Postpartum psychosis is a rare but serious psychiatric condition, affecting approximately 1 to 2 per 1000 women after childbirth. Due to its rarity, most healthcare providers have limited experience identifying its signs and symptoms, and there are currently no comprehensive, evidence-based guidelines for its treatment. While interest in postpartum mood and anxiety disorders have grown significantly over the past decade, the infrequent occurrence of postpartum psychosis—combined with inconsistent classification of the condition—has made systematic research particularly difficult. As a result, our understanding of the diagnosis, management, and underlying causes of postpartum psychosis remains limited, leaving significant gaps in knowledge surrounding this most severe form of postpartum mental illness.
The Mass General Hospital Postpartum Psychosis Project
In an effort to gain a deeper and more nuanced understanding of postpartum psychosis, Lee Cohen, MD, along with colleagues Rachel Vanderkruik, PhD, and Marlene Freeman, MD, from the Ammon-Pinizzotto Center for Women’s Mental Health, launched the Massachusetts General Hospital Postpartum Psychosis Project (MGHP3). Established in 2018, MGHP3 was designed to collect comprehensive data on the characteristics, onset, and treatment of postpartum psychosis across a large and geographically diverse population.
The overarching goal of this project is to expand our understanding of the phenomenology of postpartum psychosis within a diverse cohort of women. Specifically, the project aims to identify risk factors for postpartum psychosis, to better characterize its clinical presentation, and to advance our understanding of the underlying causes of this rare but serious condition.
Study Design
The study is actively recruiting women who experienced psychosis within six months of childbirth and are within ten years of their most recent episode of postpartum psychosis. Data is collected through a one-time structured clinical interview, conducted by phone or Zoom, which includes administration of the Mini International Neuropsychiatric Interview for Psychotic Disorders Studies and the MGHP3 Questionnaire.
A wide range of demographic and clinical information is gathered, including the timing of illness onset, symptom patterns, co-occurring conditions, psychiatric diagnoses before and after the postpartum psychosis episode, and details on treatment received.
As part of the study, participants also provide a saliva sample for genetic analysis. In addition, a subset of participants will undergo neuroimaging assessments to further investigate potential biological factors contributing to postpartum psychosis.
Preliminary Results
The research team published preliminary results from MGHP3, analyzing data from 248 participants, representing a total of 258 episodes of postpartum psychosis (PP). Participants came from 41 U.S. states, Washington D.C., and 7 countries. The time between the onset of the initial postpartum psychosis episode and the study interview varied widely, with a mean of 167.63 weeks (SD = 162.33), or approximately 3.22 years.
Clinical Characteristics
Prior to their first postpartum psychosis episode, 38.3% of participants had no prior psychiatric diagnosis, while 61.7% had at least one psychiatric diagnosis. Among those with previous psychiatric history, 12.5% had a history of psychosis before the postpartum episode, most often in the context of Bipolar I Disorder with psychotic features, Major Depressive Disorder with psychotic features, or Brief Psychotic Disorder.
At the time of their first postpartum psychosis episode, nearly 72% of participants met criteria for Bipolar I Disorder with psychotic features. Among the 248 participants, 10 experienced recurrent postpartum psychosis. Of those, 6 were again diagnosed with Bipolar I Disorder with psychotic features, while 4 were classified as Brief Psychotic Disorder.
Timing, Duration, and Symptom Patterns
The median time between delivery and symptom onset was 10 days (SD = 43.72; range: 0 to 185 days). The duration of psychotic symptoms varied:
- 1.9% reported symptoms lasting less than 1 day
- 60.9% reported symptoms lasting between 1 day and 1 month
- 24.0% experienced symptoms lasting 1 to 6 months
- 10.47% had symptoms lasting more than 6 months
Among 233 participants, the median time to return to psychiatric baseline was 25 weeks (SD = 50.21; range: 0.1 to 364 weeks). For the 10 participants with recurrent postpartum psychosis, the median time to baseline was 10 weeks (SD = 20.87; range: 0.5 to 52 weeks).
Symptom patterns also varied: 45.7% of participants reported that their symptoms waxed and waned, while 51.9% experienced more continuous symptoms.The most commonly reported symptoms were:
- Odd beliefs or delusions – 87.6%
- Persecutory delusions – 75.2%
- Delusions of reference – 55.8%
- Visual hallucinations – 52.3%
- Auditory hallucinations – 48.06%
Treatment in the Context of Postpartum Psychosis
Among the 258 postpartum psychosis episodes reported, participants received a range of treatments during the postpartum period:
- 93.0% received medication
- 74.4% were hospitalized
- 65.9% received psychotherapy
Of those who received medication, The most commonly used medications were atypical antipsychotics (81.0%), followed by SSRIs (50.0%), benzodiazepines (47.3%), lithium (28.7%), lamotrigine (18.6%), and valproic acid (9.3%).
Clinical Implications
The MGHP3 cohort represents one of the largest clinically assessed cohorts of individuals with postpartum psychosis in the United States, and one of the most comprehensive globally, including detailed diagnostic interviews that will ultimately be combined with genetic data, and neuroimaging assessments to provide an in-depth understanding of this rare but serious condition. Unlike many previous studies, which primarily relied on registry data or small clinical samples, MGHP3 offers a broader, more nuanced view of postpartum psychosis, with data drawn from a diverse cohort. This rich dataset provides critical insights into symptom onset, duration, and treatment, offering a more complete picture of the disorder.
While previous research has consistently highlighted a strong association between postpartum psychosis and bipolar disorder, the MGHP3 findings underscore the heterogeneity of the condition, revealing that postpartum psychosis may not always be linked to bipolar disorder. These findings emphasize the need for further research into the etiology, effective treatments, and long-term management of postpartum psychosis, particularly for women whose symptoms do not align with traditional bipolar presentations.
Additionally, while two-thirds of participants had symptoms lasting less than one month, approximately one-third experienced symptoms for longer than a month, highlighting the potential for prolonged illness. The risks associated with prolonged postpartum psychosis, both for the mother and the child, underscore the importance of developing treatment strategies that can minimize symptom duration and mitigate long-term effects.
The MGHP3 study lays a critical foundation for future research, offering a comprehensive view of postpartum psychosis that integrates translational reproductive neuroscience and genomic research. As further studies are conducted, this work will contribute to a more holistic understanding of postpartum psychosis and help identify effective, individualized treatments for affected women.
Ongoing Recruitment
As of December 18, 2024, 445 participants have enrolled in MGHP3.
The research team continues to actively recruit for this study. Those interested in participating in MGHP3 or referring patients can learn more HERE.
Read More
Cohen LS, Arakelian M, Church TR, Dunk MM, Gaw ML, Yoon HE, Kobylski LA, Vanderkruik R, Freeman MP. The phenomenology of postpartum psychosis: preliminary findings from the Massachusetts General Hospital Postpartum Psychosis Project. Molecular Psychiatry. 2024 Dec 6:1-8.
Cohen LS, Vanderkruik R, Arakelian M, Church TR, Dunk MM, Freeman MP. Establishment of the MGH Postpartum Psychosis Project: MGHP3. PLoS One. 2023 Feb 9; 18(2):e0281133.
