CASE-BASED LEARNING ·
MENTAL HEALTH · RISK MANAGEMENT
When the Chart Isn't Enough: What NP Students Can Learn from the
Lindsay Clancy Case
A trial making national headlines is also, quietly, a case study
every psychiatric-mental-health and primary care NP student should sit with.
In August 2026, testimony resumed in the Massachusetts murder
trial of Lindsay Clancy, the Duxbury mother charged with killing her three
children — Cora, 5; Dawson, 3; and Callan, 8 months — in January 2023. Clancy
has pleaded not guilty. Her defense argues she was in the grip of postpartum
psychosis and an underlying bipolar disorder that went unrecognized;
prosecutors argue she acted intentionally and rationally. That legal question
is for the jury, and nothing here takes a position on it.
What is useful for us, as clinicians who write prescriptions and
answer portal messages between visits, is the clinical picture that has emerged
from testimony: two nurse practitioners, a psychiatrist, telehealth visits,
portal messages arriving nearly every day, and a patient moving between at
least three different care systems that never fully talked to each other.
Whatever the jury ultimately decides about intent, the treatment timeline
itself is a case study in how postpartum mood and anxiety symptoms can be
managed by the book and still go sideways — and where the seams in our systems
tend to be.
|
⚠ Before We Go Further This is an active criminal trial and a pending civil suit
against several of the treating clinicians, who have denied the allegations
against them. Details below are drawn from court testimony reported by major
outlets and are presented for clinical education only — not as a verdict on
anyone's care, competence, or intent. Treat every clinician in this story the
way you'd want to be treated if your own worst week of charting were read
back in a courtroom. |
The
Timeline, As It Has Been Described in Testimony
Clancy's psychiatric care began in September 2022, a few months
after her third child was born, when she was started on treatment by
psychiatrist Dr. Jennifer Tufts, who saw her by telehealth 14 times over the
following four months and wrote roughly half of her approximately 30 total
psychiatric prescriptions. In late November 2022, an NP colleague transferred
Clancy's case to Rebecca Jollotta, a psychiatric nurse practitioner with a
hospital's perinatal behavioral health program, who built a treatment plan
targeting anxiety, depression, and insomnia.
From there, testimony describes a pattern many of us will
recognize: near-daily portal messages. Clancy reported intrusive thoughts,
worsening insomnia, and a frightening sense of numbness — writing at one point
that she felt she was "going to die" and didn't care. She repeatedly
asked to change medications, often stopped taking a new prescription before it
had time to work, and voiced fear of becoming dependent on the benzodiazepines
she'd been given for sleep. Jollotta's working diagnosis was postpartum depression
and anxiety, but she testified she raised an underlying bipolar disorder as a
differential as early as her first visit, based on Clancy's atypical,
non-fatigued response to Zoloft — and disclosed that concern directly to Clancy
and her husband on December 6th. A structured screen (the Mood Disorder
Questionnaire, administered by the prior NP) had come back negative, but
Jollotta testified the tool is less sensitive for bipolar II or unspecified
bipolar presentations, and continued to act on the differential afterward:
holding Prozac and titrating Seroquel on the FDA's manic-episode dosing
schedule rather than its depression schedule, even after Clancy's husband
pushed back, saying, "My wife is not bipolar."
Jollotta testified she directly and repeatedly assessed suicidal
ideation, asking about plan and intent at multiple points in December and
documenting "SI, no plan, no intent" each time, including on a joint
call with Clancy's husband. She also recommended escalation to a higher level
of care more than once — a partial hospitalization program at an out-of-state
hospital in mid-December, and inpatient options (including the mechanics of
ED-to-inpatient bed access) discussed explicitly in a December 15th call. In
late December, Clancy was evaluated at that out-of-state hospital's day
program, which recommended tapering her off one antipsychotic; on New Year's
Eve she was admitted to a psychiatric facility for inpatient stabilization and
was later discharged with an outpatient follow-up plan. Jollotta testified she
never obtained records from either institution — she did not request them, and
disputes the out-of-state hospital's record indicating they called her office
and got no response, saying she never received such a call. She did not
prescribe anything further in January. The children were killed on January 24,
2023, after which prosecutors say Clancy took some of her own medication, cut
herself, and fell from a second-floor window. A forensic toxicologist later
testified that her blood levels of the antipsychotic quetiapine were roughly
ten times below a toxic threshold — testimony the prosecution used to argue
that medication was not driving her behavior that day, while the defense
countered that a drug level is a poor stand-in for someone's overall clinical
state.
