Semi-supervised Double Deep Learning Temporal Risk Prediction (SeDDLeR) with Electronic Health Records.

BACKGROUND: Risk prediction plays a crucial role in planning for prevention, monitoring, and treatment. Electronic Health Records (EHRs) offer an expansive repository of temporal medical data encompassing both risk factors and outcome indicators essential for effective risk prediction. However, challenges emerge due to the lack of readily available gold-standard outcomes and the complex effects of various risk factors. Compounding these challenges are the false positives in diagnosis codes, and formidable task of pinpointing the onset timing in annotations. OBJECTIVE: We develop a Semi-supervised Double Deep Learning Temporal Risk Prediction (SeDDLeR) algorithm based on extensive unlabeled longitudinal Electronic Health Records (EHR) data augmented by a limited set of gold standard labels on the binary status information indicating whether the clinical event of interest occurred during the follow-up period. METHODS: The SeDDLeR algorithm calculates an individualized risk of developing future clinical events over time using each patient's baseline EHR features via the following steps: (1) construction of an initial EHR-derived surrogate as a proxy for the onset status; (2) deep learning calibration of the surrogate along gold-standard onset status; and (3) semi-supervised deep learning for risk prediction combining calibrated surrogates and gold-standard onset status. To account for missing onset time and heterogeneous follow-up, we introduce temporal kernel weighting. We devise a Gated Recurrent Units (GRUs) module to capture temporal characteristics. We subsequently assess our proposed SeDDLeR method in simulation studies and apply the method to the Massachusetts General Brigham (MGB) Biobank to predict type 2 diabetes (T2D) risk. RESULTS: SeDDLeR outperforms benchmark risk prediction methods, including Semi-parametric Transformation Model (STM) and DeepHit, with consistently best accuracy across experiments. SeDDLeR achieved the best C-statistics ( 0.815, SE 0.023; vs STM +.084, SE 0.030, P-value .004; vs DeepHit +.055, SE 0.027, P-value .024) and best average time-specific AUC (0.778, SE 0.022; vs STM + 0.059, SE 0.039, P-value .067; vs DeepHit + 0.168, SE 0.032, P-value <0.001) in the MGB T2D study. CONCLUSION: SeDDLeR can train robust risk prediction models in both real-world EHR and synthetic datasets with minimal requirements of labeling event times. It holds the potential to be incorporated for future clinical trial recruitment or clinical decision-making.

Stratifying Risk for Postpartum Depression at Time of Hospital Discharge.

Objective: Postpartum depression (PPD) represents a major contributor to postpartum morbidity and mortality. Beyond efforts at routine screening, risk stratification models could enable more targeted interventions in settings with limited resources. Thus, we aimed to develop and estimate the performance of a generalizable risk stratification model for PPD in patients without a history of depression using information collected as part of routine clinical care. Methods: We performed a retrospective cohort study of all individuals who delivered between 2017 and 2022 in one of two large academic medical centers and six community hospitals. An elastic net model was constructed and externally validated to predict PPD using sociodemographic factors, medical history, and prenatal depression screening information, all of which was known before discharge from the delivery hospitalization. Results: The cohort included 29,168 individuals; 2,703 (9.3%) met at least one criterion for postpartum depression in the 6 months following delivery. In the external validation data, the model had good discrimination and remained well-calibrated: area under the receiver operating characteristic curve 0.721 (95% CI: 0.707-0.734), Brier calibration score 0.088 (95% CI: 0.084 - 0.092). At a specificity of 90%, the positive predictive value was 28.0% (95% CI: 26.0-30.1%), and the negative predictive value was 92.2% (95% CI: 91.8-92.7%). Conclusions: These findings demonstrate that a simple machine-learning model can be used to stratify the risk for PPD before delivery hospitalization discharge. This tool could help identify patients within a practice at the highest risk and facilitate individualized postpartum care planning regarding the prevention of, screening for, and management of PPD at the start of the postpartum period and potentially the onset of symptoms.

Family history as the strongest predictor of aortic and peripheral aneurysms in patients with intracranial aneurysms.

