Predicting postpartum psychiatric admission using a machine learning approach

Ability of clinical data to predict readmission in Child and Adolescent Mental Health Services

This study addresses the challenge of predicting readmissions in Child and Adolescent Mental Health Services (CAMHS) by analyzing the predictability of readmissions over short, medium, and long term periods. Using health records spanning 35 years, which included 22,643 patients and 30,938 episodes of care, we focused on the episode of care as a central unit, defined as a referral-discharge cycle that incorporates assessments and interventions. Data pre-processing involved handling missing values, normalizing, and transforming data, while resolving issues related to overlapping episodes and correcting registration errors where possible. Readmission prediction was inferred from electronic health records (EHR), as this variable was not directly recorded. A binary classifier distinguished between readmitted and non-readmitted patients, followed by a multi-class classifier to categorize readmissions based on timeframes: short (within 6 months), medium (6 months - 2 years), and long (more than 2 years). Several predictive models were evaluated based on metrics like AUC, F1-score, precision, and recall, and the K-prototype algorithm was employed to explore similarities between episodes through clustering. The optimal binary classifier (Oversampled Gradient Boosting) achieved an AUC of 0.7005, while the multi-class classifier (Oversampled Random Forest) reached an AUC of 0.6368. The K-prototype resulted in three clusters as optimal (SI: 0.256, CI: 4473.64). Despite identifying relationships between care intensity, case complexity, and readmission risk, generalizing these findings proved difficult, partly because clinicians often avoid discharging patients likely to be readmitted. Overall, while this dataset offers insights into patient care and service patterns, predicting readmissions remains challenging, suggesting a need for improved analytical models that consider patient development, disease progression, and intervention effects.

Development of a prediction model of postpartum hospital use using an equity-focused approach

BACKGROUND

Racial inequities in maternal morbidity and mortality persist into the postpartum period, leading to a higher rate of postpartum hospital use among Black and Hispanic people. Delivery hospitalizations provide an opportunity to screen and identify people at high risk to prevent adverse postpartum outcomes. Current models do not adequately incorporate social and structural determinants of health, and some include race, which may result in biased risk stratification.

OBJECTIVE

This study aimed to develop a risk prediction model of postpartum hospital use while incorporating social and structural determinants of health and using an equity approach.

STUDY DESIGN

We conducted a retrospective cohort study using 2016-2018 linked birth certificate and hospital discharge data for live-born infants in New York City. We included deliveries from 2016 to 2017 in model development, randomly assigning 70%/30% of deliveries as training/test data. We used deliveries in 2018 for temporal model validation. We defined "Composite postpartum hospital use" as at least 1 readmission or emergency department visit within 30 days of the delivery discharge. We categorized diagnosis at first hospital use into 14 categories based on International Classification of Diseases-Tenth Revision diagnosis codes. We tested 72 candidate variables, including social determinants of health, demographics, comorbidities, obstetrical complications, and severe maternal morbidity. Structural determinants of health were the Index of Concentration at the Extremes, which is an indicator of racial-economic segregation at the zip code level, and publicly available indices of the neighborhood built/natural and social/economic environment of the Child Opportunity Index. We used 4 statistical and machine learning algorithms to predict "Composite postpartum hospital use", and an ensemble approach to predict "Cause-specific postpartum hospital use". We simulated the impact of each risk stratification method paired with an effective intervention on race-ethnic equity in postpartum hospital use.