What
This Case Is Actually Teaching Us
1. A surge in portal messages is data, not
just workload
Nearly-daily messages reporting worsening symptoms, requests to
change medication, and statements like feeling "numb" and wanting to
"die" are not noise to triage between patients — they are an
escalating clinical picture arriving one inbox message at a time. It's easy,
especially in a high-volume portal, to answer each message as its own small
problem (reassure, adjust, encourage patience) without stepping back to ask
what the trend line looks like. Build a habit of pattern-recognition: if you've
responded to the same patient three or four times in two weeks, that's a signal
to escalate the encounter type — a same-day visit, a check-in call, a safety
screen — rather than another message reply.
2. Nonadherence doesn't mean the risk has
gone away
Clancy reportedly often stopped medications early or took less
than prescribed, out of fear of side effects and benzodiazepine dependence.
That's a completely understandable patient response — and also one that muddies
the clinical picture, because a nonadherent patient can look
"undertreated" on paper while still carrying real, escalating risk.
When a patient tells you they're not taking something as prescribed, that
disclosure is worth more than a note in the chart; it's an opening to ask directly
what's driving the fear, and to revisit the plan together rather than simply
restating it.
3. Treatment-resistant insomnia and
depersonalization deserve a harder look, not just reassurance
Being told a medication "takes several weeks to work"
is accurate and often the right thing to say — until the symptom in question is
severe, refractory insomnia or a patient describing feeling emotionally numb
and detached. Those two symptoms in particular are worth treating as potential
red flags for something beyond straightforward postpartum depression: an
evolving mixed or manic state, or early depersonalization that can precede a
psychotic break. Sleep deprivation is not just uncomfortable in the postpartum
period — it is a recognized, modifiable driver of manic and psychotic episodes,
which makes it a vital sign in this population, not a side complaint.
4. A negative structured screen doesn't
close the question
The Mood Disorder Questionnaire Clancy was given came back
negative — she endorsed only 3 of the 13 listed symptoms, well short of the 7
needed to screen positive. Jollotta testified she knew the MDQ is less
sensitive for bipolar II or bipolar-not-otherwise-specified presentations, and
kept the differential open anyway, reasoning from the atypical medication
reaction itself rather than deferring entirely to the screen. That's the right
instinct, and worth naming explicitly: a validated tool's negative result
narrows the odds, it doesn't rule the condition out, and a clinician who has a
specific reason to distrust a tool's sensitivity for a given presentation is
right to keep asking questions the tool didn't ask.
5. Disclosing a differential to the family
is necessary, but not sufficient, when they push back
Jollotta told Clancy and her husband directly, in person, that
she suspected an underlying bipolar disorder — and continued titrating
medication on a bipolar treatment schedule afterward. Her husband responded,
"My wife is not bipolar," and Clancy said nothing. That moment is
worth sitting with: disclosing a differential is necessary, but a family's
discomfort with a diagnosis doesn't mean the clinical plan should quietly
soften. To her credit, testimony indicates Jollotta didn't abandon the differential
after the pushback — she kept dosing toward it. The harder, less resolved
question the case raises is what happens next when a family pushes back on a
working diagnosis: how, and how often, to revisit it, and how to keep
documentation honest about the disagreement without it reading, in hindsight,
as capitulation.
6. Care fragmentation is where risk hides
Across roughly four months, Clancy was seen by a psychiatrist
and at least two NPs by telehealth, self-referred to an out-of-state specialty
hospital, and completed a psychiatric inpatient stay. Jollotta testified she
never obtained records from the psychiatrist, the out-of-state hospital, or the
inpatient facility — not because a request was denied, but because she never
made one, relying instead on what Clancy reported about each encounter. The
out-of-state hospital's own record states they tried to reach her office and
got no callback; Jollotta disputes ever receiving that call. Whichever account
is accurate, the deeper lesson is the same: when care is split across
institutions that don't share a records system, a phone tag or a missed request
can sit invisibly for weeks with nobody positioned to notice, and "the
patient will tell me what happened" is not a substitute for the discharge
summary or a documented records request, sent and followed up on, on your own
side of the gap.
