AI and narrative embeddings detect PTSD following childbirth via birth stories

Free-text analysis using machine learning (ML)-based natural language processing (NLP) shows promise for diagnosing psychiatric conditions. Chat Generative Pre-trained Transformer (ChatGPT) has demonstrated preliminary initial feasibility for this purpose; however, whether it can accurately assess mental illness remains to be determined. This study evaluates the effectiveness of ChatGPT and the text-embedding-ada-002 (ADA) model in detecting post-traumatic stress disorder following childbirth (CB-PTSD), a maternal postpartum mental illness affecting millions of women annually, with no standard screening protocol. Using a sample of 1295 women who gave birth in the last six months and were 18+ years old, recruited through hospital announcements, social media, and professional organizations, we explore ChatGPT's and ADA's potential to screen for CB-PTSD by analyzing maternal childbirth narratives. The PTSD Checklist for DSM-5 (PCL-5; cutoff 31) was used to assess CB-PTSD. By developing an ML model that utilizes numerical vector representation of the ADA model, we identify CB-PTSD via narrative classification. Our model outperformed (F1 score: 0.81) ChatGPT and six previously published large text-embedding models trained on mental health or clinical domains data, suggesting that the ADA model can be harnessed to identify CB-PTSD. Our modeling approach could be generalized to assess other mental health disorders.

Screening for post-traumatic stress disorder following childbirth using the Peritraumatic Distress Inventory

BACKGROUND

Post-traumatic stress disorder (PTSD) following traumatic childbirth may undermine maternal and infant health, but screening for maternal childbirth-related PTSD (CB-PTSD) remains lacking. Acute emotional distress in response to a traumatic experience strongly associates with PTSD. The Peritraumatic Distress Inventory (PDI) assesses acute distress in non-postpartum individuals, but its use to classify women likely to endorse CB-PTSD is unknown.

METHODS

3039 women provided information about their mental health and childbirth experience. They completed the PDI regarding their recent childbirth event, and a PTSD symptom screen to determine CB-PTSD. We employed Exploratory Graph Analysis and bootstrapping to reveal the PDI's factorial structure and optimal cutoff value for CB-PTSD classification.

RESULTS

Factor analysis revealed two strongly correlated stable factors based on a modified version of the PDI: (1) negative emotions and (2) bodily arousal and threat appraisal. A score of 15+ on the modified PDI produced high sensitivity and specificity: 88 % with a positive CB-PTSD screen in the first postpartum months and 93 % with a negative screen.

LIMITATIONS

In this cross-sectional study, the PDI was administered at different timepoints postpartum. Future work should examine the PDI's predictive utility for screening women as closely as possible to the time of childbirth, and establish clinical cutoffs in populations after complicated deliveries.

CONCLUSIONS

Brief self-report screening concerning a woman's emotional reactions to childbirth using our modified PDI tool can detect those likely to endorse CB-PTSD in the early postpartum. This may serve as the initial step of managing symptoms to ultimately prevent chronic manifestations.

Peripartum dissociation, sense of control, postpartum posttraumatic stress disorder and emotional adjustment to motherhood in adult survivors of childhood maltreatment

Survivors of childhood maltreatment (CM) may experience difficulties in the peripartum period and in adjustment to motherhood. In this study we examined a model wherein CM is associated with maternal self-efficacy and maternal bonding three months postpartum, through mediation of peripartum dissociation and reduced sense of control during childbirth and postpartum-posttraumatic-stress disorder (P-PTSD). Women were recruited in a maternity ward within 48 h of childbirth (T1, N = 440), and contacted three-months postpartum (T2, N = 295). Participants completed self-report questionnaires: peripartum dissociation, sense of control (T1), and CM, P-PTSD, postpartum-depression, maternal self-efficacy and bonding (T2). Obstetrical data were collected from medical files. Structural equation modeling was conducted to test the hypothesized model, controlling for mode of delivery and postpartum-depression. Reported CM included child emotional neglect (CEN; 23.5%), child emotional abuse (CEA; 16.3%), child sexual abuse (CSA; 12.9%) and child physical abuse (CPA; 7.1%). CM was positively associated with peripartum dissociation and P-PTSD (p < .001). Peripartum dissociation was positively associated with P-PTSD (p < .001). P-PTSD was negatively associated with maternal self-efficacy (p < .001) and maternal bonding (p < .001). Association between CM and maternal self-efficacy and bonding was serially mediated by peripartum dissociation and P-PTSD, but not by sense of control. Findings remained significant after controlling for mode of delivery and postpartum-depression. CM is a risk factor for adjustment to motherhood, owing to its effects on peripartum dissociation and P-PTSD. Implementation of a trauma-informed approach in obstetric care and recognition of peripartum dissociative reactions are warranted.

