Labora Rounds · Newborn care

More healthy babies are going to intensive care.

Full-term. Normal birth weight. One baby, not twins. These are the newborns nobody expects to need a NICU — and between 2016 and 2024 the share of them admitted to one rose in 46 of 51 states. In Mississippi it nearly doubled.

Every number here comes from federal birth records that anyone can download. Nobody had put them together this way. So we did — all 51 states, all nine years. Find yours.

Healthy full-term newborns admitted to intensive care
2024  · 4.50 per 100 births nationally
2016 – 2024
Lower
Higher

Each square is one state, placed roughly where it sits on the map so small states stay visible. Click one to hold it. The colour scale is fixed across all nine years, so the map darkening over time is a real change, not a rescaling.

Mississippi

+86.8%
Rate per 100 births · 2016–2024
3.81
Per 100 in 2024
966
Babies, of 25,334

The part that should not be happening

The babies who were always going to need a NICU barely moved. The ones who weren’t went up three times as fast.

If newborn intensive care were simply expanding across the board, every group would rise together. It didn’t. Among babies born before 37 weeks — where roughly half go to a NICU and always have — the rate rose 5% over nine years. Among healthy full-term babies it rose 17%. Each panel below is on its own scale, because the levels are nowhere near each other; the percentage is what you compare.

National, pooled across all 51 jurisdictions. “Healthy full-term” is the cohort mapped above. “39–40 weeks only” is the same filter narrowed to the two lowest-risk weeks of gestation — it rises too, by 8.7%, which is the tightest version of the same finding.

The nine-year change

Forty-six states up, five down — and the biggest moves are in the states with the least maternity care left

Change in the share of healthy full-term newborns admitted to intensive care, 2016 to 2024. The national rate went from 3.84 to 4.50 per 100 — a 17.2% rise. Twenty-eight of the 51 moved faster than that.

What this is, and what it isn’t

This measures care delivered, not babies harmed.

A NICU admission is a decision made in a hospital. When more full-term, normal-weight, single babies get admitted year after year, there are several things it could mean, and this data cannot tell them apart:

  • More newborns are genuinely sicker at birth.
  • Hospitals are screening harder and catching more — the same babies, seen better.
  • Staffing is thinner on the regular postpartum floor, so borderline babies go up a level.
  • A hospital with a NICU has a financial reason to fill it.

We are not claiming to know which. We are saying the line moved, in nearly every state, in the same direction, over nine years — and that a change this size in who gets separated from their mother after birth deserves a name and an explanation. That is the question we think somebody should be asking.

Read the caveats first

Rhode Island’s series breaks. Its all-births rate falls from 14.7 to 4.3 and back — that is a change in how the state fills in the form, not in its hospitals. Treat it as unusable. Wyoming, Hawaii and Vermont are small enough that single-year moves are noisy.

The NICU box is a birth-certificate field. States revised their certificates on different schedules, and a state that improves its reporting will look like a state whose admissions rose. We show nine consecutive years precisely so a one-year jump is visible as a jump.

The denominator excludes blanks. We count only births where the field was actually completed — between 96% and 99% of births nationally each year, and no lower than 80% in any single state-year.

Methods

How to rebuild this yourself

Everything below is public and free. There is no proprietary input to this figure — the work was deciding what to ask for, and then asking for it nine times.

SourceCDC/NCHS Natality files, birth years 2016–2024, accessed through CDC WONDER Natality (wonder.cdc.gov/natality.html).
MeasureBirth-certificate item Admission to NICU, coded Yes / No / Unknown.
Cohort — “healthy full-term”Singleton birth (plurality = 1), 37–41 completed weeks of gestation, birth weight 2,500–3,999g. This is a filter, not a clinical judgment: it removes the newborns who are expected to need intensive care so that what remains is the group whose admission is a choice.
Cohort — “all births”Every live birth in the state-year, no restriction.
DenominatorBirths with the NICU field completed (Yes or No). Unknown and blank are excluded rather than counted as No.
RateAdmissions ÷ completed-field births × 100. National figures are pooled from the state counts, not averaged across states.
GeographyState of residence. 50 states plus the District of Columbia. Territories excluded.
Pulled11 August 2026.

Cite as: Labora Rounds, Neonatal intensive care admission among healthy full-term newborns, United States 2016–2024. Derived from CDC/NCHS Natality via CDC WONDER; pulled 11 August 2026. Reuse it, check it, argue with it — that is what it is for.

This is one measure out of thousands we watch.

Labora Rounds runs automated pipelines against federal and public data on women’s health, every day, and writes down what changes. Neonatal care is one thread. Hospital closures, obstetric workforce, coverage, and policy are others.

See the engine The county map The Record