OBJECTIVE: Intracranial aneurysms (IA) and aortic aneurysms (AA) are both abnormal dilations of arteries with familial predisposition and have been proposed to share co-prevalence and pathophysiology. Associations of IA and non-aortic peripheral aneurysms are less well-studied. The goal of the study was to understand the patterns of aortic and peripheral (extracranial) aneurysms in patients with IA, and risk factors associated with the development of these aneurysms. METHODS: 4701 patients were included in our retrospective analysis of all patients with intracranial aneurysms at our institution over the past 26 years. Patient demographics, comorbidities, and aneurysmal locations were analyzed. Univariate and multivariate analyses were performed to study associations with and without extracranial aneurysms. RESULTS: A total of 3.4% of patients (161 of 4701) with IA had at least one extracranial aneurysm. 2.8% had thoracic or abdominal aortic aneurysms. Age, male sex, hypertension, coronary artery disease, history of ischemic cerebral infarction, connective tissues disease, and family history of extracranial aneurysms in a 1st degree relative were associated with the presence of extracranial aneurysms and a higher number of extracranial aneurysms. In addition, family history of extracranial aneurysms in a second degree relative is associated with the presence of extracranial aneurysms and atrial fibrillation is associated with a higher number of extracranial aneurysms. CONCLUSION: Significant comorbidities are associated with extracranial aneurysms in patients with IA. Family history of extracranial aneurysms has the strongest association and suggests that IA patients with a family history of extracranial aneurysms may benefit from screening.

Genome-Wide Association Study of Treatment-Resistant Depression: Shared Biology With Metabolic Traits.

OBJECTIVE: Treatment-resistant depression (TRD) occurs in roughly one-third of all individuals with major depressive disorder (MDD). Although research has suggested a significant common variant genetic component of liability to TRD, with heritability estimated at 8% when compared with non-treatment-resistant MDD, no replicated genetic loci have been identified, and the genetic architecture of TRD remains unclear. A key barrier to this work has been the paucity of adequately powered cohorts for investigation, largely because of the challenge in prospectively investigating this phenotype. The objective of this study was to perform a well-powered genetic study of TRD. METHODS: Using receipt of electroconvulsive therapy (ECT) as a surrogate for TRD, the authors applied standard machine learning methods to electronic health record data to derive predicted probabilities of receiving ECT. These probabilities were then applied as a quantitative trait in a genome-wide association study of 154,433 genotyped patients across four large biobanks. RESULTS: Heritability estimates ranged from 2% to 4.2%, and significant genetic overlap was observed with cognition, attention deficit hyperactivity disorder, schizophrenia, alcohol and smoking traits, and body mass index. Two genome-wide significant loci were identified, both previously implicated in metabolic traits, suggesting shared biology and potential pharmacological implications. CONCLUSIONS: This work provides support for the utility of estimation of disease probability for genomic investigation and provides insights into the genetic architecture and biology of TRD.

Validation of an ICD-Code-Based Case Definition for Psychotic Illness Across Three Health Systems.

BACKGROUND AND HYPOTHESIS: Psychosis-associated diagnostic codes are increasingly being utilized as case definitions for electronic health record (EHR)-based algorithms to predict and detect psychosis. However, data on the validity of psychosis-related diagnostic codes is limited. We evaluated the positive predictive value (PPV) of International Classification of Diseases (ICD) codes for psychosis. STUDY DESIGN: Using EHRs at 3 health systems, ICD codes comprising primary psychotic disorders and mood disorders with psychosis were grouped into 5 higher-order groups. 1133 records were sampled for chart review using the full EHR. PPVs (the probability of chart-confirmed psychosis given ICD psychosis codes) were calculated across multiple treatment settings. STUDY RESULTS: PPVs across all diagnostic groups and hospital systems exceeded 70%: Mass General Brigham 0.72 [95% CI 0.68-0.77], Boston Children's Hospital 0.80 [0.75-0.84], and Boston Medical Center 0.83 [0.79-0.86]. Schizoaffective disorder PPVs were consistently the highest across sites (0.80-0.92) and major depressive disorder with psychosis were the most variable (0.57-0.79). To determine if the first documented code captured first-episode psychosis (FEP), we excluded cases with prior chart evidence of a diagnosis of or treatment for a psychotic illness, yielding substantially lower PPVs (0.08-0.62). CONCLUSIONS: We found that the first documented psychosis diagnostic code accurately captured true episodes of psychosis but was a poor index of FEP. These data have important implications for the case definitions used in the development of risk prediction models designed to predict or detect undiagnosed psychosis.