7. Lab values and toxicology answer a
narrower question than they seem to
The toxicologist's testimony that quetiapine levels were far
below a toxic threshold is a good real-world illustration of something worth
teaching explicitly: a blood level tells you about drug concentration at one
moment, not about a person's overall psychiatric state, especially once
psychosis, dissociation, or a manic episode is in play. Don't let a reassuring
lab value substitute for a face-to-face mental status exam, and don't assume a
low level rules out a medication's contribution to what's unfolding clinically.
Before
You Escalate the Medication Regimen: What Baseline Labs Can Rule Out
It's worth separating two very different kinds of lab testing,
because they get conflated constantly in both practice and in courtroom
testimony. Therapeutic drug monitoring — measuring the blood concentration of a
psychiatric medication itself — is genuinely not routine for most
antidepressants; it's reserved for narrow-therapeutic-index drugs like lithium
and clozapine. But baseline medical labs to rule out organic, treatable mimics
of a mood or anxiety presentation are a different question entirely, and are
considered standard of care in a thorough initial psychiatric workup —
especially postpartum, when several of these conditions are common and their
timelines overlap almost exactly with expected postpartum depression or
anxiety.
A baseline panel worth considering before you escalate a
psychiatric medication regimen:
•
TSH and
free T4 (± free T3): postpartum thyroiditis affects an estimated 5–10% of
postpartum women and classically runs a transient hyperthyroid phase (anxiety,
insomnia, palpitations, irritability) followed by a hypothyroid phase
(depression, fatigue, brain fog) — a timeline that overlaps almost perfectly
with expected PPD onset.
•
CBC:
postpartum anemia from delivery blood loss is common and can present as
fatigue, poor concentration, and low mood that gets read as purely psychiatric.
•
Basic
metabolic panel/electrolytes and renal function: rules out sodium and other
electrolyte disturbances that can cause confusion, agitation, or mood lability.
•
Fasting
glucose or HbA1c: persistent gestational diabetes or undiagnosed insulin
resistance can drive mood swings, irritability, and cognitive symptoms.
•
Vitamin
B12 and folate: deficiency impairs neurotransmitter synthesis and can present
as depression, fatigue, and, in severe cases, cognitive or psychotic-like
symptoms.
•
Vitamin D:
increasingly linked to depressive symptoms; deficiency is common postpartum,
especially with limited sun exposure and breastfeeding.
•
Iron
studies/ferritin: iron deficiency without frank anemia is linked to fatigue,
low mood, and restless sleep.
Why this belongs in the differential: if a patient presents with
severe insomnia, anxiety, and mood symptoms and this workup was never done or
documented, medication changes are being made without having excluded
reversible causes that wouldn't respond to psychiatric treatment at all —
meaning a dose could be escalated, or a drug switched, while the actual driver
goes unaddressed. A documented baseline workup (or a documented, reasoned
decision to skip it) is part of what makes a treatment plan defensible in
hindsight, not just clinically sound in the moment.
Psychotherapy
as a Modality — Not Just an Afterthought to Medication
One thing worth naming directly in this case: a patient who is
telling her clinicians that medications are making her feel worse, not better,
is giving a clear signal to lean harder into a non-pharmacologic track — not
just to keep rotating agents. Psychotherapy is not appropriate as the sole
treatment once a patient is in an acute manic, mixed, or psychotic state, or is
expressing intrusive thoughts of harm — safety stabilization comes first, and
that often means medication and a higher level of care. But for postpartum
depression and anxiety in a patient who is engaged in her own care and not in
acute crisis — which describes much of Clancy's early treatment course —
psychotherapy is first-line, evidence-based treatment in its own right, not a
soft adjunct to be mentioned once and left unaddressed.
