RESUMEN
OBJECTIVE: This study aimed to develop a novel approach using routinely collected electronic health records (EHRs) data to improve the prediction of a rare event. We illustrated this using an example of improving early prediction of an autism diagnosis, given its low prevalence, by leveraging correlations between autism and other neurodevelopmental conditions (NDCs). METHODS: To achieve this, we introduced a conditional multi-label model by merging conditional learning and multi-label methodologies. The conditional learning approach breaks a hard task into more manageable pieces in each stage, and the multi-label approach utilizes information from related neurodevelopmental conditions to learn predictive latent features. The study involved forecasting autism diagnosis by age 5.5 years, utilizing data from the first 18 months of life, and the analysis of feature importance correlations to explore the alignment within the feature space across different conditions. RESULTS: Upon analysis of health records from 18,156 children, we are able to generate a model that predicts a future autism diagnosis with moderate performance (AUROC=0.76). The proposed conditional multi-label method significantly improves predictive performance with an AUROC of 0.80 (p < 0.001). Further examination shows that both the conditional and multi-label approach alone provided marginal lift to the model performance compared to a one-stage one-label approach. We also demonstrated the generalizability and applicability of this method using simulated data with high correlation between feature vectors for different labels. CONCLUSION: Our findings underscore the effectiveness of the developed conditional multi-label model for early prediction of an autism diagnosis. The study introduces a versatile strategy applicable to prediction tasks involving limited target populations but sharing underlying features or etiology among related groups.
RESUMEN
BACKGROUND: Patients with pulmonary hypertension have a high risk of maternal morbidity and mortality. It is unknown if a trial of labor carries a lower risk of morbidity in these patients compared to a planned cesarean delivery. The objective of this study was to examine the association of delivery mode with severe maternal morbidity events during delivery hospitalization among patients with pulmonary hypertension. METHODS: This retrospective cohort study used the Premier inpatient administrative database. Patients delivering ≥25 weeks gestation from January 1, 2016, to September 30, 2020, and with pulmonary hypertension were included. The primary analysis compared intended vaginal delivery (ie, trial of labor) to intended cesarean delivery (intention to treat analysis). A sensitivity analysis was conducted comparing vaginal delivery to cesarean delivery (as treated analysis). The primary outcome was nontransfusion severe maternal morbidity during the delivery hospitalization. Secondary outcomes included blood transfusion (4 or more units) and readmission to the delivery hospital within 90 days from discharge from delivery hospitalization. RESULTS: The cohort consisted of 727 deliveries. In the primary analysis, there was no difference in nontransfusion morbidity between intended vaginal delivery and intended cesarean delivery groups (adjusted odds ratio [aOR], 0.75; 95% confidence interval [CI], 0.49-1.15). In secondary analyses, intended cesarean delivery was not associated with blood transfusion (aOR, 0.71; 95% CI, 0.34-1.50) or readmission within 90 days (aOR, 0.60; 95% CI, 0.32-1.14). In the sensitivity analysis, cesarean delivery was associated with a 3-fold higher risk of nontransfusion morbidity compared to vaginal delivery (aOR, 2.64; 95% CI, 1.54-3.93), a 3-fold higher risk of blood transfusion (aOR, 3.06; 95% CI, 1.17-7.99), and a 2-fold higher risk of readmission within 90 days (aOR, 2.20; 95% CI, 1.09-4.46) compared to vaginal delivery. CONCLUSIONS: Among pregnant patients with pulmonary hypertension, a trial of labor was not associated with a higher risk of morbidity compared to an intended cesarean delivery. One-third of patients who required an intrapartum cesarean delivery had a morbidity event, demonstrating the increased risk of adverse events in this group.
Asunto(s)
Hipertensión Pulmonar , Embarazo , Femenino , Humanos , Estudios Retrospectivos , Hipertensión Pulmonar/diagnóstico , Hipertensión Pulmonar/epidemiología , Hipertensión Pulmonar/terapia , Parto Obstétrico/efectos adversos , Cesárea/efectos adversos , PartoRESUMEN
BACKGROUND: Electronic Health Records (EHR) are widely used to develop clinical prediction models (CPMs). However, one of the challenges is that there is often a degree of informative missing data. For example, laboratory measures are typically taken when a clinician is concerned that there is a need. When data are the so-called Not Missing at Random (NMAR), analytic strategies based on other missingness mechanisms are inappropriate. In this work, we seek to compare the impact of different strategies for handling missing data on CPMs performance. METHODS: We considered a predictive model for rapid inpatient deterioration as an exemplar implementation. This model incorporated twelve laboratory measures with varying levels of missingness. Five labs had missingness rate levels around 50%, and the other seven had missingness levels around 90%. We included them based on the belief that their missingness status can be highly informational for the prediction. In our study, we explicitly compared the various missing data strategies: mean imputation, normal-value imputation, conditional imputation, categorical encoding, and missingness embeddings. Some of these were also combined with the last observation carried forward (LOCF). We implemented logistic LASSO regression, multilayer perceptron (MLP), and long short-term memory (LSTM) models as the downstream classifiers. We compared the AUROC of testing data and used bootstrapping to construct 95% confidence intervals. RESULTS: We had 105,198 inpatient encounters, with 4.7% having experienced the deterioration outcome of interest. LSTM models generally outperformed other cross-sectional models, where embedding approaches and categorical encoding yielded the best results. For the cross-sectional models, normal-value imputation with LOCF generated the best results. CONCLUSION: Strategies that accounted for the possibility of NMAR missing data yielded better model performance than those did not. The embedding method had an advantage as it did not require prior clinical knowledge. Using LOCF could enhance the performance of cross-sectional models but have countereffects in LSTM models.
