Background:The "fourth trimester," or postpartum time period, remains a critical phase of pregnancy that significantly impacts parents and newborns. Care poses challenges due to complex individual needs as well as low attendance rates at routine appointments. A comprehensive technological solution could provide a holistic and equitable solution to meet care goals. Objective:This paper describes the development of patient engagement data with a novel postpartum conversational agent that uses natural language processing to support patients post partum. Methods:We report on the development of a postpartum conversational agent from concept to usable product as well as the patient engagement with this technology. Content for the program was developed using patient- and provider-based input and clinical algorithms. Our program offered 2-way communication to patients and details on physical recovery, lactation support, infant care, and warning signs for problems. This was iterated upon by our core clinical team and an external expert clinical panel before being tested on patients. Patients eligible for discharge around 24 hours after delivery who had delivered a singleton full-term infant vaginally were offered use of the program. Patient demographics, accuracy, and patient engagement were collected over the first 6 months of use. Results:A total of 290 patients used our conversational agent over the first 6 months, of which 112 (38.6%) were first time parents and 162 (56%) were Black. In total, 286 (98.6%) patients interacted with the platform at least once, 271 patients (93.4%) completed at least one survey, and 151 (52%) patients asked a question. First time parents and those breastfeeding their infants had higher rates of engagement overall. Black patients were more likely to promote the program than White patients (P=.047). The overall accuracy of the conversational agent during the first 6 months was 77%. Conclusions:It is possible to develop a comprehensive, automated postpartum conversational agent. The use of such a technology to support patients postdischarge appears to be acceptable with very high engagement and patient satisfaction.

Synopsis Uteroenteric fistulae in the setting of degenerating leiomyomas may present with small bowel obstruction and require multidisciplinary surgical management to perform fertility‐sparing myomectomy.

INTRODUCTION: The fourth trimester is a time of great physical and emotional challenge. To help patients navigate this period, we created the Healing at Home (H@H) program, which combined expedited discharge with a 24/7 bidirectional postpartum support chatbot. METHODS: Clinical outcomes, abstracted from medical record and patient report via text message, were evaluated for patients enrolled in H@H (March 2020 to September 2021) compared to patients delivering prior to program implementation (July to December 2019). Enrolled patients had a full-term singleton vaginal delivery of a non-small-for-gestational-age infant without postpartum hemorrhage (>1,000 cc), a low-risk sepsis risk score, a low- to intermediate-risk bilirubin, and lack of significant maternal comorbidities (eg, preeclampsia, preexisting diabetes). Institutional review board waiver for this quality improvement study was obtained. RESULTS: There were 490 patients enrolled in H@H versus 309 in preintervention group. The groups were similar in parity, race, ethnicity, and insurance status. There was no difference in the primary outcome, postpartum visit attendance (76.9% versus 76.1%, P =.77). H@H patients had a shorter postpartum length of stay (33.3 versus 46.7 hours, P <.001), higher rates of exclusive breastfeeding at 1 week (59.2% versus 57.0%, P =.02), and a higher rate of contraceptive use (55% versus 44%, P =.01), with no difference in maternal unscheduled visits or hospital readmissions. CONCLUSION: This innovative postpartum program was associated with a shorter length of stay and improvement in contraceptive acceptance and breastfeeding without increasing unscheduled visits. Future studies should evaluate the cost-effectiveness of this program and expansion to higher-risk patients including those with cesarean delivery and medical comorbidities.