Validation of an ICD-code-based case definition for psychotic illness across three health systems.

Background and Hypothesis: Early detection of psychosis is critical for improving outcomes. Algorithms to predict or detect psychosis using electronic health record (EHR) data depend on the validity of the case definitions used, typically based on diagnostic codes. Data on the validity of psychosis-related diagnostic codes is limited. We evaluated the positive predictive value (PPV) of International Classification of Diseases (ICD) codes for psychosis. Study Design: Using EHRs at three health systems, ICD codes comprising primary psychotic disorders and mood disorders with psychosis were grouped into five higher-order groups. 1,133 records were sampled for chart review using the full EHR. PPVs (the probability of chart-confirmed psychosis given ICD psychosis codes) were calculated across multiple treatment settings. Study Results: PPVs across all diagnostic groups and hospital systems exceeded 70%: Massachusetts General Brigham 0.72 [95% CI 0.68-0.77], Boston Children's Hospital 0.80 [0.75-0.84], and Boston Medical Center 0.83 [0.79-0.86]. Schizoaffective disorder PPVs were consistently the highest across sites (0.80-0.92) and major depressive disorder with psychosis were the most variable (0.57-0.79). To determine if the first documented code captured first-episode psychosis (FEP), we excluded cases with prior chart evidence of a diagnosis of or treatment for a psychotic illness, yielding substantially lower PPVs (0.08-0.62). Conclusions: We found that the first documented psychosis diagnostic code accurately captured true episodes of psychosis but was a poor index of FEP. These data have important implications for the development of risk prediction models designed to predict or detect undiagnosed psychosis.

HistoriView: Implementation and Evaluation of a Novel Approach to Review a Patient Using a Scalable Space-Efficient Timeline without Zoom Interactions.

BACKGROUND: Timelines have been used for patient review. While maintaining a compact overview is important, merged event representations caused by the intricate and voluminous patient data bring event recognition, access ambiguity, and inefficient interaction problems. Handling large patient data efficiently is another challenge. OBJECTIVE: This study aims to develop a scalable, efficient timeline to enhance patient review for research purposes. The focus is on addressing the challenges presented by the intricate and voluminous patient data. METHODS: We propose a high-throughput, space-efficient HistoriView timeline for an individual patient. For a compact overview, it uses nonstacking event representation. An overlay detection algorithm, y-shift visualization, and popup-based interaction facilitate comprehensive analysis of overlapping datasets. An i2b2 HistoriView plugin was deployed, using split query and event reduction approaches, delivering the entire history efficiently without losing information. For evaluation, 11 participants completed a usability survey and a preference survey, followed by qualitative feedback. To evaluate scalability, 100 randomly selected patients over 60 years old were tested on the plugin and were compared with a baseline visualization. RESULTS: Most participants found that HistoriView was easy to use and learn and delivered information clearly without zooming. All preferred HistoriView over a stacked timeline. They expressed satisfaction on display, ease of learning and use, and efficiency. However, challenges and suggestions for improvement were also identified. In the performance test, the largest patient had 32,630 records, which exceeds the baseline limit. HistoriView reduced it to 2,019 visual artifacts. All patients were pulled and visualized within 45.40 seconds. Visualization took less than 3 seconds for all. DISCUSSION AND CONCLUSION: HistoriView allows complete data exploration without exhaustive interactions in a compact overview. It is useful for dense data or iterative comparisons. However, issues in exploring subconcept records were reported. HistoriView handles large patient data preserving original information in a reasonable time.

Offspring cardiometabolic outcomes and postnatal growth trajectories after exposure to maternal SARS-CoV-2 infection.

