The Clancy Mistrial Settles a Legal Question, Not a Clinical One

We wrote in August about what testimony in this case described: two clinicians treating the same patient at the same time with no contact and no shared records, two hospital stays whose records never reached the outpatient providers, and a medication history that no single clinician was holding in full. That was When the Record Doesn't Follow the Patient: Documentation Lessons from the Clancy Trial. Nothing in the last three weeks changes any of it, and the reason is worth spelling out.

What the mistrial decided

Legally, one thing: twelve jurors could not agree unanimously. The note the foreperson sent read, "It is with a heavy heart that we report we are unable to come to a unanimous decision, and we will not be able to." Judge Sullivan had already delivered the Tuey-Rodriguez charge, the instruction Massachusetts courts give a deadlocked jury to encourage further deliberation. When the deadlock held, he declared a mistrial.

A hung jury is not an acquittal. The Commonwealth keeps the ability to prosecute the case again. Plymouth County District Attorney Timothy Cruz has not said whether he will retry the three counts of first-degree murder or pursue lesser charges, and a hearing is set for September 29 to take up scheduling. Clancy remains in state psychiatric custody, where she has been held for more than three years.

The reported split needs care. Defense attorney Kevin Reddington told reporters the jury stood 11-1 in favor of acquittal. Prosecutors disputed that reading of the foreperson's notes, arguing they did not establish which way the majority leaned. The jury was not polled, so the direction of the split is an assertion by one side and a contested one.

The Massachusetts standard explains how a case with this much psychiatric evidence could still hang. A defendant lacks criminal responsibility if, as a result of mental disease or defect, she could not appreciate the wrongfulness of her conduct or could not conform her conduct to the requirements of the law. Once the issue is properly before the jury, the burden stays with the Commonwealth to prove criminal responsibility beyond a reasonable doubt. Nobody asked the jury whether Clancy was ill; both sides agreed she was. The question was whether the Commonwealth had disproven, to a criminal standard, that the illness took away her capacity to appreciate wrongfulness or to conform her conduct.

The disagreement the jury inherited

The expert witnesses split on a question that has nothing to do with this defendant: when the postpartum period ends.

Prosecution experts applied a four-week onset window and concluded that Clancy was depressed but not experiencing postpartum psychosis. Defense experts argued that specialists commonly treat the postpartum period as running at least a year, which matters here because the youngest child was eight months old. That is not a courtroom artifact invented for trial. Postpartum psychosis has no diagnostic entry of its own in the DSM. It survives as a peripartum-onset specifier with a narrow window, and the clinicians and researchers who treat the condition have argued for years that the window is too short and the omission is expensive: it keeps the condition out of training, out of research funding, and out of the vocabulary of the average outpatient provider. Veerle Bergink, one of the perinatal psychiatrists who followed the trial, put the consequence bluntly: we do not train our doctors in it, and it is massively missed.

So a jury was asked to resolve beyond a reasonable doubt a definitional question the field has not resolved for itself. The deadlock is not surprising. It is also the single most useful thing this trial surfaced for practicing clinicians, because unlike the verdict, that disagreement shows up in ordinary charts. Two clinicians looking at the same patient at eight months postpartum can reach different conclusions about whether the presentation is even in scope, and both can defend the reasoning.

What the record showed, with or without a verdict

Two facts from the trial record got less attention than they deserved.

The prescribing. Reporting on the record describes roughly 30 prescriptions covering 13 different psychiatric medications over about four months, written across more than one prescriber. Experts disagree about what that means. Some read it as misdiagnosis and overprescribing; others read it as ordinary trial and error in a patient who was not responding, with the added risk that antidepressants can precipitate mania in someone with bipolar illness. Both readings depend on the same precondition: somebody has to be able to see the whole list to tell which one is happening. According to testimony, nobody was.

The escalation threshold. Her Edinburgh Postnatal Depression Scale scores worsened over time, and she called suicide hotlines twice, and she was not hospitalized, because she did not describe a specific plan. That outcome is not unusual and not obviously wrong at any individual point of contact. It is how the threshold works in most systems: a patient can deteriorate steadily, reach for help twice, and still fail to meet the operational definition of an emergency on any given day. The deterioration is visible only in the sequence, and the sequence lives in the record or nowhere.

What to carry into your own charts

The August post covered releases, medication reconciliation from the source, and documenting failed coordination attempts. Four additions, all of them prompted by what came out in the weeks since.

Document the trajectory, not the snapshot. One EPDS score is a data point. Three scores with dates in the same note are a clinical finding, and they are the only version of the picture a covering clinician or an ED physician can read quickly. If your EHR buries prior scores in a flowsheet, restate them in the narrative.

Write the escalation threshold down before you need it. Name in the chart what would change your disposition: what symptom, what score movement, what report from a partner or parent, and what you would do. A threshold defined in advance survives a reassuring 30-minute appointment. One defined in the moment usually does not.

Ask about the window instead of assuming it. Delivery date, sleep, personal and family history of bipolar disorder, and any prior postpartum episode, documented as specifics. A personal history of bipolar disorder is the largest known risk factor for postpartum psychosis, raising risk by roughly a hundredfold in the estimates cited during the trial. And the four-week question is now a live one in your chart as much as in a courtroom: if you are ruling something out because the patient is six or eight months postpartum, write down that you did and why.

Name who is holding the medication list. When there is more than one prescriber, the note should say which clinician owns reconciliation. If the answer is that nobody does, that is itself a finding worth recording, and it is a more honest chart than one that leaves the question unasked.

The distinction from the first post still carries the most disposition weight of anything on this list: intrusive thoughts that are ego-dystonic obsessions are a different clinical situation from thoughts arising out of a loss of contact with reality, and the note should say which one you found and what you asked to get there.

The part that is not clinical

Postpartum Support International responded to the mistrial by calling for three changes: state legislation recognizing childbirth-related mental illness in the criminal law, expanded access to emergency perinatal psychiatric care, and public education about a condition that is treatable when it is caught. On the first, the United States is close to alone among peer countries. More than two dozen nations have infanticide statutes that route mothers with postpartum mental illness toward psychiatric commitment rather than prison, most of them descended from English legislation of the 1920s and 1930s. Illinois passed the only comparable American law in 2018, and it operates at sentencing, after a conviction. Massachusetts has considered similar legislation and has not passed it.

PSI also cites the number that makes the access question urgent: when postpartum psychosis goes unrecognized and untreated, the risk of infanticide runs somewhere around 1 to 4 percent.

Massachusetts is not a state that ignores the problem. It requires pediatricians, obstetricians, and primary care physicians to screen for postpartum depression. What it does not have is a mother-baby psychiatric unit where a mother can be admitted without being separated from her infant, and it shares the national shortage of providers trained in perinatal mental health. Screening mandates without somewhere to send a positive screen generate documented positives and not much else.

Where this leaves us

Asked after the mistrial whether the case was about the health care system, Cruz said it was not: it was about Lindsay Clancy and what she did on January 24, 2023. As a description of what a criminal trial is for, that is right. It is also why the questions this case raised for clinicians were never going to be answered in that courtroom, and will not be answered in a second one either.

Whether there is a retrial is not ours to decide. Whether the record follows the patient is.