•
Cognitive
Behavioral Therapy (CBT): strong evidence base specifically for postpartum
depression and anxiety; targets catastrophic thinking about the baby's safety
and all-or-nothing beliefs about being a “good mother”; structured and
time-limited (roughly 8–16 sessions), which fits a population with little time
and high fatigue. Also the modality of choice for postpartum anxiety/OCD
presentations — intrusive thoughts about harm are common in postpartum anxiety
and OCD and are usually not a sign of psychosis, a distinction worth making
explicitly since it's one of the most misunderstood areas in perinatal mental
health.
•
Interpersonal
Therapy (IPT): specifically validated for postpartum depression; targets role
transition (the identity shift into motherhood, loss of prior roles) and
interpersonal strain (partner conflict, lack of support) — exactly the
territory driving much of postpartum distress that medication alone doesn't
touch. Many perinatal specialists consider it the best-fitted modality for PPD
for this reason.
•
Behavioral
activation: a lower-intensity option — structured re-engagement with valued
activity and social contact — useful when a patient is too overwhelmed or
sleep-deprived for the full cognitive work of CBT, or as a bridge while
awaiting therapy access.
•
Group-based
perinatal support: peer support groups (clinician-facilitated or otherwise)
reduce isolation, itself a major risk amplifier; a valuable adjunct, not a
substitute for individual treatment in moderate-to-severe presentations.
A reasonable prescribing sequence this case argues for: confirm
baseline labs are done; for mild-to-moderate presentations, offer psychotherapy
(CBT or IPT) as first-line with medication as adjunct or second-line if therapy
access is delayed; for moderate-to-severe presentations or limited therapy
access, run medication and therapy concurrently rather than medication alone
with therapy deferred; and if a patient reports medications are worsening
rather than helping, treat that as a cue to lean harder into the
non-pharmacologic arm and reassess the diagnosis — not simply to try another
drug in the same category. Document the therapy referral itself, whether it was
accepted, and any access barriers (insurance, waitlist, childcare) — a
documented offer of therapy a patient couldn't access reads very differently in
hindsight than no mention of therapy at all.
|
✓ Clinical Bottom Line •
A
cluster of near-daily portal messages describing worsening mood, sleep, or
intrusive thoughts is an escalation signal — move the patient into a
higher-touch encounter, don't just keep answering messages. •
Treatment-resistant
insomnia and reports of emotional numbness/depersonalization in the
postpartum period warrant explicit screening for mania, psychosis, and
bipolar spectrum illness — not just "give it more time." •
Medication
nonadherence is a conversation starter, not a data point to file away — ask
what's driving it. •
A
reassuring toxicology or lab result narrows one question; it doesn't close
the broader risk assessment. •
A
baseline metabolic/endocrine workup (TSH, CBC, BMP, B12/folate, vitamin D)
belongs early in a postpartum mood or anxiety presentation — several organic
mimics share the exact same timeline as expected PPD onset. •
A severe
(not mild) activation reaction to an SSRI, especially with
treatment-resistant insomnia, is a cue to formally screen for bipolar
spectrum illness — but know the MDQ's limits: a negative screen doesn't rule
out bipolar II or NOS presentations, and a strong clinical suspicion can
outweigh a negative structured tool. •
A
patient reporting that medication is making things worse, not better, is a
cue to intensify the psychotherapy track (CBT/IPT) alongside or instead of
another medication change — not just to keep rotating agents. •
When a
family pushes back on a disclosed differential diagnosis, that pushback is a
cue to document the disagreement and revisit it deliberately — not a cue to
quietly let the working diagnosis drop. •
After
any ED visit, specialty consult, or inpatient stay, send and follow up on an
actual records request on your own side — don't rely on the other institution
to call, and don't rely on patient self-report to reconstruct what happened
elsewhere. •
Document
your clinical reasoning at each medication change, not just the change itself
— the "why now" is what protects both the patient and you. |
|
◆ A Nuance Worth Sitting With It would be easy to read this case and conclude the answer is
simply "escalate everything, screen for everything, refer
everyone." In practice, most patients reporting postpartum insomnia,
anxiety, and medication side effects are exactly what they appear to be — and
treating every one of them as an impending psychiatric emergency would be its
own kind of harm, eroding trust and overwhelming a system that's already
short on perinatal psychiatric capacity. The clinical skill this case is
really asking us to build isn't more suspicion across the board; it's a
sharper eye for the small number of signals — refractory insomnia,
depersonalization, a diagnosis that isn't responding the way it should, care
split across systems that aren't talking — that mean this particular patient
needs more than the standard next step. |
Board
& Practice Review: Postpartum Mood and Psychotic Disorders
Key distinctions to have cold
•
Postpartum
blues: onset within days of delivery, peaks around day 3–5, resolves within two
weeks without treatment — no functional impairment beyond mild
tearfulness/irritability.