OBJECTIVE: The objective of this study is to determine whether in utero exposure to SARS-CoV-2 is associated with increased risk for a cardiometabolic diagnosis by 18 months of age. METHODS: This retrospective electronic health record (EHR)-based cohort study included the live-born offspring of all individuals who delivered during the COVID-19 pandemic (April 1, 2020-December 31, 2021) at eight hospitals in Massachusetts. Offspring exposure was defined as a positive maternal SARS-CoV-2 polymerase chain reaction test during pregnancy. The primary outcome was presence of an ICD-10 code for a cardiometabolic disorder in offspring EHR by 18 months. Weight-, length-, and BMI-for-age z scores were calculated and compared at 6-month intervals from birth to 18 months. RESULTS: A total of 29,510 offspring (1599 exposed and 27,911 unexposed) were included. By 18 months, 6.7% of exposed and 4.4% of unexposed offspring had received a cardiometabolic diagnosis (crude odds ratio [OR] 1.47 [95% CI: 1.10 to 1.94], p = 0.007; adjusted OR 1.38 [1.06 to 1.77], p = 0.01). Exposed offspring had a significantly greater mean BMI-for-age z score versus unexposed offspring at 6 months (z score difference 0.19 [95% CI: 0.10 to 0.29], p < 0.001; adjusted difference 0.04 [-0.06 to 0.13], p = 0.4). CONCLUSIONS: Exposure to maternal SARS-CoV-2 infection was associated with an increased risk of receiving a cardiometabolic diagnosis by 18 months preceded by greater BMI-for-age at 6 months.

LATTE: Label-efficient incident phenotyping from longitudinal electronic health records.

Electronic health record (EHR) data are increasingly used to support real-world evidence studies but are limited by the lack of precise timings of clinical events. Here, we propose a label-efficient incident phenotyping (LATTE) algorithm to accurately annotate the timing of clinical events from longitudinal EHR data. By leveraging the pre-trained semantic embeddings, LATTE selects predictive features and compresses their information into longitudinal visit embeddings through visit attention learning. LATTE models the sequential dependency between the target event and visit embeddings to derive the timings. To improve label efficiency, LATTE constructs longitudinal silver-standard labels from unlabeled patients to perform semi-supervised training. LATTE is evaluated on the onset of type 2 diabetes, heart failure, and relapses of multiple sclerosis. LATTE consistently achieves substantial improvements over benchmark methods while providing high prediction interpretability. The event timings are shown to help discover risk factors of heart failure among patients with rheumatoid arthritis.

Development and multi-site external validation of a generalizable risk prediction model for bipolar disorder.

Bipolar disorder is a leading contributor to disability, premature mortality, and suicide. Early identification of risk for bipolar disorder using generalizable predictive models trained on diverse cohorts around the United States could improve targeted assessment of high risk individuals, reduce misdiagnosis, and improve the allocation of limited mental health resources. This observational case-control study intended to develop and validate generalizable predictive models of bipolar disorder as part of the multisite, multinational PsycheMERGE Network across diverse and large biobanks with linked electronic health records (EHRs) from three academic medical centers: in the Northeast (Massachusetts General Brigham), the Mid-Atlantic (Geisinger) and the Mid-South (Vanderbilt University Medical Center). Predictive models were developed and valid with multiple algorithms at each study site: random forests, gradient boosting machines, penalized regression, including stacked ensemble learning algorithms combining them. Predictors were limited to widely available EHR-based features agnostic to a common data model including demographics, diagnostic codes, and medications. The main study outcome was bipolar disorder diagnosis as defined by the International Cohort Collection for Bipolar Disorder, 2015. In total, the study included records for 3,529,569 patients including 12,533 cases (0.3%) of bipolar disorder. After internal and external validation, algorithms demonstrated optimal performance in their respective development sites. The stacked ensemble achieved the best combination of overall discrimination (AUC = 0.82-0.87) and calibration performance with positive predictive values above 5% in the highest risk quantiles at all three study sites. In conclusion, generalizable predictive models of risk for bipolar disorder can be feasibly developed across diverse sites to enable precision medicine. Comparison of a range of machine learning methods indicated that an ensemble approach provides the best performance overall but required local retraining. These models will be disseminated via the PsycheMERGE Network website.