Although 50 years represents middle age among uninfected individuals, studies have shown that persons living with HIV (PWH) begin to demonstrate elevated risk for serious falls and fragility fractures in the sixth decade; the proportions of these outcomes attributable to modifiable factors are unknown.

BACKGROUND:Older (older than 50 years) persons living with HIV (PWH) are at elevated risk for falls. We explored how well our algorithm for predicting falls in a general population of middle-aged Veterans (age 45-65 years) worked among older PWH who use antiretroviral therapy (ART) and whether model fit improved with inclusion of specific ART classes. METHODS:This analysis included 304,951 six-month person-intervals over a 15-year period (2001-2015) contributed by 26,373 older PWH from the Veterans Aging Cohort Study who were taking ART. Serious falls (those falls warranting a visit to a health care provider) were identified by external cause of injury codes and a machine-learning algorithm applied to radiology reports. Potential predictors included a fall within the past 12 months, demographics, body mass index, Veterans Aging Cohort Study Index 2.0 score, substance use, and measures of multimorbidity and polypharmacy. We assessed discrimination and calibration from application of the original coefficients (model derived from middle-aged Veterans) to older PWH and then reassessed by refitting the model using multivariable logistic regression with generalized estimating equations. We also explored whether model performance improved with indicators of ART classes. RESULTS:With application of the original coefficients, discrimination was good (C-statistic 0.725; 95% CI: 0.719 to 0.730) but calibration was poor. After refitting the model, both discrimination (C-statistic 0.732; 95% CI: 0.727 to 0.734) and calibration were good. Including ART classes did not improve model performance. CONCLUSIONS:After refitting their coefficients, the same variables predicted risk of serious falls among older PWH nearly and they had among middle-aged Veterans.

BACKGROUND:The extensive research on falls and fragility fractures among persons living with HIV (PWH) has not explored the association between serious falls and subsequent fragility fracture. We explored this association. SETTING:Veterans Aging Cohort Study. METHODS:This analysis included 304,951 6-month person- intervals over a 15-year period (2001-2015) contributed by 26,373 PWH who were 50+ years of age (mean age 55 years) and taking antiretroviral therapy (ART). Serious falls (those falls significant enough to result in a visit to a health care provider) were identified by the external cause of injury codes and a machine learning algorithm applied to radiology reports. Fragility fractures were identified using ICD9 codes and included hip fracture, vertebral fractures, and upper arm fracture and were modeled with multivariable logistic regression with generalized estimating equations. RESULTS:After adjustment, serious falls in the previous year were associated with increased risk of fragility fracture [odds ratio (OR) 2.10; 95% confidence interval (CI): 1.83 to 2.41]. The use of integrase inhibitors was the only ART risk factor (OR 1.17; 95% CI: 1.03 to 1.33). Other risk factors included the diagnosis of alcohol use disorder (OR 1.49; 95% CI: 1.31 to 1.70) and having a prescription for an opioid in the previous 6 months (OR 1.40; 95% CI: 1.27 to 1.53). CONCLUSIONS:Serious falls within the past year are strongly associated with fragility fractures among PWH on ART-largely a middle-aged population-much as they are among older adults in the general population.

