research White Papers 1–4 (skupie.com/research), Aug–Nov 2026, for the findability measurements and the summarization principle.
Cite this paper
Skupie™ is an awareness and education product. It does not provide medical advice, diagnosis or treatment.
© 2026 Skupie Inc. Share and quote for non-commercial purposes with attribution to Skupie Research and a link to skupie.com/research. Skupie™ is a trademark of Skupie Inc.