Loading today’s stories
Science1 distinct publisher3 min readPublished
About one pregnancy in ten now carries the diagnosis, and the specialists quoted say the count is at once an undercount and partly a product of better detection, which is what makes the slope hard to read.
The Scientist · Science desk
product
Three Salmonella Outbreaks, One Failure Mode: Nobody Knew Where the Ingredient Went1 distinct publisher
science
Kindergarten shot records now name the exposed states: Idaho off 15 points, Maine up three1 distinct publisher
science
Cyclosporiasis cases are easing, but one state's count is 95% of the national total1 distinct publisher
science
US life expectancy hit 79 in 2024, and the cushion over 2019 is about 73 days1 distinct publisher
Compiled by The ScientistSomething wrong?How this is made
Kilpatrick describes preeclampsia as a systemic response to the placenta, with an abnormal rise in a protein called sFlt-1, which slows placental growth, appearing to be the trigger [16]. Blood pressure is the downstream signal that happens to be cheap to measure, which is why prenatal care organizes itself around it. Diagnosis then requires both readings and evidence of organ damage in urine or blood [9], so the diagnosis is an event that has to occur while someone is looking.
Spread across the eight years, the increase compounds at roughly 7.1 percent a year [1]. If the one-in-ten figure [2] describes 2024, dividing by 1.73 puts 2016 near 5.8 percent, about one pregnancy in 17 [2]. Each cohort has been slightly worse than the one before it for eight years running, rather than one bad year dragging an average.
Two statements in the reporting pull against each other, and they constrain different quantities. Sibai's incomplete-certificate problem [13] concerns the level: the true rate is above the recorded one. Kilpatrick's point that some of the increase reflects better recognition and diagnosis [18] concerns the slope: if the instrument improved over the window, the biological rise is smaller than the recorded one. Nobody here quantifies the split. The candidate drivers Kilpatrick lists, older first pregnancies, more births after 40, more IVF multiples, rising BMI [6], are each plausible on their own [7] and none is decomposed into a share of the change.
The single case in the account does one job and no more: it shows that a normal reading days earlier does not exclude onset [4]. It cannot tell you how often that path is taken, because one case has no denominator. What would settle the surveillance question is the fraction of severe cases that declare themselves between scheduled visits, and the reporting leaves that fraction uncounted.
The postpartum window is thinner still. Hourly checks in the hours after delivery [5] describe an inpatient shift, not a discharge plan. Sibai says gestational hypertension, once diagnosed, is supposed to be monitored carefully in the woman's interest and not only the fetus's [14], and that emphasis has moved toward reducing maternal morbidity [15]. Given that blood pressure problems in pregnancy are among the leading causes of illness and death in American mothers [10], the measurement that matters is what happens to surveillance after a patient goes home, and the reporting stops short of that.
The 2023 approval of a test based on the sFlt-1 ratio [17] is genuinely interesting for that reason: a biochemical trigger does not depend on a pressure spike arriving during office hours. What's missing is how the ratio behaves in routine use, at what threshold, with what rate of false alarms, and whether a positive result changes management when the next appointment is a fortnight out. I would still rather have it than not, conditional on those numbers existing somewhere.
The unglamorous fix is the one Sibai names: record how many weeks the pregnancy reached before birth [14]. Until the trend line carries gestational age at delivery, it can report that the diagnosis is more common while staying silent on whether the women it counts are getting sicker.
Ranked by verification strength, evidence, and original report placement.
According to the U.S. Centers for Disease Control and Prevention, the rate of high blood pressure in pregnancy increased by 73 percent between 2016 and 2024.
That rate is about one in 10 women.
Cases of gestational hypertension, high blood pressure occurring for the first time during pregnancy, have been rising since 1989, the first year the condition was recorded on birth certificates.
In 2019, Megan O'Grady was 32 weeks pregnant when she arrived at hospital with a blood pressure of 220/120 mm Hg and severe preeclampsia; at a doctor visit a few days earlier her blood pressure had been completely normal, and she gave birth more than a month before her due date.
O'Grady recalls that hospital staff had to check her blood pressure every hour after her son was born.
As pregnancies progress, gestational hypertension increases the risk of preeclampsia, which also leads to liver and kidney problems.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · September 2, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One magazine, one federal count, two specialists
The headline slope rests entirely on Scientific American's relay of a CDC report that is never linked, quantified in absolute cases, or dated at the one-in-ten end. What raises the floor is that the two specialists quoted are named, credentialed and on the record, and the 2023 approval is a checkable event; what keeps the ceiling low is that the causal story, the disparities and the possible undercount are all single-voice assertions with no supporting numbers.
Nothing here counts uptake
We can say a test was approved in 2023 and that the field's emphasis has shifted toward maternal morbidity. We cannot say how many clinics run the sFlt-1 ratio, how often, or with what effect, because the reporting never touches volumes, coverage or protocols. Inferring uptake from an approval date would be invention.
The percentage travels further than its denominator
A 73 percent rise is a headline that survives without context; the base rate it grew from, the year the one-in-ten figure describes, and the share attributable to better detection are all missing, and only the last is even acknowledged. That said, this is a modest gap rather than a large one, because Scientific American publishes its own undercutting caveats instead of hiding them, and the clinical stakes it describes are not exaggerated.
A disclosed stake in the test being praised
The specialist who explains the sFlt-1 mechanism and the 2023 test also helped get that test approved, and Scientific American says so in the same breath, which is the honest handling. The pull is real but declared, and it points in a predictable direction: the expert best positioned to frame the trend as a detection story also has standing in the detection technology. A subscription appeal sits mid-article, a routine commercial interest worth naming.
Direction firm, magnitude soft
That pregnancy hypertension is rising and that it kills is about as settled as this reporting can make it; the two specialists agree on the direction and one of them describes managing the consequences at the bedside. The size of the rise is another matter. When the same story tells you the count is too low and that part of the increase is better looking, 73 percent is a number to hold loosely.