BACKGROUND/OBJECTIVES:Due to high rates of multimorbidity, polypharmacy, and hazardous alcohol and opioid use, middle-aged Veterans are at risk for serious falls (those prompting a visit with a healthcare provider), posing significant risk to their forthcoming geriatric health and quality of life. We developed and validated a predictive model of the 6-month risk of serious falls among middle-aged Veterans. DESIGN:Cohort study. SETTING:Veterans Health Administration (VA). PARTICIPANTS:Veterans, aged 45 to 65 years, who presented for care within the VA between 2012 and 2015 (N = 275,940). EXPOSURES:The exposures of primary interest were substance use (including alcohol and prescription opioid use), multimorbidity, and polypharmacy. Hazardous alcohol use was defined as an Alcohol Use Disorders Identification Test - Consumption (AUDIT-C) score of 3 or greater for women and 4 or greater for men. We used International Classification of Diseases, Ninth Revision (ICD-9), codes to identify alcohol and illicit substance use disorders and identified prescription opioid use from pharmacy fill-refill data. We included counts of chronic medications and of physical and mental health comorbidities. MEASUREMENTS:We identified serious falls using external cause of injury codes and a machine-learning algorithm that identified serious falls in radiology reports. We used multivariable logistic regression with general estimating equations to calculate risk. We used an integrated predictiveness curve to identify intervention thresholds. RESULTS:Most of our sample (54%) was aged 60 years or younger. Duration of follow-up was up to 4 years. Veterans who fell were more likely to be female (11% vs 7%) and White (72% vs 68%). They experienced 43,641 serious falls during follow-up. We identified 16 key predictors of serious falls and five interaction terms. Model performance was enhanced by addition of opioid use, as evidenced by overall category-free net reclassification improvement of 0.32 (P < .001). Discrimination (C-statistic = 0.76) and calibration were excellent for both development and validation data sets. CONCLUSION:We developed and internally validated a model to predict 6-month risk of serious falls among middle-aged Veterans with excellent discrimination and calibration.

Background: Electronic health records (EHRs) are a rich source of health information; however social determinants of health, including incarceration, and how they impact health and health care disparities can be hard to extract. Objective: The main objective of this study was to compare sensitivity and specificity of patient self-report with various methods of identifying incarceration exposure using the EHR. Research Design: Validation study using multiple data sources and types. Subjects: Participants of the Veterans Aging Cohort Study (VACS), a national observational cohort based on data from the Veterans Health Administration (VHA) EHR that includes all human immunodeficiency virus–infected patients in care (47,805) and uninfected patients (99,060) matched on region, age, race/ethnicity, and sex. Measures and Data Sources: Self-reported incarceration history compared with: (1) linked VHA EHR data to administrative data from a state Department of Correction (DOC), (2) linked VHA EHR data to administrative data on incarceration from Centers for Medicare and Medicaid Services (CMS), (3) VHA EHR-specific identifier codes indicative of receipt of VHA incarceration reentry services, and (4) natural language processing (NLP) in unstructured text in VHA EHR. Results: Linking the EHR to DOC data: sensitivity 2.5%, specificity 100%; linking the EHR to CMS data: sensitivity 7.9%, specificity 99.3%; VHA EHR-specific identifier for receipt of reentry services: sensitivity 7.3%, specificity 98.9%; and NLP, sensitivity 63.5%, specificity 95.9%. Conclusions: NLP tools hold promise as a feasible and valid method to identify individuals with exposure to incarceration in EHR. Future work should expand this approach using a larger body of documents and refinement of the methods, which may further improve operating characteristics of this method.

BACKGROUND:Medication classes, polypharmacy, and hazardous alcohol and illicit substance abuse may exhibit stronger associations with serious falls among persons living with HIV (PLWH) than with uninfected comparators. We investigated whether these associations differed by HIV status. SETTING:Veterans Aging Cohort Study. METHODS:We used a nested case-control design. Cases (N = 13,530) were those who fell. Falls were identified by external cause of injury codes and a machine-learning algorithm applied to radiology reports. Cases were matched to controls (N = 67,060) by age, race, sex, HIV status, duration of observation, and baseline date. Risk factors included medication classes, count of unique non-antiretroviral therapy (non-ART) medications, and hazardous alcohol and illicit substance use. We used unconditional logistic regression to evaluate associations. RESULTS:Among PLWH, benzodiazepines [odds ratio (OR) 1.24; 95% confidence interval (CI) 1.08 to 1.40] and muscle relaxants (OR 1.29; 95% CI: 1.08 to 1.46) were associated with serious falls but not among uninfected (P > 0.05). In both groups, key risk factors included non-ART medications (per 5 medications) (OR 1.20, 95% CI: 1.17 to 1.23), illicit substance use/abuse (OR 1.44; 95% CI: 1.34 to 1.55), hazardous alcohol use (OR 1.30; 95% CI: 1.23 to 1.37), and an opioid prescription (OR 1.35; 95% CI: 1.29 to 1.41). CONCLUSION:Benzodiazepines and muscle relaxants were associated with serious falls among PLWH. Non-ART medication count, hazardous alcohol and illicit substance use, and opioid prescriptions were associated with serious falls in both groups. Prevention of serious falls should focus on reducing specific classes and absolute number of medications and both alcohol and illicit substance use.