Asunto(s)
Registros Electrónicos de Salud , Humanos , Deterioro Clínico , Modelos Estadísticos , Técnicas de Laboratorio ClínicoRESUMEN
Recent work has shown that predictive models can be applied to structured electronic health record (EHR) data to stratify autism likelihood from an early age (<1 year). Integrating clinical narratives (or notes) with structured data has been shown to improve prediction performance in other clinical applications, but the added predictive value of this information in early autism prediction has not yet been explored. In this study, we aimed to enhance the performance of early autism prediction by using both structured EHR data and clinical narratives. We built models based on structured data and clinical narratives separately, and then an ensemble model that integrated both sources of data. We assessed the predictive value of these models from Duke University Health System over a 14-year span to evaluate ensemble models predicting later autism diagnosis (by age 4â¯years) from data collected from ages 30 to 360â¯days. Our sample included 11,750 children above by age 3â¯years (385 meeting autism diagnostic criteria). The ensemble model for autism prediction showed superior performance and at age 30â¯days achieved 46.8% sensitivity (95% confidence interval, CI: 22.0%, 52.9%), 28.0% positive predictive value (PPV) at high (90%) specificity (CI: 2.0%, 33.1%), and AUC4 (with at least 4-year follow-up for controls) reaching 0.769 (CI: 0.715, 0.811). Prediction by 360â¯days achieved 44.5% sensitivity (CI: 23.6%, 62.9%), and 13.7% PPV at high (90%) specificity (CI: 9.6%, 18.9%), and AUC4 reaching 0.797 (CI: 0.746, 0.840). Results show that incorporating clinical narratives in early autism prediction achieved promising accuracy by age 30â¯days, outperforming models based on structured data only. Furthermore, findings suggest that additional features learned from clinician narratives might be hypothesis generating for understanding early development in autism.
Asunto(s)
Trastorno Autístico , Registros Electrónicos de Salud , Niño , Humanos , Lactante , Preescolar , Trastorno Autístico/diagnóstico , Valor Predictivo de las Pruebas , Narración , ElectrónicaRESUMEN
Importance: Stroke is the fifth-highest cause of death in the US and a leading cause of serious long-term disability with particularly high risk in Black individuals. Quality risk prediction algorithms, free of bias, are key for comprehensive prevention strategies. Objective: To compare the performance of stroke-specific algorithms with pooled cohort equations developed for atherosclerotic cardiovascular disease for the prediction of new-onset stroke across different subgroups (race, sex, and age) and to determine the added value of novel machine learning techniques. Design, Setting, and Participants: Retrospective cohort study on combined and harmonized data from Black and White participants of the Framingham Offspring, Atherosclerosis Risk in Communities (ARIC), Multi-Ethnic Study for Atherosclerosis (MESA), and Reasons for Geographical and Racial Differences in Stroke (REGARDS) studies (1983-2019) conducted in the US. The 62â¯482 participants included at baseline were at least 45 years of age and free of stroke or transient ischemic attack. Exposures: Published stroke-specific algorithms from Framingham and REGARDS (based on self-reported risk factors) as well as pooled cohort equations for atherosclerotic cardiovascular disease plus 2 newly developed machine learning algorithms. Main Outcomes and Measures: Models were designed to estimate the 10-year risk of new-onset stroke (ischemic or hemorrhagic). Discrimination concordance index (C index) and calibration ratios of expected vs observed event rates were assessed at 10 years. Analyses were conducted by race, sex, and age groups. Results: The combined study sample included 62â¯482 participants (median age, 61 years, 54% women, and 29% Black individuals). Discrimination C indexes were not significantly different for the 2 stroke-specific models (Framingham stroke, 0.72; 95% CI, 0.72-073; REGARDS self-report, 0.73; 95% CI, 0.72-0.74) vs the pooled cohort equations (0.72; 95% CI, 0.71-0.73): differences 0.01 or less (P values >.05) in the combined sample. Significant differences in discrimination were observed by race: the C indexes were 0.76 for all 3 models in White vs 0.69 in Black women (all P values <.001) and between 0.71 and 0.72 in White men and between 0.64 and 0.66 in Black men (all P values ≤.001). When stratified by age, model discrimination was better for younger (<60 years) vs older (≥60 years) adults for both Black and White individuals. The ratios of observed to expected 10-year stroke rates were closest to 1 for the REGARDS self-report model (1.05; 95% CI, 1.00-1.09) and indicated risk overestimation for Framingham stroke (0.86; 95% CI, 0.82-0.89) and pooled cohort equations (0.74; 95% CI, 0.71-0.77). Performance did not significantly improve when novel machine learning algorithms were applied. Conclusions and Relevance: In this analysis of Black and White individuals without stroke or transient ischemic attack among 4 US cohorts, existing stroke-specific risk prediction models and novel machine learning techniques did not significantly improve discriminative accuracy for new-onset stroke compared with the pooled cohort equations, and the REGARDS self-report model had the best calibration. All algorithms exhibited worse discrimination in Black individuals than in White individuals, indicating the need to expand the pool of risk factors and improve modeling techniques to address observed racial disparities and improve model performance.