•
Postpartum
depression (PPD): onset typically within the first month but can emerge up to a
year postpartum; meets criteria for a major depressive episode with peripartum
onset per DSM-5-TR; screen with validated tools such as the Edinburgh Postnatal
Depression Scale (EPDS).
•
Postpartum
psychosis: not a standalone DSM-5-TR code — captured as a
specifier/presentation of bipolar disorder, brief psychotic disorder, or
(rarely) schizophrenia with peripartum onset; rapid onset, often within the
first 2 weeks postpartum; psychiatric emergency requiring immediate evaluation,
usually inpatient.
•
Key red
flags that should raise suspicion for an evolving bipolar or psychotic process
rather than uncomplicated PPD: refractory insomnia despite treatment, rapid
mood lability, depersonalization/derealization, disorganized thinking,
new-onset paranoia, or any expressed intrusive thoughts of harm.
•
Antidepressant-induced
activation syndrome: agitation, akathisia-like restlessness, insomnia, and
irritability emerging after starting or increasing an SSRI/SNRI — a signal to
pause and reassess for underlying bipolarity before adding or switching within
the same drug class.
•
Sleep
deprivation is both a symptom and a risk multiplier in postpartum mood
disorders — it should be actively treated and tracked, not treated as an
expected inconvenience of new parenthood.
DSM-5-TR: the diagnostic detail that
actually matters here
It's easy to assume all three of the phenomena above have
equally formal diagnostic criteria. They don't — and knowing exactly where each
one sits in DSM-5-TR is what separates a clean clinical explanation from a
hand-wave.
•
Antidepressant-induced
activation syndrome has no DSM-5 code at all. It is a recognized adverse drug
reaction described in FDA labeling and the pharmacology literature, not a
diagnosable psychiatric disorder — it's identified by temporal association (onset
or worsening tied to starting/increasing the drug) and confirmed by improvement
on dose reduction or discontinuation, not by a criteria checklist. If akathisia
is the dominant feature, DSM-5-TR does have a specific code, Medication-Induced
Acute Akathisia, under medication-induced movement disorders — but that covers
the motor restlessness alone, not the full activation picture.
•
An SSRI
unmasking latent bipolar disorder does have formal DSM-5-TR grounding, and the
criteria matter: a manic or hypomanic episode that emerges during
antidepressant treatment can be coded as Substance/Medication-Induced Bipolar
and Related Disorder — but DSM-5-TR includes a deliberate exclusion clause
stating that if the full manic or hypomanic episode persists at a syndromal
level beyond the expected physiological effect of the medication, that
persistence is itself sufficient evidence for a true Bipolar I or Bipolar II
diagnosis, not merely a substance-induced one. In plain terms: DSM-5 explicitly
treats a mania that doesn't resolve when the drug's effect should have worn off
as evidence the medication revealed a real underlying bipolar disorder, rather
than manufactured a temporary one. This is the exact clinical argument behind
the “unmasking” theory raised in cases like this.
•
Postpartum
psychosis has no standalone DSM-5-TR diagnosis — there is no listed disorder by
that name. What exists instead is a “with peripartum onset” specifier, which
can attach to a Major Depressive Episode or Manic Episode (with onset in
pregnancy or within 4 weeks postpartum, including mood-congruent or
mood-incongruent psychotic features), or to Brief Psychotic Disorder (psychotic
symptoms lasting 1 day to 1 month with full return to baseline functioning,
also with a peripartum-onset specifier).