RESULTS

The overall incidence of postpartum hospital use was 5.7%; the incidences among Black, Hispanic, and White people were 8.8%, 7.4%, and 3.3%, respectively. The most common diagnoses for hospital use were general perinatal complications (17.5%), hypertension/eclampsia (12.0%), nongynecologic infections (10.7%), and wound infections (8.4%). Logistic regression with least absolute shrinkage and selection operator selection retained 22 predictor variables and achieved an area under the receiver operating curve of 0.69 in the training, 0.69 in test, and 0.69 in validation data. Other machine learning algorithms performed similarly. Selected social and structural determinants of health features included the Index of Concentration at the Extremes, insurance payor, depressive symptoms, and trimester entering prenatal care. The "Cause-specific postpartum hospital use" model selected 6 of the 14 outcome diagnoses (acute cardiovascular disease, gastrointestinal disease, hypertension/eclampsia, psychiatric disease, sepsis, and wound infection), achieving an area under the receiver operating curve of 0.75 in training, 0.77 in test, and 0.75 in validation data using a cross-validation approach. Models had slightly lower performance in Black and Hispanic subgroups. When simulating use of the risk stratification models with a postpartum intervention, identifying high-risk individuals with the "Composite postpartum hospital use" model resulted in the greatest reduction in racial-ethnic disparities in postpartum hospital use, compared with the "Cause-specific postpartum hospital use" model or a standard approach to identifying high-risk individuals with common pregnancy complications.

CONCLUSION

The "Composite postpartum hospital use" prediction model incorporating social and structural determinants of health can be used at delivery discharge to identify persons at risk for postpartum hospital use.

Recentering responsible and explainable artificial intelligence research on patients: implications in perinatal psychiatry

In the setting of underdiagnosed and undertreated perinatal depression (PD), Artificial intelligence (AI) solutions are poised to help predict and treat PD. In the near future, perinatal patients may interact with AI during clinical decision-making, in their patient portals, or through AI-powered chatbots delivering psychotherapy. The increase in potential AI applications has led to discussions regarding responsible AI and explainable AI (XAI). Current discussions of RAI, however, are limited in their consideration of the patient as an active participant with AI. Therefore, we propose a patient-centered, rather than a patient-adjacent, approach to RAI and XAI, that identifies autonomy, beneficence, justice, trust, privacy, and transparency as core concepts to uphold for health professionals and patients. We present empirical evidence that these principles are strongly valued by patients. We further suggest possible design solutions that uphold these principles and acknowledge the pressing need for further research about practical applications to uphold these principles.

Prediction of patient admission and readmission in adults from a Colombian cohort with bipolar disorder using artificial intelligence

Introduction

Bipolar disorder (BD) is a chronically progressive mental condition, associated with a reduced quality of life and greater disability. Patient admissions are preventable events with a considerable impact on global functioning and social adjustment. While machine learning (ML) approaches have proven prediction ability in other diseases, little is known about their utility to predict patient admissions in this pathology.

Aim

To develop prediction models for hospital admission/readmission within 5 years of diagnosis in patients with BD using ML techniques.

Methods

The study utilized data from patients diagnosed with BD in a major healthcare organization in Colombia. Candidate predictors were selected from Electronic Health Records (EHRs) and included sociodemographic and clinical variables. ML algorithms, including Decision Trees, Random Forests, Logistic Regressions, and Support Vector Machines, were used to predict patient admission or readmission. Survival models, including a penalized Cox Model and Random Survival Forest, were used to predict time to admission and first readmission. Model performance was evaluated using accuracy, precision, recall, F1 score, area under the receiver operating characteristic curve (AUC) and concordance index.

Results

The admission dataset included 2,726 BD patients, with 354 admissions, while the readmission dataset included 352 patients, with almost half being readmitted. The best-performing model for predicting admission was the Random Forest, with an accuracy score of 0.951 and an AUC of 0.98. The variables with the greatest predictive power in the Recursive Feature Elimination (RFE) importance analysis were the number of psychiatric emergency visits, the number of outpatient follow-up appointments and age. Survival models showed similar results, with the Random Survival Forest performing best, achieving an AUC of 0.95. However, the prediction models for patient readmission had poorer performance, with the Random Forest model being again the best performer but with an AUC below 0.70.

Conclusion

ML models, particularly the Random Forest model, outperformed traditional statistical techniques for admission prediction. However, readmission prediction models had poorer performance. This study demonstrates the potential of ML techniques in improving prediction accuracy for BD patient admissions.