Asunto(s)
Población Negra , Disparidades en Atención de Salud , Prejuicio , Medición de Riesgo , Accidente Cerebrovascular , Población Blanca , Femenino , Humanos , Masculino , Persona de Mediana Edad , Aterosclerosis/epidemiología , Enfermedades Cardiovasculares/epidemiología , Ataque Isquémico Transitorio/epidemiología , Estudios Retrospectivos , Accidente Cerebrovascular/diagnóstico , Accidente Cerebrovascular/epidemiología , Accidente Cerebrovascular/etnología , Medición de Riesgo/normas , Reproducibilidad de los Resultados , Factores Sexuales , Factores de Edad , Factores Raciales/estadística & datos numéricos , Población Negra/estadística & datos numéricos , Población Blanca/estadística & datos numéricos , Estados Unidos/epidemiología , Aprendizaje Automático/normas , Sesgo , Prejuicio/prevención & control , Disparidades en Atención de Salud/etnología , Disparidades en Atención de Salud/normas , Disparidades en Atención de Salud/estadística & datos numéricos , Simulación por Computador/normas , Simulación por Computador/estadística & datos numéricosRESUMEN
BACKGROUND/OBJECTIVES: Sleep measures, such as duration and onset timing, are associated with adiposity outcomes among children. Recent research among adults has considered variability in sleep and wake onset times, with the Sleep Regularity Index (SRI) as a comprehensive metric to measure shifts in sleep and wake onset times between days. However, little research has examined regularity and adiposity outcomes among children. This study examined the associations of three sleep measures (i.e., sleep duration, sleep onset time, and SRI) with three measures of adiposity (i.e., body mass index [BMI], waist circumference, and waist-to-height ratio [WHtR]) in a pediatric sample. SUBJECTS/METHODS: Children (ages 4-13 years) who were part of the U.S. Newborn Epigenetic STudy (NEST) participated. Children (N = 144) wore an ActiGraph for 1 week. Sleep measures were estimated from actigraphy data. Weight, height, and waist circumference were measured by trained researchers. BMI and WHtR was calculated with the objectively measured waist and height values. Multiple linear regression models examined associations between child sleep and adiposity outcomes, controlling for race/ethnicity, child sex, age, mothers' BMI and sleep duration. RESULTS: When considering sleep onset timing and duration, along with demographic covariates, sleep onset timing was not significantly associated with any of the three adiposity measures, but a longer duration was significantly associated with a lower BMI Z-score (ß = -0.29, p < 0.001), waist circumference (ß = -0.31, p < 0.001), and WHtR (ß = -0.38, p < 0.001). When considering SRI and duration, duration remained significantly associated with the adiposity measures. The SRI and adiposity associations were in the expected direction, but were non-significant, except the SRI and WHtR association (ß = -0.16, p = 0.077) was marginally non-significant. CONCLUSIONS: Sleep duration was consistently associated with adiposity measures in children 4-13 years of age. Pediatric sleep interventions should focus first on elongating nighttime sleep duration, and examine if this improves child adiposity outcomes.