To identify incidence of new onset postpartum hypertensive disorders of pregnancy (HDP) in a low risk obstetric population undergoing remote blood pressure monitoring This retrospective cohort study examined clinical and healthcare utilization outcomes among 316 low-risk patients enrolled in remote blood pressure monitoring as part of a bidirectional text messaging program for postpartum support between September 2021 and May 2022. Patients with singleton pregnancy and no diagnosis of chronic hypertension or HDP antenatally or during delivery hospitalization were eligible for inclusion. The primary outcome was incidence of new onset postpartum HDP post-delivery hospitalization, defined by ACOG guidelines. Secondary outcomes included severity of HDP, chief complaint leading to diagnosis (blood pressures vs symptoms), hospital readmission, emergency department utilization, and postpartum visit attendance. Of 316 patients, nine patients (2.9%) were diagnosed with HDP after discharge from delivery hospitalization (Table 2). Four patients were diagnosed due to blood pressures identified through remote monitoring. Four patients (1.2%) were readmitted postpartum, all for HDP by blood pressure and symptom criteria (median 8 days postpartum). Five patients were diagnosed with HDP but not readmitted (median 6 days postpartum). There were no significant clinical differences between patients with and without postpartum HDP or between patients with and without postpartum readmission (data not shown). In a cohort of low risk obstetric patients undergoing remote blood pressure monitoring, 2.9% developed new onset HDP after discharge from delivery hospitalization. 44% of cases were diagnosed due to blood pressures from remote monitoring. Remote blood pressure monitoring in low risk patients may identify postpartum HDP at higher than expected rates. Increasing accurate diagnosis of postpartum HDP is essential to reducing maternal morbidity and mortality and improving counseling and management for future pregnancies.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

BACKGROUND: The COVID-19 pandemic led to the rapid uptake of telemedicine services, which have been shown to be potentially cost -saving and of comparable quality to in-person care for certain populations. However, there are some concerns regarding the feasibility of implementation for marginalized populations, and the impact of widespread implementation of these services on health disparities has not been well studied.OBJECTIVE: This study aimed to assess the impact of telehealth implementation on postpartum care during the COVID-19 pandemic on racial disparities in visit attendance and completion of postpartum care goals.STUDY DESIGN: In this retrospective cohort study at a single tertiary care center, differences in outcomes between all Black and non-Black patients who had scheduled postpartum visits before and after telehealth implementation for postpartum care were compared. The primary outcome was postpartum visit attendance. The secondary outcomes included postpartum depression screening, contraception selection, breastfeeding status, completion of postpartum 2-hour glucose tolerance test, and cardiology follow-up for hypertensive disorders of pregnancy. In multivariable analysis, interaction terms were used to evaluate the differential impact of telehealth implementation by race.RESULTS: Of 1579 patients meeting the inclusion criteria (780 in the preimplementation group and 799 in the postimplementation group), 995 (63%) self-identified as Black. In the preimplementation period, Black patients were less likely to attend a postpartum visit than non-Black patients (63.9% in Black patients vs 88.7% in non-Black patients; adjusted odds ratio, 0.48; 95% confidence interval, 0.29-0.79). In the postimplementation period, there was no difference in postpartum visit attendance by race (79.1% in Black patients vs 88.6% in non-Black patients; adjusted odds ratio, 0.74; 95% confidence interval, 0.45 -1.21). In addition, significant differences across races in postpartum depression screening during the preimplementation period became nonsignificant in the postimplementation period. Telehealth implementation for postpartum care significantly reduced racial disparities in postpartum visit attendance (interaction P=.005).CONCLUSION: Telehealth implementation for postpartum care during the COVID-19 pandemic was associated with decreased racial disparities in postpartum visit attendance.