Early trajectory of clinical global impression as a transdiagnostic predictor of psychiatric hospitalisation: a retrospective cohort study

BACKGROUND

Identifying patients most at risk of psychiatric hospitalisation is crucial to improving service provision and patient outcomes. Existing predictors focus on specific clinical scenarios and are not validated with real-world data, limiting their translational potential. This study aimed to determine whether early trajectories of Clinical Global Impression Severity are predictors of 6 month risk of hospitalisation.

METHODS

This retrospective cohort study used data from the NeuroBlu database, an electronic health records network from 25 US mental health-care providers. Patients with an ICD-9 or ICD-10 code of major depressive disorder, bipolar disorder, generalised anxiety disorder, post-traumatic stress disorder, schizophrenia or schizoaffective disorder, ADHD, or personality disorder were included. Using this cohort, we assessed whether clinical severity and instability (operationalised using Clinical Global Impression Severity measurements) during a 2-month period were predictors of psychiatric hospitalisation within the next 6 months.

FINDINGS

36 914 patients were included (mean age 29·7 years [SD 17·5]; 21 156 [57·3%] female, 15 748 [42·7%] male; 20 559 [55·7%] White, 4842 [13·1%] Black or African American, 286 [0·8%] Native Hawaiian or other Pacific Islander, 300 [0·8%] Asian, 139 [0·4%] American Indian or Alaska Native, 524 (1·4%) other or mixed race, and 10 264 [27·8%] of unknown race). Clinical severity and instability were independent predictors of risk of hospitalisation (adjusted hazard ratio [HR] 1·09, 95% CI 1·07-1·10 for every SD increase in instability; 1·11, 1·09-1·12 for every SD increase in severity; p<0·0001 for both). These associations were consistent across all diagnoses, age groups, and in both males and females, as well as in several robustness analyses, including when clinical severity and clinical instability were based on the Patient Health Questionnaire-9 rather than Clinical Global Impression Severity measurements. Patients in the top half of the cohort for both clinical severity and instability were at an increased risk of hospitalisation compared with those in the bottom half along both dimensions (HR 1·45, 95% CI 1·39-1·52; p<0·0001).

INTERPRETATION

Clinical instability and severity are independent predictors of future risk of hospitalisation, across diagnoses, age groups, and in both males and females. These findings could help clinicians make prognoses and screen patients who are most likely to benefit from intensive interventions, as well as help health-care providers plan service provisions by adding additional detail to risk prediction tools that incorporate other risk factors.

FUNDING

National Institute for Health and Care Research, National Institute for Health and Care Research Oxford Health Biomedical Research Centre, Medical Research Council, Academy of Medical Sciences, and Holmusk.

Identification of Postpartum Depression in Electronic Health Records: Validation in a Large Integrated Health Care System

BACKGROUND

The accuracy of electronic health records (EHRs) for identifying postpartum depression (PPD) is not well studied.

OBJECTIVE

This study aims to evaluate the accuracy of PPD reporting in EHRs and compare the quality of PPD data collected before and after the implementation of the International Classification of Diseases, Tenth Revision (ICD-10) coding in the health care system.

METHODS

Information on PPD was extracted from a random sample of 400 eligible Kaiser Permanente Southern California patients' EHRs. Clinical diagnosis codes and pharmacy records were abstracted for two time periods: January 1, 2012, through December 31, 2014 (International Classification of Diseases, Ninth Revision [ICD-9] period), and January 1, 2017, through December 31, 2019 (ICD-10 period). Manual chart reviews of clinical records for PPD were considered the gold standard and were compared with corresponding electronically coded diagnosis and pharmacy records using sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Kappa statistic was calculated to measure agreement.