Asunto(s)
Adiposidad , Sueño , Adolescente , Adulto , Índice de Masa Corporal , Niño , Preescolar , Humanos , Recién Nacido , Obesidad , Circunferencia de la CinturaRESUMEN
BACKGROUND: Web-based crowdfunding has become a popular method to raise money for medical expenses, and there is growing research interest in this topic. However, crowdfunding data are largely composed of unstructured text, thereby posing many challenges for researchers hoping to answer questions about specific medical conditions. Previous studies have used methods that either failed to address major challenges or were poorly scalable to large sample sizes. To enable further research on this emerging funding mechanism in health care, better methods are needed. OBJECTIVE: We sought to validate an algorithm for identifying 11 disease categories in web-based medical crowdfunding campaigns. We hypothesized that a disease identification algorithm combining a named entity recognition (NER) model and word search approach could identify disease categories with high precision and accuracy. Such an algorithm would facilitate further research using these data. METHODS: Web scraping was used to collect data on medical crowdfunding campaigns from GoFundMe (GoFundMe Inc). Using pretrained NER and entity resolution models from Spark NLP for Healthcare in combination with targeted keyword searches, we constructed an algorithm to identify conditions in the campaign descriptions, translate conditions to International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes, and predict the presence or absence of 11 disease categories in the campaigns. The classification performance of the algorithm was evaluated against 400 manually labeled campaigns. RESULTS: We collected data on 89,645 crowdfunding campaigns through web scraping. The interrater reliability for detecting the presence of broad disease categories in the campaign descriptions was high (Cohen κ: range 0.69-0.96). The NER and entity resolution models identified 6594 unique (276,020 total) ICD-10-CM codes among all of the crowdfunding campaigns in our sample. Through our word search, we identified 3261 additional campaigns for which a medical condition was not otherwise detected with the NER model. When averaged across all disease categories and weighted by the number of campaigns that mentioned each disease category, the algorithm demonstrated an overall precision of 0.83 (range 0.48-0.97), a recall of 0.77 (range 0.42-0.98), an F1 score of 0.78 (range 0.56-0.96), and an accuracy of 95% (range 90%-98%). CONCLUSIONS: A disease identification algorithm combining pretrained natural language processing models and ICD-10-CM code-based disease categorization was able to detect 11 disease categories in medical crowdfunding campaigns with high precision and accuracy.
Asunto(s)
Colaboración de las Masas , Algoritmos , Colaboración de las Masas/métodos , Atención a la Salud , Humanos , Reproducibilidad de los ResultadosRESUMEN
An extensive epidemiological literature indicates that increased exposure to tobacco retail outlets (TROs) places never smokers at greater risk for smoking uptake and current smokers at greater risk for increased consumption and smoking relapse. Yet research into the mechanisms underlying this effect has been limited. This preliminary study represents the first effort to examine the neurobiological consequences of exposure to personally relevant TROs among both smokers (n = 17) and nonsmokers (n = 17). Individuals carried a global positioning system (GPS) tracker for 2 weeks. Traces were used to identify TROs and control outlets that fell inside and outside their ideographically defined activity space. Participants underwent functional MRI (fMRI) scanning during which they were presented with images of these storefronts, along with similar store images from a different county and rated their familiarity with these stores. The main effect of activity space was additive with a Smoking status × Store type interaction, resulting in smokers exhibiting greater neural activation to TROs falling inside activity space within the parahippocampus, precuneus, medial prefrontal cortex, and dorsal anterior insula. A similar pattern was observed for familiarity ratings. Together, these preliminary findings suggest that the otherwise distinct neural systems involved in self-orientation/self-relevance and smoking motivation may act in concert and underlie TRO influence on smoking behavior. This study also offers a novel methodological framework for evaluating the influence of community features on neural activity that can be readily adapted to study other health behaviors.
Asunto(s)
Fumar Cigarrillos/psicología , Mercadotecnía , Fumadores/psicología , Productos de Tabaco , Tabaquismo/diagnóstico por imagen , Adolescente , Adulto , Femenino , Humanos , Imagen por Resonancia Magnética , Masculino , Persona de Mediana Edad , Motivación , Fumar , Adulto JovenRESUMEN
BACKGROUND: Viewing their habitual smoking environments increases smokers' craving and smoking behaviors in laboratory settings. A deep learning approach can differentiate between habitual smoking versus nonsmoking environments, suggesting that it may be possible to predict environment-associated smoking risk from continuously acquired images of smokers' daily environments. OBJECTIVE: In this study, we aim to predict environment-associated risk from continuously acquired images of smokers' daily environments. We also aim to understand how model performance varies by location type, as reported by participants. METHODS: Smokers from Durham, North Carolina and surrounding areas completed ecological momentary assessments both immediately after smoking and at randomly selected times throughout the day for 2 weeks. At each assessment, participants took a picture of their current environment and completed a questionnaire on smoking, craving, and the environmental setting. A convolutional neural network-based model was trained to predict smoking, craving, whether smoking was permitted in the current environment and whether the participant was outside based on images of participants' daily environments, the time since their last cigarette, and baseline data on daily smoking habits. Prediction performance, quantified using the area under the receiver operating characteristic curve (AUC) and average precision (AP), was assessed for out-of-sample prediction as well as personalized models trained on images from days 1 to 10. The models were optimized for mobile devices and implemented as a smartphone app. RESULTS: A total of 48 participants completed the study, and 8008 images were acquired. The personalized models were highly effective in predicting smoking risk (AUC=0.827; AP=0.882), craving (AUC=0.837; AP=0.798), whether smoking was permitted in the current environment (AUC=0.932; AP=0.981), and whether the participant was outside (AUC=0.977; AP=0.956). The out-of-sample models were also effective in predicting smoking risk (AUC=0.723; AP=0.785), whether smoking was permitted in the current environment (AUC=0.815; AP=0.937), and whether the participant was outside (AUC=0.949; AP=0.922); however, they were not effective in predicting craving (AUC=0.522; AP=0.427). Omitting image features reduced AUC by over 0.1 when predicting all outcomes except craving. Prediction of smoking was more effective for participants whose self-reported location type was more variable (Spearman ρ=0.48; P=.001). CONCLUSIONS: Images of daily environments can be used to effectively predict smoking risk. Model personalization, achieved by incorporating information about daily smoking habits and training on participant-specific images, further improves prediction performance. Environment-associated smoking risk can be assessed in real time on a mobile device and can be incorporated into device-based smoking cessation interventions.
