Science1 distinct publisher3 min readPublished
Writing in STAT, two forensic and reproductive psychiatrists argue that liability fear will drive general psychiatrists and OB-GYNs away from new mothers, a prediction resting on one workforce number and on what trainees have told them.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The mechanism on offer is an asymmetry in what gets noticed. An admission that was not needed produces a mother separated from her infant and a line in a chart; a missed postpartum psychosis, which the authors describe as usually a medical emergency warranting hospitalization, can produce a trial [10]. Clinicians following this case are being shown the second failure mode in detail and the first hardly at all.
That makes the predicted direction plausible, though the piece supplies no rate to attach to it. Its behavioural evidence is that trainees have told the authors they are worried about their future practices, and that colleagues have asked whether the public will now regard virtual visits as inappropriate for maternal mental health care [7][9][17].
The one hard quantity is the workforce. About 500 reproductive psychiatrists practise in the United States [1], which works out to roughly ten per state if they were spread evenly, and they are not [15]. The subspecialty's usual role is consultative: answering diagnostic questions and, more often, medication safety questions from general psychiatrists [13]. This is a channel built for second opinions, and absorbing a caseload that other clinicians decide to stop carrying is a different scale of job altogether.
Base rates matter for the second half of the argument. Anxiety, obsessive-compulsive symptoms and mood symptoms are the common perinatal presentations, and psychosis is much less common [10]; intrusive thoughts and depressive symptoms usually do not require hospitalization [11]. A threshold recalibrated to catch the rare emergency therefore does most of its work on the frequent presentations, which is why the authors expect unneeded admissions and inappropriate child protective services referrals, and expect mothers to respond by reporting less [12]. How much less is the open question here. Non-disclosure leaves no claim line and no chart entry, so the outcome they are most worried about is also the hardest one to find in any dataset.
What would be measurable: perinatal psychiatric visit volumes by specialty, the telehealth share of those visits, and admission rates for perinatal patients, tracked across the publicity window. None of that appears here, and it is worth saying plainly that this is an opinion piece by two clinicians, not a study.
They also have a position in the remedy. The authors argue for awareness and education rather than defensive medicine, and they helped build the National Curriculum in Reproductive Psychiatry along with dozens of colleagues, and led an American Academy of Psychiatry and the Law practice resource [14]. That is a disclosed interest rather than a flaw in the reasoning, but it shapes which of the available responses gets recommended.
Where I land: the supply figure is the load-bearing fact, and it holds independently of how the trial ends [1][2]. Whether liability fear moves generalist behaviour at a scale that matters is an empirical question nobody in this piece has measured, and defensive practice tends to appear in scheduling patterns well before anyone writes a policy about it.
Ranked by verification strength, evidence, and original report placement.
Only about 500 reproductive psychiatrists are practising in the United States.
As a result of the small subspecialty workforce, general psychiatrists (as in the Clancy case) and OB-GYNs are often the frontline providers for perinatal mental illnesses; both are trained to assess and treat these disorders but do not have a dedicated focus on them.
The criminal case was with the jury at the time of publication; Clancy and her ex-husband have also taken civil action; and the criminal case, civil action and media coverage have all focused on the alleged failures of the clinicians Clancy saw.
The authors describe the clinicians at issue as those whom Clancy saw before killing her three children.
The authors report that, anecdotally, trainees have expressed concerns to them about their future practices.
The defense has criticised aspects of Clancy's care, suggesting the treating psychiatrist missed diagnoses, that telehealth visits are inherently inferior, and that inappropriate medication changes were made.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 28, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
Text watermarks land on 2 December. The detection they imply does not.1 distinct publisher
science
Four years of Dobbs data: one prediction in four survived contact with the numbers1 distinct publisher
science
"A beautiful mouse story, sold as human advice": muscle scientists contest protein restriction1 distinct publisher
science
Specialists outbilled primary care doctors on Medicare's new relationship-care add-on1 distinct publisher
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 column, one number
Every fact in this story comes from a single signed STAT opinion piece, and inside it exactly one figure does any work: roughly 500 reproductive psychiatrists in the United States, attributed to no count or registry. The clinical background is credible expertise stated without citation. The causal chain that gives the piece its point — courtroom scrutiny, then clinicians backing away from new mothers — rests on what trainees and colleagues have said to the two authors, which they label anecdotal themselves.
Nothing counted
The behaviour the authors are worried about is precisely what nobody in this story has measured: no referral volumes, no telehealth visit counts, no hospitalization or child-protective-services rates, and no before-and-after on which clinicians accept perinatal patients. Two teaching artefacts are named — a national curriculum and a professional practice resource — but the piece says who built them, never who uses them.
Prediction outruns the record
The dek-level worry — generalists walking away from mothers, hospitals over-admitting them, virtual care falling out of favour — is presented as a live consequence of an ongoing trial, while the observation underneath it is two psychiatrists recalling anxious trainees and inquiring colleagues. STAT flags the piece as opinion and the authors hedge with 'raises the possibility,' which is honest. Still, the distance between 'colleagues asked us' and 'mothers will find treatment even harder to obtain' is the entire argument, and nothing in this reporting crosses it.
The remedy is the authors' own work
The bylines disclose the interest plainly rather than hiding it: a forensic psychiatrist who is president-elect of the American Academy of Psychiatry and the Law and a past president of the same body, both subspecialists in the field the piece argues deserves wider reach, and both credited with the National Curriculum in Reproductive Psychiatry and the AAPL practice resource they recommend as the answer. Arguing for education over defensive medicine is a defensible position; it also happens to be an argument for the curriculum they wrote, and no independent party in this story seconds it.
Clear about itself, unverifiable in substance
We know exactly what was argued, by whom, where and when — the piece is signed, dated, and labelled opinion, so our read of the framing is firm. What we cannot do is test it: one publisher, no trial record, no clinician or health-system response, and no data on the practice patterns at issue. Confidence in the description is high; confidence in the prediction is not ours to give.