BACKGROUND:Medical practitioners use survival models to explore and understand the relationships between patients' covariates (e.g. clinical and genetic features) and the effectiveness of various treatment options. Standard survival models like the linear Cox proportional hazards model require extensive feature engineering or prior medical knowledge to model treatment interaction at an individual level. While nonlinear survival methods, such as neural networks and survival forests, can inherently model these high-level interaction terms, they have yet to be shown as effective treatment recommender systems.METHODS:We introduce DeepSurv, a Cox proportional hazards deep neural network and state-of-the-art survival method for modeling interactions between a patient's covariates and treatment effectiveness in order to provide personalized treatment recommendations.RESULTS:We perform a number of experiments training DeepSurv on simulated and real survival data. We demonstrate that DeepSurv performs as well as or better than other state-of-the-art survival models and validate that DeepSurv successfully models increasingly complex relationships between a patient's covariates and their risk of failure. We then show how DeepSurv models the relationship between a patient's features and effectiveness of different treatment options to show how DeepSurv can be used to provide individual treatment recommendations. Finally, we train DeepSurv on real clinical studies to demonstrate how it's personalized treatment recommendations would increase the survival time of a set of patients.CONCLUSIONS:The predictive and modeling capabilities of DeepSurv will enable medical researchers to use deep neural networks as a tool in their exploration, understanding, and prediction of the effects of a patient's characteristics on their risk of failure.

PurposeTo estimate medical device utilization needed to detect safety differences among implantable cardioverter defibrillators (ICDs) generator models and compare these estimates to utilization in practice. MethodsWe conducted repeated sample size estimates to calculate the medical device utilization needed, systematically varying device-specific safety event rate ratios and significance levels while maintaining 80% power, testing 3 average adverse event rates (3.9, 6.1, and 12.6 events per 100 person-years) estimated from the American College of Cardiology's 2006 to 2010 National Cardiovascular Data Registry of ICDs. We then compared with actual medical device utilization. ResultsAt significance level 0.05 and 80% power, 34% or fewer ICD models accrued sufficient utilization in practice to detect safety differences for rate ratios <1.15 and an average event rate of 12.6 events per 100person-years. For average event rates of 3.9 and 12.6 events per 100 person-years, 30% and 50% of ICD models, respectively, accrued sufficient utilization for a rate ratio of 1.25, whereas 52% and 67% for a rate ratio of 1.50. Because actual ICD utilization was not uniformly distributed across ICD models, the proportion of individuals receiving any ICD that accrued sufficient utilization in practice was 0% to 21%, 32% to 70%, and 67% to 84% for rate ratios of 1.05, 1.15, and 1.25, respectively, for the range of 3 average adverse event rates. ConclusionsSmall safety differences among ICD generator models are unlikely to be detected through routine surveillance given current ICD utilization in practice, but large safety differences can be detected for most patients at anticipated average adverse event rates.