Asunto(s)
Cese del Hábito de Fumar , Productos de Tabaco , Humanos , Fumadores , Fumar , Fumar TabacoRESUMEN
PURPOSE OF REVIEW: Individuals with attention-deficit hyperactivity disorder (ADHD) may be unusually sensitive to screen media technology (SMT), from television to mobile devices. Although an association between ADHD and SMT use has been confirmed, its importance is uncertain partly due to variability in the way SMT has been conceptualized and measured. Here, we identify distinct, quantifiable dimensions of SMT use and review possible links to ADHD to facilitate more precise, reproducible investigation. RECENT FINDINGS: Display characteristics, media multitasking, device notifications, SMT addiction, and media content all may uniquely impact the ADHD phenotype. Each can be investigated with a digital health approach and counteracted with device-based interventions. Novel digital therapeutics for ADHD demonstrate that specific forms of SMT can also have positive effects. Further study should quantify how distinct dimensions of SMT use relate to ADHD. SMT devices themselves can serve as a self-monitoring study platform and deliver digital interventions.
Asunto(s)
Trastorno por Déficit de Atención con Hiperactividad/psicología , Conducta Adictiva , Trastorno por Déficit de Atención con Hiperactividad/terapia , Humanos , Tiempo de PantallaRESUMEN
BACKGROUND: Fatigue is a prevalent and functionally disabling symptom for individuals living with multiple sclerosis (MS) which is poorly understood and multifactorial in etiology. Bladder dysfunction is another common MS symptom which limits social engagement and quality of life. To manage bladder issues, individuals with MS tend to limit their fluid intake, which may contribute to a low-hydration (LoH) state and fatigue. OBJECTIVE: To evaluate the relationship between patient-reported MS fatigue, bladder dysfunction, and hydration status. METHODS: We performed a prospective cross-sectional study in 50 women with MS. Participants submitted a random urine sample and completed several fatigue-related surveys. Using a urine specific gravity (USG) threshold of 1.015, we classified MS subjects into two groups: high-hydration (HiH) and LoH states. RESULTS: LoH status was more common in MS subjects with bladder dysfunction. Statistically significant differences in self-reported Fatigue Performance Scale were observed between HiH and LoH subjects (p = 0.022). USG was significantly correlated with fatigue as measured by the MS Fatigue Severity Scale (FSS) score (r = 0.328, p = 0.020). CONCLUSION: Hydration status correlates with self-reported fatigue, with lower fatigue scores found in those with HiH status (USG < 1.015).
Asunto(s)
Deshidratación/fisiopatología , Esclerosis Múltiple/fisiopatología , Vejiga Urinaria Neurogénica/fisiopatología , Equilibrio Hidroelectrolítico , Adulto , Estudios Transversales , Deshidratación/epidemiología , Conducta de Ingestión de Líquido , Fatiga , Femenino , Humanos , Persona de Mediana Edad , Esclerosis Múltiple/epidemiología , Estudios Prospectivos , Índice de Severidad de la Enfermedad , Vejiga Urinaria Neurogénica/epidemiología , Incontinencia Urinaria/epidemiología , Incontinencia Urinaria/fisiopatologíaRESUMEN
BACKGROUND: The Multiple Sclerosis Walking Scale (MSWS-12) is the predominant patient-reported measure of multiple sclerosis (MS) -elated walking ability, yet it had not been analyzed using item response theory (IRT), the emerging standard for patient-reported outcome (PRO) validation. This study aims to reduce MSWS-12 measurement error and facilitate computerized adaptive testing by creating an IRT model of the MSWS-12 and distributing it online. METHODS: MSWS-12 responses from 284 subjects with MS were collected by mail and used to fit and compare several IRT models. Following model selection and assessment, subpopulations based on age and sex were tested for differential item functioning (DIF). RESULTS: Model comparison favored a one-dimensional graded response model (GRM). This model met fit criteria and explained 87 % of response variance. The performance of each MSWS-12 item was characterized using category response curves (CRCs) and item information. IRT-based MSWS-12 scores correlated with traditional MSWS-12 scores (r = 0.99) and timed 25-foot walk (T25FW) speed (r = -0.70). Item 2 showed DIF based on age (χ 2 = 19.02, df = 5, p < 0.01), and Item 11 showed DIF based on sex (χ 2 = 13.76, df = 5, p = 0.02). CONCLUSIONS: MSWS-12 measurement error depends on walking ability, but could be lowered by improving or replacing items with low information or DIF. The e-MSWS-12 includes IRT-based scoring, error checking, and an estimated T25FW derived from MSWS-12 responses. It is available at https://ms-irt.shinyapps.io/e-MSWS-12 .