INTRODUCTION: Postpartum depression (PPD) is costly for women, families, and the health care system. Early identification is critical. We sought to evaluate PPD screening at delivery discharge and compare ongoing rates of positive screening among women at risk of PPD. METHODS: In September 2020, we implemented universal Edinburgh Postpartum Depression Scale (EPDS) screening postpartum (PP) prior to discharge. A score of 9–12 and>12 was moderate and high risk for PPD, respectively. All who scored≥9 met with social workers and were referred for outpatient resources. From September to December 2020, women who scored≥9 were contacted for re-screening at 1–3 weeks PP and at 6-weeks PP visit. RESULTS: A total of 1,256 women underwent immediate PPD screening. Of those, 165 (13.1%) scored ≥9, with 63 (5.0%) of these scoring >12. Two percent (n=24) indicated thoughts of self-harm. Over 50% (n=80) had no prior history of mental illness. Among women who scored≥9 immediately PP and were reached for re-screening, 58 of 111 (52.5%) scored ≥9 within 3 weeks and 48 of 93 (51.6%) scored ≥9 at their PP visit. Among women who scored>12 immediately PP, 32 of 42 (76.2%) scored ≥9 at 3 weeks and 24 of 31 (77.4%) scored ≥9 at their PP visit. Only 41 (24.8%) had initiated therapy by their PP visit. CONCLUSION: Immediate PP EPDS screening is feasible and captures women at risk of PPD. More than half of women who had an elevated score immediately PP continued to score high at their PP visit. Timely screening and access to mental health services is critical and should be integrated into PP care.

Gaulton JS, et al. BMJ Innov 2021;0:1–5. doi:10.1136/bmjinnov-2021-000791 Neonatology, Jefferson Health–Abington, Abington, Pennsylvania, USA HUP Obstetrics and Gynecology, Penn Medicine, Philadelphia, Pennsylvania, USA Center for Health Care Innovation, Penn Medicine, Philadelphia, Pennsylvania, USA Division of Neonatology, Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA Office of the Dean of the University of Pennsylvania, Penn Medicine, Philadelphia, Pennsylvania, USA Emergency Medicine, University of Pennsylvania Health System, Philadelphia, Pennsylvania, USA

### Summary box #### What are the new findings? #### How might it impact on healthcare in the future? The rise of healthcare chatbots using artificial intelligence (AI) to understand unconstrained natural language input and reply with appropriate answers presents an emerging field of research, but few published studies on this topic include structured evaluation of efficacy or safety.1 In the past few months, healthcare has seen COVID-19 accelerate the adoption of digital health solutions to enable more timely care.2–4 Before AI chatbots can be deployed in healthcare applications, they need to be appropriately ‘trained’ on clinically relevant data.5 We will discuss the context that led to the development of a practical training method for a healthcare AI chatbot that efficiently improves chatbot accuracy and patient safety. The Healing at Home programme at the Hospital of the University of Pennsylvania (HUP) coordinates prioritised discharge and digital access to care for mothers and newborns.6 The American College of Obstetricians and Gynecologists recommends more immediate contact between obstetricians and patients to support postpartum care as an ongoing process, especially during the ‘fourth trimester’ after discharge.7–9 Literature shows many examples of texting interventions improving access to perinatal care.10–15 Healing at Home developed a postpartum support chatbot named ‘Penny’ in a partnership between a multidisciplinary clinical team from HUP, …

The millennial pregnant patient expects an innovative approach to prenatal care. Patients are reaching to peer support online communities or engaging in direct-to-consumer mobile applications during their pregnancy. Currently developed solutions show promise, however, the clinical impact and generalizability of these solutions remains unclear. Technology has the potential to decrease health care disparities, improve patient and provider satisfaction as well as clinical outcomes. In this article we discuss traditional models of prenatal education as well and suggest how obstetricians should consider utilizing technology as an approach to provide prenatal education to their patients.

APPLIED MATHEMATICS Correction for ‘‘On the behavior of a capillary surface in a wedge,’’ by Paul Concus and Robert Finn, which appeared in issue 2, June 1, 1969, of Proc Natl Acad Sci USA (63:292–299). Owing to a printer’s error, the initial sentence of the abstract appeared as ‘‘Estimates above and below are obtained for the height of the equilibrium-free surface of a liquid’’ rather than the intended ‘‘Estimates above and below are obtained for the height of the equilibrium free surface of a liquid.’’ Also, the first sentence of the introduction, ‘‘Consider a volume of liquid that partially fills a cylindrical container Z, forming an equilibriumfree surface S, as determined by surface and gravitational forces’’ should have read ‘‘Consider a volume of liquid that partially fills a cylindrical container Z, forming an equilibrium free surface S, as determined by surface and gravitational forces.’’ These errors can lead to basic misunderstanding as to the nature of the problem addressed in the article. The online version of this article has been corrected.