RESULTS

Overall agreement between the identification of depression using combined diagnosis codes and pharmacy records with that of medical record review was strong (κ=0.85, sensitivity 98.3%, specificity 83.3%, PPV 93.7%, NPV 95.0%). Using only diagnosis codes resulted in much lower sensitivity (65.4%) and NPV (50.5%) but good specificity (88.6%) and PPV (93.5%). Separately, examining agreement between chart review and electronic coding among diagnosis codes and pharmacy records showed sensitivity, specificity, and NPV higher with prescription use records than with clinical diagnosis coding for PPD, 96.5% versus 72.0%, 96.5% versus 65.0%, and 96.5% versus 65.0%, respectively. There was no notable difference in agreement between ICD-9 (overall κ=0.86) and ICD-10 (overall κ=0.83) coding periods.

CONCLUSIONS

PPD is not reliably captured in the clinical diagnosis coding of EHRs. The accuracy of PPD identification can be improved by supplementing clinical diagnosis with pharmacy use records. The completeness of PPD data remained unchanged after the implementation of the ICD-10 diagnosis coding.

Trends of perinatal mental health referrals and psychiatric admissions in Queensland

BACKGROUND

Perinatal depression is often underdiagnosed; consequently, many women suffer perinatal depression without follow-up care. Screening for depressive symptoms during the perinatal period has been recommended in Australia to increase detection and follow-up of women suffering from depressive symptoms. Screening rates have gradually increased over the last decades in Australia.

OBJECTIVE

To explore trends in referrals of women to community mental health services during the perinatal period, and prenatal and postnatal admissions to psychiatric units, among those who gave birth in Queensland between 2009 and 2015.

METHOD

Retrospective analyses of data from three linked state-wide administrative data collections. Trend analyses using adjusted Poisson regression models examined 426,242 births. Outcome variables included referrals to specialised mental health services; women admitted with a mood disorder during the second half of their pregnancy and during the first 3 months of the postnatal period; and women admitted with non-affective psychosis disorders during the second half of their pregnancy and during the first 3 months of the postnatal period.

RESULTS

We found an increase in mental health referrals during the perinatal period over time (adjusted incidence rate ratio, 1.07; 95% confidence interval, [1.06, 1.08]) and a decrease in admissions with mood disorders during the first 3 months of the postnatal period (adjusted incidence rate ratio, 0.95; 95% confidence interval, [0.94, 0.98]). We did not find any changes in rates of admission for other outcomes.

CONCLUSION

Since the introduction of universal screening in Queensland, referrals for mental health care during the perinatal period have increased, while admissions for mood disorders in the first 3 months after delivery decreased.

Postpartum Psychosis

PURPOSE OF REVIEW

Postpartum psychosis is a psychiatric emergency that can affect the health and life of mothers, infants, and families. Postpartum psychosis (PPP) is distinct from non-postpartum psychosis in many ways, and it is crucial to study and understand PPP to identify, treat, and possibly prevent this condition. We therefore sought to review the latest research findings about PPP with the intention of updating readers about the latest evidence base.

RECENT FINDINGS

Multiple physiologic pathways have been implicated in the development of PPP, and further understanding these pathways may allow for early detection and treatment. Risk assessment and treatment should include consideration of the woman patient but also the mother-infant dyad and the larger family. It is our hope that this review of research updates in postpartum psychosis may inform clinical practice and promote specialized, evidence-based diagnosis, risk assessment, and treatment.

A Scoping Review of the Use of Machine Learning in Health Economics and Outcomes Research: Part 2-Data From Nonwearables

OBJECTIVES

Despite the increasing interest in applying machine learning (ML) methods in health economics and outcomes research (HEOR), stakeholders face uncertainties in when and how ML can be used. We reviewed the recent applications of ML in HEOR.

METHODS

We searched PubMed for studies published between January 2020 and March 2021 and randomly chose 20% of the identified studies for the sake of manageability. Studies that were in HEOR and applied an ML technique were included. Studies related to wearable devices were excluded. We abstracted information on the ML applications, data types, and ML methods and analyzed it using descriptive statistics.

