Postpartum psychosis (PP) is a severe condition affecting 1 to 2 in 1000 individuals after childbirth, characterized by sudden onset of delusions, hallucinations, and mood symptoms. It requires urgent treatment to ensure the safety of both mother and child. While often linked to bipolar disorder, not all people with PP develop recurrent psychiatric issues. Treatment typically includes inpatient care and medication, though some women recover fully while others experience chronic illness.
Research on PP is limited, particularly regarding diverse patient cohorts and long-term outcomes. To address this, the Massachusetts General Hospital Postpartum Psychosis Project (MGHP3) was established in 2018 to collect data on the condition's symptoms, onset, and treatment across a large, geographically diverse group. This report presents findings from the largest U.S.-based cohort studied to date.
MGHP3 enrolls women who have experienced a psychotic episode within 6 months of delivery in the 10 years prior to the interview. Participants complete structured interviews to collect data on psychiatric history, episode specifics, delivery, and medication use. DNA samples are also taken for genetic analysis. Additionally, starting in July 2020, participants are asked to share their lived experiences with PP, which are analyzed qualitatively to complement the quantitative data.
As of November 3, 2022, 328 subjects had enrolled in MGHP3. 248 participants were included in this analysis, accounting for 258 episodes of PP. Individuals who were lost to follow-up, ineligible, withdrawn, or experienced mania without psychotic features were excluded. Participants hailed from 41 states, Washington D.C., and 7 countries. The amount of time that passed between the onset of index PP episode and the time of interview varied greatly, with a mean of 167.63 weeks (standard deviation, SD=162.33), or approximately 3.22 years.*
Clinical Characteristics
Prior to their first episode of PP, 38.3% (n=95) of participants had no previous psychiatric diagnoses, while 61.7% (n=153) had at least one. Among them, 12.5% (n=31) had experienced psychosis before, mainly in the context of Bipolar I Disorder with psychotic features (n=28), Major Depressive Disorder with psychotic features (n=2), or Brief Psychotic Disorder (n=1). At the time of their index PP episode, nearly 72% (n=178) of participants met the criteria for Bipolar I Disorder with psychotic features. Of the 248 participants, 10 experienced recurrent PP, with 6 of these cases also associated with Bipolar I Disorder with psychotic features, while the remaining 4 were classified as Brief Psychotic Disorder.
Timing, Length, and Type of Symptoms
The median number of days between delivery and symptom onset (n=253) was 10 days (SD=43.72, MIN/MAX=0/185). The duration of psychotic symptoms varied, with 60.9% (n=157) reporting symptoms lasting 1 day to 1 month, 24.0% (n=62) lasting 1 month to 6 months, 10.47% (n=27) lasting over 6 months, and 1.9% (n=5) lasting less than 1 day. Among 233 participants, the median time to return to a psychiatric baseline was 25 weeks (SD=50.21; MIN/MAX=0.1/364). For those with recurrent PP (n=10), the median time to baseline was 10 weeks (SD=20.87; MIN/MAX=0.5/52). Nearly half of participants (45.7%; n=118) reported that their symptoms waxed and waned, while 51.9% (n=134) experienced more consistent symptoms. The most common symptoms reported were odd beliefs or delusions (87.6%; n=226), persecutory delusions (75.2%; n=194), and delusions of reference (55.8%; n=144). Visual hallucinations were reported by 52.3% (n=135) and auditory hallucinations by 48.06% (n=124).
Treatment Received in the Context of PP Episode
Participants reported receiving various treatments in the postpartum period following their PP episode(s) (N=258). Most participants received medication (93.0%; n=240), 74.4% were hospitalized (n=192), and 65.89% participated in psychotherapy (n=170). Among those who received medication, 81.0% (n=209) took an atypical antipsychotic, 50.0% (n=129) took an SSRI, 28.7% (n=74) took a mood stabilizer like lithium, 9.3% (n=24) took valproic acid, 18.6% (n=48) took lamotrigine, and 47.3% (n=122) took a benzodiazepine. For atypical antipsychotics (n=209), 41.6% (n=87) took quetiapine, 40.7% (n=85) took olanzapine, and 27.3% (n=57) took risperidone. Among those on SSRIs (n=129), 67.4% (n=87) took sertraline and 27.9% (n=36) took escitalopram. Of those who took a benzodiazepine (n=122), 64.8% (n=79) took lorazepam and 31.2% (n=38) took clonazepam.
This study provides detailed patient-reported data on postpartum psychosis from a large cohort, offering insights into symptom onset, duration, and treatment. The sample includes women from diverse geographic locations and psychiatric backgrounds, with a focus on those with various diagnoses beyond bipolar disorder, including Major Depressive Disorder with psychotic features and Brief Psychotic Disorder. The study found significant variability in the onset and duration of PP symptoms, with a median onset of 10 days postpartum. Symptoms commonly included delusions, persecutory beliefs, and visual hallucinations, with nearly half of participants reporting symptoms that waxed and waned. Treatment typically involved medication, such as atypical antipsychotics, and hospitalization.
These findings highlight the heterogeneity of PP, suggesting that PP may not always be associated with bipolar disorder, and emphasize the need for further research into effective treatments and the long-term management of PP. Future studies could investigate the relationship between PP and other mental health conditions, explore the role of prophylactic treatment, and assess treatment barriers. The study's limitations include potential recall bias due to the retrospective nature of the study and the use of self-reported data, but it offers a broad view of PP across diverse patient backgrounds. These data lay the foundation for future studies aimed at enhancing our understanding of postpartum psychosis, including translational reproductive neuroscience and genomic research in well-characterized populations of affected women.
*Several protocol deviations were made to include participants who had their episode more than 10 years before the interview date but were confident in strong recollection of their episodes.
Citation: Cohen, L. S., Arakelian, M., Church, T. R., Dunk, M. M., Gaw, M. L., Yoon, H. E., ... & Freeman, M. P. (2024). The phenomenology of postpartum psychosis: preliminary findings from the Massachusetts General Hospital Postpartum Psychosis Project. Molecular Psychiatry, 1-8.
Written by: Hannah Yoon, Clinical Research Coordinator
Comments