Measuring disrespect and abuse during childbirth in a high-resource country: Development and validation of a German self-report tool

INTRODUCTION

Increasing evidence on disrespect and abuse during childbirth has led to growing concern about the quality of care childbearing women are experiencing. To provide quantitative evidence of disrespect and abuse during childbirth services in Germany a validated measurement tool is needed.

RESEARCH AIM

The aim of this research project was the development and psychometric validation of a survey tool in the German language that measures disrespect and abuse of women during childbirth.

METHODS

A survey tool was created including the following measures: German adaptations of the short and long form of the "Mothers on Respect" (MOR) index (MOR-7 and MOR-G); the "Mothers' Autonomy in Decision Making" (MADM) scale; a mistreatment-index (MIST-I) comprising indicators of mistreatment during childbirth; and a set of items that measure experiences of discrimination during maternity care. Internal consistency reliability and construct validity of the scales were assessed using Cronbach's alpha, unweighted least squares factor analysis and non-parametric correlation analysis with a scale that measures a related construct, the Posttraumatic Symptom Scale - Self Report (PSS-SR) scale. We distributed the survey online, recruiting through snowball sampling via social media. A selection bias towards women who had experienced disrespect and abuse during their birth was intended and expedient for tool validation. The final sample of participants (n = 2045) had given birth in Germany between 2009 and 2018.

FINDINGS

More than 77% of the study participants reported at least one form of mistreatment with non-consented care being the most commonly reported type of mistreatment, followed by physical violence, violation of physical privacy, verbal abuse and neglect. All included scales showed good psychometric properties with high Cronbach's alphas (0.95 for both MOR versions and 0.96 for MADM). Factor analysis generated one factor scales with high factor loadings (0.75 to 0.92 for MOR-7; 0.37 to 0.90 for MOR-G and 0.83 to 0.92 for MADM). MOR-7, MOR-G, MADM and MIST-I scores were significantly (p<0.001) correlated with PSS-SR scores (Spearman's rho -0.70, -0.61 and 0.68 for MOR-G, MADM and the MIST-I, respectively).

CONCLUSIONS

This study presents a valid and reliable instrument for the quantitative assessment of disrespect and abuse during childbirth in Germany. Childbearing women's experiences of disrespect and abuse are a relevant phenomenon in German hospital based maternity care. Disrespect and abuse during childbirth appear to contribute to post-traumatic symptoms and may be associated with severe mental health problems postpartum.

Traumatic childbirth experiences, effects and coping: A qualitative study

OBJECTIVE

Childbirth experience and perception of birth are personal. It is important to understand how the traumatic childbirth experiences affected women and for the nursing care that would be given to them. The aim of the research was to determine the childbirth experiences perceived by women as traumatic, how they were affected by these experiences, and how they coped.

METHODS

The study was a descriptive phenomenological approach study. The data were obtained through semi-structured, in-depth individual interviews. Content analysis was used in the analysis of the data. The mothers were contacted by using the social media platforms. Sixteen women were included in the research who expressed that they had traumatic childbirth experiences.

RESULTS

It was determined that women experienced severe fear and pain during childbirth, that they were in unsuitable maternity hospital environments, and that care was not taken for their privacy. At the same time, it was determined that they experienced disorders in the postpartum period connected to their traumatic birth experiences, that they postponed their pregnancy plans due to their fear of childbirth that developed connected to traumatic childbirth or that they did not want to become pregnant, and that they did not receive sufficient support from health personnel.

CONCLUSION

It is necessary to provide privacy during childbirth, to take into consideration the decisions of women, and to provide for their continuous support by midwives.

Assessment of childbirth-related post traumatic stress disorder in Australian mothers: Psychometric properties of the City Birth Trauma Scale

BACKGROUND

The City Birth Trauma Scale (BiTS; Ayers, Wright & Thornton, 2018) is self-report measure of Post-Traumatic Stress Disorder (PTSD) symptoms following childbirth, based on DSM-5 criteria. We report on the first study of the psychometric properties of the BiTS in the Australian population.

METHODS

Participants were mothers of infants aged 0-12 months (N = 705), who completed the BiTS and measures of related constructs. Confirmatory factor analysis was performed to assess the factor structure of the BiTS. Examination of the reliability, convergent, divergent and discriminant validity and acceptability of the BiTS was also examined.