RESULTS

We retrieved 805 articles, of which 161 (20%) were randomly chosen. Ninety-two of the random sample met the eligibility criteria. We found that ML was primarily used for predicting future events (86%) rather than current events (14%). The most common response variables were clinical events or disease incidence (42%) and treatment outcomes (22%). ML was less used to predict economic outcomes such as health resource utilization (16%) or costs (3%). Although electronic medical records (35%) were frequently used for model development, claims data were used less frequently (9%). Tree-based methods (eg, random forests and boosting) were the most commonly used ML methods (31%).

CONCLUSIONS

The use of ML techniques in HEOR is growing rapidly, but there remain opportunities to apply them to predict economic outcomes, especially using claims databases, which could inform the development of cost-effectiveness models.

On AI Approaches for Promoting Maternal and Neonatal Health in Low Resource Settings: A Review

A significant challenge for hospitals and medical practitioners in low- and middle-income nations is the lack of sufficient health care facilities for timely medical diagnosis of chronic and deadly diseases. Particularly, maternal and neonatal morbidity due to various non-communicable and nutrition related diseases is a serious public health issue that leads to several deaths every year. These diseases affecting either mother or child can be hospital-acquired, contracted during pregnancy or delivery, postpartum and even during child growth and development. Many of these conditions are challenging to detect at their early stages, which puts the patient at risk of developing severe conditions over time. Therefore, there is a need for early screening, detection and diagnosis, which could reduce maternal and neonatal mortality. With the advent of Artificial Intelligence (AI), digital technologies have emerged as practical assistive tools in different healthcare sectors but are still in their nascent stages when applied to maternal and neonatal health. This review article presents an in-depth examination of digital solutions proposed for maternal and neonatal healthcare in low resource settings and discusses the open problems as well as future research directions.

Technology-Based Approaches for Supporting Perinatal Mental Health

PURPOSE OF REVIEW

This review explores advances in the utilization of technology to address perinatal mood and anxiety disorders (PMADs). Specifically, we sought to assess the range of technologies available, their application to PMADs, and evidence supporting use.

RECENT FINDINGS

We identified a variety of technologies with promising capacity for direct intervention, prevention, and augmentation of clinical care for PMADs. These included wearable technology, electronic consultation, virtual and augmented reality, internet-based cognitive behavioral therapy, and predictive analytics using machine learning. Available evidence for these technologies in PMADs was almost uniformly positive. However, evidence for use in PMADs was limited compared to that in general mental health populations. Proper attention to PMADs has been severely limited by issues of accessibility, affordability, and patient acceptance. Increased use of technology has the potential to address all three of these barriers by facilitating modes of communication, data collection, and patient experience.

Sensing and Artificial Intelligent Maternal-Infant Health Care Systems: A Review

Currently, information and communication technology (ICT) allows health institutions to reach disadvantaged groups in rural areas using sensing and artificial intelligence (AI) technologies. Applications of these technologies are even more essential for maternal and infant health, since maternal and infant health is vital for a healthy society. Over the last few years, researchers have delved into sensing and artificially intelligent healthcare systems for maternal and infant health. Sensors are exploited to gauge health parameters, and machine learning techniques are investigated to predict the health conditions of patients to assist medical practitioners. Since these healthcare systems deal with large amounts of data, significant development is also noted in the computing platforms. The relevant literature reports the potential impact of ICT-enabled systems for improving maternal and infant health. This article reviews wearable sensors and AI algorithms based on existing systems designed to predict the risk factors during and after pregnancy for both mothers and infants. This review covers sensors and AI algorithms used in these systems and analyzes each approach with its features, outcomes, and novel aspects in chronological order. It also includes discussion on datasets used and extends challenges as well as future work directions for researchers.

Machine Learning Methods for Predicting Postpartum Depression: Scoping Review

BACKGROUND

Machine learning (ML) offers vigorous statistical and probabilistic techniques that can successfully predict certain clinical conditions using large volumes of data. A review of ML and big data research analytics in maternal depression is pertinent and timely, given the rapid technological developments in recent years.

OBJECTIVE

This study aims to synthesize the literature on ML and big data analytics for maternal mental health, particularly the prediction of postpartum depression (PPD).