Adipocyte differentiation is controlled by many transcription factors, but few known downstream targets of these factors are necessary for adipogenesis. Here we report that retinol saturase (RetSat), which is an enzyme implicated in the generation of dihydroretinoid metabolites, is induced during adipogenesis and is directly regulated by the transcription factor peroxisome proliferator activated receptor γ (PPARγ). Ablation of RetSat dramatically inhibited adipogenesis but, surprisingly, this block was not overcome by the putative product of RetSat enzymatic activity. On the other hand, ectopic RetSat with an intact, but not a mutated, FAD/NAD dinucleotide-binding motif increased endogenous PPARγ transcriptional activity and promoted adipogenesis. Indeed, RetSat was not required for adipogenesis when cells were provided with exogenous PPARγ ligands. In adipose tissue, RetSat is expressed in adipocytes but is unexpectedly downregulated in obesity, most likely owing to infiltration of macrophages that we demonstrate to repress RetSat expression. Thiazolidinedione treatment reversed low RetSat expression in adipose tissue of obese mice. Thus, RetSat plays an important role in the biology of adipocytes, where it favors normal differentiation, yet is reduced in the obese state. RetSat is thus a novel target for therapeutic intervention in metabolic disease.

Nuclear receptor corepressor 1 (Ncor1) is an activator for the enzyme histone deacetylase 3 (Hdac3) that is required for embryogenesis, but its physiological functions are unknown. Now experiments in knock-out mice lacking Ncor1 show that disruption of the Ncor1–Hdac3 interaction causes aberrant regulation of clock genes and results in abnormal circadian behaviour — with a sleep–wake cycle closer to 23 hours than the normal 24. These mice are also leaner than normal and more insulin sensitive as a result of increased energy expenditure. Loss of a functional Ncor1–Hdac3 complex in vivo changes the oscillatory patterns of several metabolic genes, demonstrating that circadian regulation of metabolism is critical for normal energy balance. Targeting of the Ncor1–Hdac3 enzyme could be a highly specific intervention in diseases of nutritional stress such as obesity and diabetes. This paper shows that specific genetic disruption of the Ncor–HdaC3 interaction in mice causes aberrant regulation of clock genes and results in abnormal circadian behaviour. These mice are also leaner and more insulin sensitive due to increased energy expenditure. Loss of a functional Ncor–HdaC3 complex in vivo changes the oscillatory patterns of several metabolic genes, demonstrating that circadian regulation of metabolism is critical for normal energy balance. Rhythmic changes in histone acetylation at circadian clock genes suggest that temporal modulation of gene expression is regulated by chromatin modifications1,2,3. Furthermore, recent studies demonstrate a critical relationship between circadian and metabolic physiology4,5,6,7. The nuclear receptor corepressor 1 (Ncor1) functions as an activating subunit for the chromatin modifying enzyme histone deacetylase 3 (Hdac3)8. Lack of Ncor1 is incompatible with life, and hence it is unknown whether Ncor1, and particularly its regulation of Hdac3, is critical for adult mammalian physiology9. Here we show that specific, genetic disruption of the Ncor1–Hdac3 interaction in mice causes aberrant regulation of clock genes and results in abnormal circadian behaviour. These mice are also leaner and more insulin-sensitive owing to increased energy expenditure. Unexpectedly, loss of a functional Ncor1–Hdac3 complex in vivo does not lead to sustained increases in known catabolic genes, but instead significantly alters the oscillatory patterns of several metabolic genes, demonstrating that circadian regulation of metabolism is critical for normal energy balance. These findings indicate that activation of Hdac3 by Ncor1 is a nodal point in the epigenetic regulation of circadian and metabolic physiology.