Asunto(s)
Evaluación de la Discapacidad , Esclerosis Múltiple/terapia , Perfil de Impacto de Enfermedad , Caminata/fisiología , Adulto , Anciano , Anciano de 80 o más Años , Femenino , Humanos , Masculino , Persona de Mediana EdadRESUMEN
Recently developed survival analysis methods improve upon existing approaches by predicting the probability of event occurrence in each of a number pre-specified (discrete) time intervals. By avoiding placing strong parametric assumptions on the event density, this approach tends to improve prediction performance, particularly when data are plentiful. However, in clinical settings with limited available data, it is often preferable to judiciously partition the event time space into a limited number of intervals well suited to the prediction task at hand. In this work, we develop Adaptive Discretization for Event PredicTion (ADEPT) to learn from data a set of cut points defining such a partition. We show that in two simulated datasets, we are able to recover intervals that match the underlying generative model. We then demonstrate improved prediction performance on three real-world observational datasets, including a large, newly harmonized stroke risk prediction dataset. Finally, we argue that our approach facilitates clinical decision-making by suggesting time intervals that are most appropriate for each task, in the sense that they facilitate more accurate risk prediction.
RESUMEN
Purpose: To evaluate the diagnostic performance of a robotically aligned optical coherence tomography (RAOCT) system coupled with a deep learning model in detecting referable posterior segment pathology in OCT images of emergency department patients. Methods: A deep learning model, RobOCTNet, was trained and internally tested to classify OCT images as referable versus non-referable for ophthalmology consultation. For external testing, emergency department patients with signs or symptoms warranting evaluation of the posterior segment were imaged with RAOCT. RobOCTNet was used to classify the images. Model performance was evaluated against a reference standard based on clinical diagnosis and retina specialist OCT review. Results: We included 90,250 OCT images for training and 1489 images for internal testing. RobOCTNet achieved an area under the curve (AUC) of 1.00 (95% confidence interval [CI], 0.99-1.00) for detection of referable posterior segment pathology in the internal test set. For external testing, RAOCT was used to image 72 eyes of 38 emergency department patients. In this set, RobOCTNet had an AUC of 0.91 (95% CI, 0.82-0.97), a sensitivity of 95% (95% CI, 87%-100%), and a specificity of 76% (95% CI, 62%-91%). The model's performance was comparable to two human experts' performance. Conclusions: A robotically aligned OCT coupled with a deep learning model demonstrated high diagnostic performance in detecting referable posterior segment pathology in a cohort of emergency department patients. Translational Relevance: Robotically aligned OCT coupled with a deep learning model may have the potential to improve emergency department patient triage for ophthalmology referral.
Asunto(s)
Aprendizaje Profundo , Humanos , RetinaRESUMEN
Sex differences in the age of autism diagnosis during childhood have been documented consistently but remain poorly understood. In this study, we used electronic health records data from a diverse, academic medical center to quantify differences in the age of autism diagnosis between boys and girls and identify associations between the age of diagnosis and co-occurring neurodevelopmental, psychiatric, and medical conditions. An established computable phenotype was used to identify all autism diagnoses within the Duke University Health System between 2014 and 2021. Co-occurring neurodevelopmental and psychiatric diagnoses as well as visits to specific medical and supportive services were identified in the 2 years prior to the autism diagnosis. Cox proportional hazards models were fitted to quantify associations between diagnosis age and sex with and without controlling for the presence of each co-occurring diagnosis and visit type. Records from 1438 individuals (1142 boys and 296 girls) were included. Girls were more likely to be diagnosed either before age 3 ( χ 2 = 497.720, p < 0.001) or after age 11 ( χ 2 = 4.014, p = 0.047), whereas boys were more likely to be diagnosed between ages 3 and 11 ( χ 2 = 5.532, p = 0.019). Visits for anxiety ( χ 2 = 4.200, p = 0.040) and mood disorders ( χ 2 = 7.033, p = 0.008) were more common in girls and associated with later autism diagnosis (HR = 0.615, p < 0.001; and HR = 0.493, p < 0.001). Visits for otolaryngology were more common in boys and associated with an earlier autism diagnosis (HR = 1.691, p < 0.001). After controlling for these conditions, associations between sex and diagnosis age were reduced and not statistically significant. These results show that the age of autism diagnosis differs in girls compared to boys, but these differences were neutralized when controlling for co-occurring neurodevelopmental and psychiatric conditions prior to autism diagnosis. Understanding sex differences and the possible mediating role of other diagnoses may suggest targets for intervention to promote earlier and more equitable diagnosis.