Background: Machine learning methods may complement traditional analytic methods for medical device surveillance.Methods and results: Using data from the National Cardiovascular Data Registry for implantable cardioverter-defibrillators (ICDs) linked to Medicare administrative claims for longitudinal follow-up, we applied three statistical approaches to safety-signal detection for commonly used dual-chamber ICDs that used two propensity score (PS) models: one specified by subject-matter experts (PS-SME), and the other one by machine learning-based selection (PS-ML). The first approach used PS-SME and cumulative incidence (time-to-event), the second approach used PS-SME and cumulative risk (Data Extraction and Longitudinal Trend Analysis [DELTA]), and the third approach used PS-ML and cumulative risk (embedded feature selection). Safety-signal surveillance was conducted for eleven dual-chamber ICD models implanted at least 2,000 times over 3 years. Between 2006 and 2010, there were 71,948 Medicare fee-for-service beneficiaries who received dual-chamber ICDs. Cumulative device-specific unadjusted 3-year event rates varied for three surveyed safety signals: death from any cause, 12.8%-20.9%; nonfatal ICD-related adverse events, 19.3%-26.3%; and death from any cause or nonfatal ICD-related adverse event, 27.1%-37.6%. Agreement among safety signals detected/not detected between the time-to-event and DELTA approaches was 90.9% (360 of 396, kappa=0.068), between the time-to-event and embedded feature-selection approaches was 91.7% (363 of 396, kappa=-0.028), and between the DELTA and embedded feature selection approaches was 88.1% (349 of 396, kappa=-0.042).Conclusion: Three statistical approaches, including one machine learning method, identified important safety signals, but without exact agreement. Ensemble methods may be needed to detect all safety signals for further evaluation during medical device surveillance.

We propose a method to reduce variance in treatment effect estimates in the setting of high-dimensional data. In particular, we introduce an approach for learning a metric to be used in matching treatment and control groups. The metric reduces variance in treatment effect estimates by weighting covariates related to the outcome and filtering out unrelated covariates.

Objective: The study sought to quantify coordination of epilepsy care, over time, between neurologists and other health care providers using social network analysis (SNA).Methods: The Veterans Health Administration (VA) instituted an Epilepsy Center of Excellence (ECOE) model in 2008 to enhance care coordination between neurologists and other health care providers. Provider networks in the 16 VA ECOE facilities (hub sites) were compared to a subset of 33 VA facilities formally affiliated (consortium sites) and 14 unaffiliated VA facilities. The number of connections between neurologists and each provider (node degree) was measured by shared epilepsy patients and tallied to generate estimates at the facility level separately within and across facilities. Mixed models were used to compare change of facility-level node degree over time across the three facility types, adjusted for number of providers per facility.Results: Over the time period 2000-2013, epilepsy care coordination both within and across facilities significantly increased. These increases were seen in all three types of facilities namely hub, consortium, and unaffiliated site, relatively equally. The increase in connectivity was more dramatic with providers across facilities compared to providers within the same facilities.Conclusion: Establishment of the ECOE hub and spoke model contributed to an increase in epilepsy care coordination both within and across facilities from 2000 to 2013, but there was substantial variation across different facilities. SNA is a tool that may help measure coordination of specialty care. Published by Elsevier Inc.

Objective. To determine if 1) patients have distinct affective reaction patterns to medication information, and 2) whether there is an association between affective reaction patterns and willingness to take medication. Methods. We measured affect in real time as subjects listened to a description of benefits and side effects for a hypothetical new medication. Subjects moved a dial on a handheld response system to indicate how they were feeling from Very Good to Very Bad. Patterns of reactions were identified using a cluster-analytic statistical approach for multiple time series. Subjects subsequently rated their willingness to take the medication on a 7-point Likert scale. Associations between subjects' willingness ratings and affect patterns were analyzed. Additional analyses were performed to explore the role of race/ethnicity regarding these associations. Results. Clusters of affective reactions emerged that could be classified into 4 patterns: Moderate positive reactions to benefits and negative reactions to side effects (n = 186), Pronounced positive reactions to benefits and negative reactions to side effects (n = 110), feeling consistently Good (n = 58), and feeling consistently close to Neutral (n = 33). Mean (standard error) willingness to take the medication was greater among subjects feeling consistently Good 4.72 (0.20) compared with those in the Moderate 3.76 (0.11), Pronounced 3.68 (0.14), and Neutral 3.62 (0.26) groups. Black subjects with a Pronounced pattern were less willing to take the medication compared with both Hispanic (P = 0.0270) and White subjects (P = 0.0001) with a Pronounced pattern. Conclusion. Patients' affective reactions to information were clustered into specific patterns. Reactions varied by race/ethnicity and were associated with treatment willingness. Ultimately, a better understanding of how patients react to information may help providers develop improved methods of communication.