METHODS

We used a scoping review methodology using the Arksey and O'Malley framework to rapidly map research activity in ML for predicting PPD. Two independent researchers searched PsycINFO, PubMed, IEEE Xplore, and the ACM Digital Library in September 2020 to identify relevant publications in the past 12 years. Data were extracted from the articles' ML model, data type, and study results.

RESULTS

A total of 14 studies were identified. All studies reported the use of supervised learning techniques to predict PPD. Support vector machine and random forest were the most commonly used algorithms in addition to Naive Bayes, regression, artificial neural network, decision trees, and XGBoost (Extreme Gradient Boosting). There was considerable heterogeneity in the best-performing ML algorithm across the selected studies. The area under the receiver operating characteristic curve values reported for different algorithms were support vector machine (range 0.78-0.86), random forest method (0.88), XGBoost (0.80), and logistic regression (0.93).

CONCLUSIONS

ML algorithms can analyze larger data sets and perform more advanced computations, which can significantly improve the detection of PPD at an early stage. Further clinical research collaborations are required to fine-tune ML algorithms for prediction and treatment. ML might become part of evidence-based practice in addition to clinical knowledge and existing research evidence.

Overview of the role of big data in mental health: A scoping review (Preprint).. [DOI: 10.2196/preprints.32424] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

BACKGROUND

Big Data offers promise in the field of mental health and plays an important part when it comes to automation, analysis and prevention of mental health disorders

OBJECTIVE

The purpose of this scoping review is to explore how big data was exploited in mental health. This review specifically addresses both the volume, velocity, veracity and variety of collected data as well as how data was attained, stored, managed, and kept private and secure.

METHODS

Six databases were searched to find relevant articles. PRISMA Extension for Scoping Reviews (PRISMA-ScR) was used as a guideline methodology to develop a comprehensive scoping review.

RESULTS

General and Big Data features were extracted from the studies reviewed. Various technologies were noted when it comes to using Big Data in mental health with depression and anxiety being the focus of most of the studies. Some of these included Machine Learning (ML) models in 22 studies of which Random Forest (RF) was the most widely used. Logistic Regression (LR) was used in 4 studies, and Support Vector Machine (SVM) was used in 3 studies.

CONCLUSIONS

In order to utilize Big Data as a way to mitigate mental health disorders and prevent their appearance altogether a great effort is still needed. Integration and analysis of Big Data, doctors and researchers alike can find patterns in otherwise difficult to identify data by making use of AI and Machine Learning techniques. Similarly, machine learning and artificial intelligence can be used to automate the analytical process.

Automatic detection and characterization of quantitative phase images of thalassemic red blood cells using a mask region-based convolutional neural network

SIGNIFICANCE

Label-free quantitative phase imaging is a promising technique for the automatic detection of abnormal red blood cells (RBCs) in real time. Although deep-learning techniques can accurately detect abnormal RBCs from quantitative phase images efficiently, their applications in diagnostic testing are limited by the lack of transparency. More interpretable results such as morphological and biochemical characteristics of individual RBCs are highly desirable.

AIM

An end-to-end deep-learning model was developed to efficiently discriminate thalassemic RBCs (tRBCs) from healthy RBCs (hRBCs) in quantitative phase images and segment RBCs for single-cell characterization.

APPROACH

Two-dimensional quantitative phase images of hRBCs and tRBCs were acquired using digital holographic microscopy. A mask region-based convolutional neural network (Mask R-CNN) model was trained to discriminate tRBCs and segment individual RBCs. Characterization of tRBCs was achieved utilizing SHapley Additive exPlanation analysis and canonical correlation analysis on automatically segmented RBC phase images.

RESULTS

The implemented model achieved 97.8% accuracy in detecting tRBCs. Phase-shift statistics showed the highest influence on the correct classification of tRBCs. Associations between the phase-shift features and three-dimensional morphological features were revealed.

CONCLUSIONS

The implemented Mask R-CNN model accurately identified tRBCs and segmented RBCs to provide single-RBC characterization, which has the potential to aid clinical decision-making.