Asunto(s)
Trastorno del Espectro Autista , Trastorno Autístico , Trastornos Generalizados del Desarrollo Infantil , Niño , Humanos , Masculino , Femenino , Preescolar , Trastorno del Espectro Autista/complicaciones , Trastorno del Espectro Autista/diagnóstico , Trastorno del Espectro Autista/epidemiología , Caracteres Sexuales , AnsiedadRESUMEN
BACKGROUND: Women with cardiomyopathies are at risk for pregnancy complications. The optimal mode of delivery in these patients is guided by expert opinion and limited small studies. OBJECTIVES: The objective of this study is to examine the association of delivery mode with severe maternal morbidity events during delivery hospitalization and readmissions among patients with cardiomyopathies. METHODS: The Premier inpatient administrative database was used to conduct a retrospective cohort study of pregnant patients with a diagnosis of a cardiomyopathy. Utilizing a target trial emulation strategy, the primary analysis compared outcomes among patients exposed to intended vaginal delivery vs intended cesarean delivery (intention to treat). A secondary analysis compared outcomes among patients who delivered vaginally vs by cesarean (as-treated). Outcomes examined were nontransfusion severe maternal morbidity during the delivery hospitalization, blood transfusion, and readmission. RESULTS: The cohort consisted of 2,921 deliveries. In the primary analysis (intention to treat), there was no difference in nontransfusion morbidity (adjusted OR [aOR]: 1.17; 95% CI: 0.91-1.51), blood transfusion (aOR: 1.27; 95% CI: 0.81-1.98), or readmission (aOR: 1.03; 95% CI: 0.73-1.44) between intended vaginal delivery and intended cesarean delivery. In the as-treated analysis, cesarean delivery was associated with a 2-fold higher risk of nontransfusion morbidity (aOR: 2.44; 95% CI: 1.85-3.22) and blood transfusion (aOR: 2.26; 95% CI: 1.34-3.81) when compared with vaginal delivery. CONCLUSIONS: In patients with cardiomyopathies, a trial of labor does not confer a higher risk of maternal morbidity, blood transfusion, or readmission compared with planned cesarean delivery.
Asunto(s)
Cardiomiopatías , Insuficiencia Cardíaca , Embarazo , Humanos , Femenino , Estudios Retrospectivos , Insuficiencia Cardíaca/etiología , Parto Obstétrico , Cesárea/efectos adversos , Cardiomiopatías/epidemiología , Cardiomiopatías/etiologíaRESUMEN
Importance: Autism detection early in childhood is critical to ensure that autistic children and their families have access to early behavioral support. Early correlates of autism documented in electronic health records (EHRs) during routine care could allow passive, predictive model-based monitoring to improve the accuracy of early detection. Objective: To quantify the predictive value of early autism detection models based on EHR data collected before age 1 year. Design, Setting, and Participants: This retrospective diagnostic study used EHR data from children seen within the Duke University Health System before age 30 days between January 2006 and December 2020. These data were used to train and evaluate L2-regularized Cox proportional hazards models predicting later autism diagnosis based on data collected from birth up to the time of prediction (ages 30-360 days). Statistical analyses were performed between August 1, 2020, and April 1, 2022. Main Outcomes and Measures: Prediction performance was quantified in terms of sensitivity, specificity, and positive predictive value (PPV) at clinically relevant model operating thresholds. Results: Data from 45â¯080 children, including 924 (1.5%) meeting autism criteria, were included in this study. Model-based autism detection at age 30 days achieved 45.5% sensitivity and 23.0% PPV at 90.0% specificity. Detection by age 360 days achieved 59.8% sensitivity and 17.6% PPV at 81.5% specificity and 38.8% sensitivity and 31.0% PPV at 94.3% specificity. Conclusions and Relevance: In this diagnostic study of an autism screening test, EHR-based autism detection achieved clinically meaningful accuracy by age 30 days, improving by age 1 year. This automated approach could be integrated with caregiver surveys to improve the accuracy of early autism screening.