Previous research has shown that neural networks can model survival data in situations in which some patients' death times are unknown, e.g. right-censored. However, neural networks have rarely been shown to outperform their linear counterparts such as the Cox proportional hazards model. In this paper, we run simulated experiments and use real survival data to build upon the risk-regression architecture proposed by Faraggi and Simon. We demonstrate that our model, DeepSurv, not only works as well as other survival models but actually outperforms in predictive ability on survival data with linear and nonlinear risk functions. We then show that the neural network can also serve as a recommender system by including a categorical variable representing a treatment group. This can be used to provide personalized treatment recommendations based on an individual's calculated risk. We provide an open source Python module that implements these methods in order to advance research on deep learning and survival analysis.

OBJECTIVE:To identify patients in a human immunodeficiency virus (HIV) study cohort who have fallen by applying supervised machine learning methods to radiology reports of the cohort.METHODS:We used the Veterans Aging Cohort Study Virtual Cohort (VACS-VC), an electronic health record-based cohort of 146 530 veterans for whom radiology reports were available (N=2 977 739). We created a reference standard of radiology reports, represented each report by a feature set of words and Unified Medical Language System concepts, and then developed several support vector machine (SVM) classifiers for falls. We compared mutual information (MI) ranking and embedded feature selection approaches. The SVM classifier with MI feature selection was chosen to classify all radiology reports in VACS-VC.RESULTS:Our SVM classifier with MI feature selection achieved an area under the curve score of 97.04 on the test set. When applied to all the radiology reports in VACS-VC, 80 416 of these reports were classified as positive for a fall. Of these, 11 484 were associated with a fall-related external cause of injury code (E-code) and 68 932 were not, corresponding to 29 280 patients with potential fall-related injuries who could not have been found using E-codes.DISCUSSION:Feature selection was crucial to improving the classifier's performance. Feature selection with MI allowed us to select the number of discriminative features to use for classification, in contrast to the embedded feature selection method, in which the number of features is chosen automatically.CONCLUSION:Machine learning is an effective method of identifying patients who have suffered a fall. The development of this classifier supplements the clinical researcher's toolkit and reduces dependence on under-coded structured electronic health record data.

Any closed, connected Riemannian manifold M can be smoothly embedded by its Laplacian eigenfunction maps into R-m for some m. We call the smallest such m the maximal embedding dimension of M. We show that the maximal embedding dimension of M is bounded from above by a constant depending only on the dimension of M, a lower bound for injectivity radius, a lower bound for Ricci curvature, and a volume bound. We interpret this result for the case of surfaces isometrically immersed in 1183, showing that the maximal embedding dimension only depends on bounds for the Gaussian curvature, mean curvature, and surface area. Furthermore, we consider the relevance of these results for shape registration. (C) 2014 Elsevier Inc. All rights reserved.

We prove consistency results for two types of density estimators on a closed, connected Riemannian manifold under suitable regularity conditions. The convergence rates are consistent with those in Euclidean space as well as those obtained for a previously proposed class of kernel density estimators on closed Riemannian manifolds. The first estimator is the uniform mixture of heat kernels centered at each observation, a natural extension of the usual Gaussian estimator to Riemannian manifolds. The second is an approximate heat kernel (AHK) estimator that is motivated by more practical considerations, where observations occur on a manifold isometrically embedded in Euclidean space whose structure or heat kernel may not be completely known. We also provide some numerical evidence that the predicted convergence rate is attained for the AHK estimator.

We introduce a class of spectral shape signatures constructed from symmetric functions on the eigenfunctions of the Laplacian exponentially weighted by their eigenvalues. Such a construction is motivated by problems that arise in the use of the eigenfunctions for shape comparison, such as indeterminacies in the choice of signs and the particular ordering in which the eigenfunctions are presented. The spectral invariants are applied to the analysis of Alzheimer's disease (AD) data collected by the Alzheimer's Disease Neuroimaging Initiative, in particular, to the problem of determining whether the signatures can aid in early detection of AD through morphology and imaging.