RESULTS

Confirmatory factor analysis supported a bi-factor model of Birth-related Symptoms (BRS) and General Symptoms (GS) of post-partum PTSD as well as a global CB-PTSD factor. Internal consistency was found for the BiTS total scale and two proposed subscales (BRS and GS). BiTS total scores were significantly associated with an established measure of PTSD, providing support for convergent validity. Evidence of discriminant validity was examined by comparing the BiTS to an established measure of postpartum depression.

LIMITATIONS

The present sample may over-represent participants with traumatic birth experiences in comparison to the general public. Furthermore, use of self-report measures limits the capacity to confirm the diagnostic status of participants.

CONCLUSION

These findings suggest that the BiTS is a valid and reliable measure of childbirth-related PTSD, suited for use in postpartum populations. Total scores on the measure may be informative for clinical and research purposes, while evidence suggests strong support for interpretation of subscale scores.

Delivery mode is associated with maternal mental health following childbirth

Childbirth is a life-transforming event often followed by a time of heightened psychological vulnerability in the mother. There is a growing recognition of the importance of obstetrics aspects in maternal well-being with the way of labor potentially influencing psychological adjustment following parturition or failure thereof. Empirical scrutiny on the association between mode of delivery and postpartum well-being remains limited. We studied 685 women who were on average 3 months following childbirth and collected information concerning mode of delivery and pre- and postpartum mental health. Analysis of variance revealed that women who had cesarean section or vaginal instrumental delivery had higher somatization, obsessive compulsive, depression, and anxiety symptom levels than those who had natural or vaginal delivery as well as overall general distress, controlling for premorbid mental health, maternal age, education, primiparity, and medical complication in newborn. Women who underwent unplanned cesarean also had higher levels of childbirth-related PTSD symptoms excluding those with vaginal instrumental. The risk for endorsing psychiatric symptoms reflecting clinically relevant cases increased by twofold following unplanned cesarean and was threefold for probable childbirth-related PTSD. Maternal well-being following childbirth is associated with the experienced mode of delivery. Increasing awareness in routine care of the implications of operative delivery and obstetric interventions in delivery on a woman's mental health is needed. Screening at-risk women could improve the quality of care and prevent enduring symptoms. Research is warranted on the psychological and biological factors implicated in the mode of delivery and their role in postpartum adjustment.

Fear and Anxiety Disorders Related to Childbirth: Epidemiological and Therapeutic Issues

PURPOSE OF REVIEW

This paper seeks to identify the risk factors of fear of childbirth (FOC) and posttraumatic stress disorder (PTSD) related to birth and reviews the efficacy of their respective screening tools and therapeutic interventions.

RECENT FINDINGS

Biofeedback, hypnosis, internet-based cognitive behavioral therapy, and antenatal education are promising treatments for FOC. Training midwives to address traumatic birth experiences could help in preventing PTSD. A shorter more pragmatic screening tool for FOC than the Wijma Delivery Expectancy/Experience Questionnaire (WDEQ) is needed. Women with PTSD attributed a mismatch between the expected mode of delivery (MOD) and the actual MOD as the cause of their trauma. A history of mental health disorders, lack of social support, previous negative birth experiences, and MOD are correlated to FOC and postpartum PTSD. Psycho-education and CBT-based treatments have been found to reduce levels of FOC and PTSD.

A simple model for prediction postpartum PTSD in high-risk pregnancies

This study aimed to examine the prevalence and possible antepartum risk factors of complete and partial post-traumatic stress disorder (PTSD) among women with complicated pregnancies and to define a predictive model for postpartum PTSD in this population. Women attending the high-risk pregnancy outpatient clinics at Sheba Medical Center completed the Edinburgh Postnatal Depression Scale (EPDS) and a questionnaire regarding demographic variables, history of psychological and psychiatric treatment, previous trauma, previous childbirth, current pregnancy medical and emotional complications, fears from childbirth, and expected pain. One month after delivery, women were requested to repeat the EPDS and complete the Post-traumatic Stress Diagnostic Scale (PDS) via telephone interview. The prevalence rates of postpartum PTSD (9.9 %) and partial PTSD (11.9 %) were relatively high. PTSD and partial PTSD were associated with sadness or anxiety during past pregnancy or childbirth, previous very difficult birth experiences, preference for cesarean section in future childbirth, emotional crises during pregnancy, increased fear of childbirth, higher expected intensity of pain, and depression during pregnancy. We created a prediction model for postpartum PTSD which shows a linear growth in the probability for developing postpartum PTSD when summing these seven antenatal risk factors. Postpartum PTSD is extremely prevalent after complicated pregnancies. A simple questionnaire may aid in identifying at-risk women before childbirth. This presents a potential for preventing or minimizing postpartum PTSD in this population.