•
The
specifier's 4-week window is a real, clinically debated limitation: the
research and clinical literature describe postpartum psychosis emerging
anywhere from days to several months postpartum, most commonly in the first two
weeks but not confined to a 4-week cutoff. A patient who becomes psychotic at 6
or 8 weeks postpartum doesn't cleanly qualify for the specifier at all, even
though it's clinically the same phenomenon — clinicians often end up using
Other Specified Bipolar and Related Disorder or Other Specified Schizophrenia
Spectrum and Other Psychotic Disorder to code around that gap.
•
In
practice, most cases clinically labeled “postpartum psychosis” ultimately
resolve to a diagnosis of Bipolar I Disorder, manic or mixed episode, with
psychotic features, with peripartum onset — the majority of women who
experience it are later found to have bipolar spectrum illness, which is part
of why screening for bipolarity (not just depression) in a postpartum patient
with severe or atypical symptoms is diagnostically, not just cautiously,
indicated.
•
The
practical consequence of this diagnostic gap: because there's no standalone
code, there's no dedicated, universally adopted screening pathway for
postpartum psychosis the way the EPDS exists for depression. Risk
identification depends heavily on clinician index of suspicion rather than a
structured instrument — advocacy groups such as Postpartum Support
International have specifically pushed for a distinct DSM classification,
arguing the current structure also complicates research tracking and, in some payer
systems, urgent access to care.
Documentation habits this case reinforces
•
Record the
reasoning behind every dose change or medication switch, not only the new order
— what symptom prompted it, what was discussed, what was ruled out.
•
Log every
portal message touchpoint as part of the clinical record and note when a
pattern of contact triggers escalation to an in-person or higher-acuity
encounter.
•
Document
any attempt (successful or not) to obtain outside records after an ED visit,
consult, or inpatient stay — an unanswered records request is still worth
charting.
•
Explicitly
document a negative screen for safety concerns (self-harm, harm to others,
psychosis) at visits where risk is being actively considered — silence in the
chart reads, in hindsight, as "never asked."
For
the Student or New Grad Reading This
It is tempting, reading a case like this from the outside, to
feel certain you'd have caught what apparently got missed. Be careful with that
certainty. Every signal described above is far easier to see laid out in a
timeline than it is buried inside a fourteen-visit relationship with a
sleep-deprived, understandably frustrated patient who is telling you, in good
faith, that she just needs a different medication. The lesson isn't that any
one clinician failed a test the rest of us would pass. It's that postpartum
psychiatric care sits at an intersection of high stakes, high symptom overlap,
and fragmented systems — and that building deliberate habits (structured
screening, proactive record requests, a lower threshold to escalate refractory
insomnia or depersonalization) is how we compensate for the fact that no single
visit, message, or lab value will ever tell the whole story on its own.
References
CNN.
“Lindsay Clancy repeatedly asked to change medications, psychiatric nurse
says.” August 2026.
NBC10
Boston. “Lindsay Clancy trial focuses on medications she was prescribed.”
August 2026.
CBS
News Boston. “Lindsay Clancy trial focuses on psychiatric care she received
before killing children.” August 2026.
ABC
News / Good Morning America. “Lindsay Clancy trial: Toxicology expert testifies
that drug levels were low.” August 2026.
ABC
News. “‘I feel like I’m going to die’: Nurse practitioner testifies about
Lindsay Clancy messages.” August 2026.
The
Boston Globe. “Nurse practitioner who treated Lindsay Clancy testifies about
Duxbury mother’s desperation to improve amid postpartum struggles.” August
2026.
PBS
NewsHour. “Lindsay Clancy trial turns focus to medications prescribed before
she killed her children.” August 2026.
NPR.
“Lindsay Clancy’s trial highlights gaps in understanding, treating postpartum
psychosis.” August 2026.
American
Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders,
Fifth Edition, Text Revision (DSM-5-TR). 2022.
Commonwealth
v. Lindsay Clancy. Plymouth Superior Court trial testimony of Rebecca Jollotta,
CNP, August 2026.
American
College of Obstetricians and Gynecologists (ACOG). Screening and Diagnosis of
Mental Health Conditions During Pregnancy and Postpartum, Clinical Practice
Guideline. 2023.
This post is for clinical education
purposes only and does not constitute legal commentary, medical advice, or a
judgment of any individual involved in ongoing litigation. Case facts are drawn
from public court reporting and may evolve as the trial continues.