Asunto(s)
Trastorno Autístico , Niño , Humanos , Adulto , Lactante , Trastorno Autístico/diagnóstico , Trastorno Autístico/epidemiología , Registros Electrónicos de Salud , Estudios Retrospectivos , Valor Predictivo de las Pruebas , Encuestas y CuestionariosRESUMEN
The mixture cure model allows failure probability to be estimated separately from failure timing in settings wherein failure never occurs in a subset of the population. In this paper, we draw on insights from representation learning and causal inference to develop a neural network based mixture cure model that is free of distributional assumptions, yielding improved prediction of failure timing, yet still effectively disentangles information about failure timing from information about failure probability. Our approach also mitigates effects of selection biases in the observation of failure and censoring times on estimation of the failure density and censoring density, respectively. Results suggest this approach could be applied to distinguish factors predicting failure occurrence versus timing and mitigate biases in real-world observational datasets.
RESUMEN
STUDY OBJECTIVES: Caffeine use is ubiquitous among adolescents and may be harmful to sleep, with downstream implications for health and development. Research has been limited by self-reported and/or aggregated measures of sleep and caffeine collected at a single time point. This study examines bidirectional associations between daily caffeine consumption and electroencephalogram-measured sleep among adolescents and explores whether these relationships depend on timing of caffeine use. METHODS: Ninety-eight adolescents aged 11-17 (mean =14.38, standard deviation = 1.77; 50% female) participated in 7 consecutive nights of at-home sleep electroencephalography and completed a daily diary querying morning, afternoon, and evening caffeine use. Linear mixed-effects regressions examined relationships between caffeine consumption and total sleep time, sleep-onset latency, sleep efficiency, wake after sleep onset, and time spent in sleep stages. Impact of sleep indices on next-day caffeine use was also examined. RESULTS: Increased total caffeine consumption was associated was increased sleep-onset latency (ß = .13; 95% CI = .06, .21; P < .001) and reduced total sleep time (ß = -.17; 95% confidence interval [CI] = -.31, -.02; P = .02), sleep efficiency (ß = -1.59; 95% CI = -2.51, -.67; P < .001), and rapid eye movement sleep (ß = -.12; 95% CI = -.19, -.05; P < .001). Findings were driven by afternoon and evening caffeine consumption. Reduced sleep efficiency was associated with increased afternoon caffeine intake the following day (ß = -.006; 95% CI = -.012, -.001; P = .01). CONCLUSIONS: Caffeine consumption, especially afternoon and evening use, impacts several aspects of adolescent sleep health. In contrast, most sleep indicators did not affect next-day caffeine use, suggesting multiple drivers of adolescent caffeine consumption. Federal mandates requiring caffeine content labeling and behavioral interventions focused on reducing caffeine intake may support adolescent sleep health. CITATION: Lunsford-Avery JR, Kollins SH, Kansagra S, Wang KW, Engelhard MM. Impact of daily caffeine intake and timing on electroencephalogram-measured sleep in adolescents. J Clin Sleep Med. 2022;18(3):877-884.
Asunto(s)
Cafeína , Sueño , Adolescente , Cafeína/efectos adversos , Niño , Electroencefalografía , Femenino , Humanos , Masculino , Polisomnografía , Sueño REMRESUMEN
OBJECTIVE: Sleep is vital to supporting adolescent behavioral health and functioning; however, sleep disturbances remain under-recognized and undertreated in many health care settings. One barrier is the complexity of sleep, which makes it difficult for providers to determine which aspects-beyond sleep duration-may be most important to assess and treat to support adolescent health. This study examined associations between 2 sleep indices (regularity and timing) and adolescent behavioral health and functioning over and above the impact of shortened/fragmented sleep. METHOD: Eighty-nine adolescents recruited from the community (mean age = 14.04, 45% female participants) completed 7 days/nights of actigraphy and, along with a parent/guardian, reported on behavioral health (internalizing and externalizing symptoms) and psychosocial functioning. Stepwise linear regressions examined associations between sleep timing and regularity and behavioral/functional outcomes after accounting for shortened/fragmented sleep. RESULTS: Delayed sleep timing was associated with greater self-reported internalizing (F[6,82] = 11.57, p = 0.001) and externalizing (F[6,82] = 11.12, p = 0.001) symptoms after accounting for shortened/fragmented sleep. Irregular sleep was associated with greater self-reported and parent-reported externalizing symptoms (self: F[7,81] = 6.55, p = 0.01; parent: F[7,80] = 6.20, p = 0.01) and lower psychosocial functioning (self: F[7,81] = 6.03, p = 0.02; parent: F[7,78] = 3.99, p < 0.05) after accounting for both shortened/fragmented sleep and delayed sleep timing. CONCLUSION: Sleep regularity and timing may be critical for understanding the risk of poor behavioral health and functional deficits among adolescents and as prevention and intervention targets. Future work should focus on developing and evaluating convenient, low-cost, and effective methods for addressing delayed and/or irregular adolescent sleep patterns in real-world health care settings.