RESUMO
BACKGROUND: Validating new algorithms, such as methods to disentangle intrinsic treatment risk from risk associated with experiential learning of novel treatments, often requires knowing the ground truth for data characteristics under investigation. Since the ground truth is inaccessible in real world data, simulation studies using synthetic datasets that mimic complex clinical environments are essential. We describe and evaluate a generalizable framework for injecting hierarchical learning effects within a robust data generation process that incorporates the magnitude of intrinsic risk and accounts for known critical elements in clinical data relationships. METHODS: We present a multi-step data generating process with customizable options and flexible modules to support a variety of simulation requirements. Synthetic patients with nonlinear and correlated features are assigned to provider and institution case series. The probability of treatment and outcome assignment are associated with patient features based on user definitions. Risk due to experiential learning by providers and/or institutions when novel treatments are introduced is injected at various speeds and magnitudes. To further reflect real-world complexity, users can request missing values and omitted variables. We illustrate an implementation of our method in a case study using MIMIC-III data for reference patient feature distributions. RESULTS: Realized data characteristics in the simulated data reflected specified values. Apparent deviations in treatment effects and feature distributions, though not statistically significant, were most common in small datasets (n < 3000) and attributable to random noise and variability in estimating realized values in small samples. When learning effects were specified, synthetic datasets exhibited changes in the probability of an adverse outcomes as cases accrued for the treatment group impacted by learning and stable probabilities as cases accrued for the treatment group not affected by learning. CONCLUSIONS: Our framework extends clinical data simulation techniques beyond generation of patient features to incorporate hierarchical learning effects. This enables the complex simulation studies required to develop and rigorously test algorithms developed to disentangle treatment safety signals from the effects of experiential learning. By supporting such efforts, this work can help identify training opportunities, avoid unwarranted restriction of access to medical advances, and hasten treatment improvements.
Assuntos
Aprendizado Profundo , Humanos , Simulação por Computador , AlgoritmosRESUMO
Determining biomarkers of responses to environmental exposures and evaluating whether they predict respiratory outcomes may help optimize environmental and medical approaches to childhood asthma. Relative mitochondrial (mt) DNA abundance and other potential mitochondrial indicators of oxidative stress may provide a sensitive metric of the child's shifting molecular responses to its changing environment. We leveraged two urban childhood cohorts (Environmental Control as Add-on Therapy in Childhood Asthma (ECATCh); Columbia Center for Children's Environmental Health (CCCEH)) to ascertain whether biomarkers in buccal mtDNA associate with airway inflammation and altered lung function over 6 months of time and capture biologic responses to multiple external stressors such as indoor allergens and fine particulate matter (PM2.5). Relative mtDNA content was amplified by qPCR and methylation of transfer RNA phenylalanine/rRNA 12S (TF/RNR1), cytochrome c oxidase (CO1), and carboxypeptidase O (CPO) was measured by pyrosequencing. Data on residential exposures and respiratory outcomes were harmonized between the two cohorts. Repeated measures and multiple regression models were utilized to assess relationships between mitochondrial biomarkers, respiratory outcomes, and residential exposures (PM2.5, allergens), adjusted for potential confounders and time-varying asthma. We found across the 6 month visits, a 0.64 fold higher level of TF/RNR1 methylation was detected among those with asthma in comparison to those without asthma ((parameter estimate (PE) 0.64, standard error 0.28, p = 0.03). In prospective analyses, CPO methylation was associated with subsequent reduced forced vital capacity (FVC; PE -0.03, standard error 0.01, p = 0.02). Bedroom dust mouse allergen, but not indoor PM2.5, was associated with higher methylation of TF/RNR1 (PE 0.015, standard error 0.006, p = 0.01). Select mtDNA measures in buccal cells may indicate children's responses to toxic environmental exposures and associate selectively with asthma and lung function.
Assuntos
Asma , Mucosa Bucal , Criança , Humanos , Animais , Camundongos , Estudos Prospectivos , Asma/epidemiologia , DNA Mitocondrial , Biomarcadores , Material Particulado/toxicidadeRESUMO
In the use of medical device procedures, learning effects have been shown to be a critical component of medical device safety surveillance. To support their estimation of these effects, we evaluated multiple methods for modeling these rates within a complex simulated dataset representing patients treated by physicians clustered within institutions. We employed unique modeling for the learning curves to incorporate the learning hierarchy between institution and physicians and then modeled them within established methods that work with hierarchical data such as generalized estimating equations (GEE) and generalized linear mixed effect models. We found that both methods performed well, but that the GEE may have some advantages over the generalized linear mixed effect models for ease of modeling and a substantially lower rate of model convergence failures. We then focused more on using GEE and performed a separate simulation to vary the shape of the learning curve as well as employed various smoothing methods to the plots. We concluded that while both hierarchical methods can be used with our mathematical modeling of the learning curve, the GEE tended to perform better across multiple simulated scenarios in order to accurately model the learning effect as a function of physician and hospital hierarchical data in the use of a novel medical device. We found that the choice of shape used to produce the 'learning-free' dataset would be dataset specific, while the choice of smoothing method was negligibly different from one another. This was an important application to understand how best to fit this unique learning curve function for hierarchical physician and hospital data. Copyright © 2017 John Wiley & Sons, Ltd.
Assuntos
Equipamentos e Provisões , Curva de Aprendizado , Modelos Estatísticos , Cateterismo Cardíaco/efeitos adversos , Simulação por Computador , Interpretação Estatística de Dados , Educação Médica , Feminino , Humanos , Modelos Lineares , Masculino , MédicosRESUMO
OBJECTIVES: We sought to compare 2 contrast-induced nephropathy (CIN) risk prediction models in a validation cohort using a consensus definition. BACKGROUND: Contrast-induced nephropathy (CIN) is independently associated with mortality following percutaneous coronary intervention (PCI). Multiple prediction models for the development of CIN have been published using heterogeneous outcome definitions. METHODS: We analyzed 5,540 patients who underwent PCI from January 2005 to June 2012 at a single academic medical center. The primary outcome was development of CIN, defined as an increase in serum creatinine of ≥0.5 mg/dl or a relative increase of ≥25% from baseline. Receiver operator characteristic (ROC) curves were used to evaluate the discriminatory power of Mehran and WBH prediction models. RESULTS: The mean age of our cohort was 68 ± 12 years. The mean baseline creatinine was 1.2 ± 0.53 mg/dl (eGFR 73 ± 27 ml/min). The mean contrast volume used was 212 ± 92 ml. CIN occurred in 436 patients (7.9%). The Mehran risk score demonstrated better discrimination than the William Beaumont Hospital (WBH) risk score to predict the occurrence of CIN (c statistic: 0.82 vs. 0.73, respectively). Mortality at 30 days was approximately 8 times higher among patients with CIN as compared to those without (14.7% vs. 1.8% P < 0.01). CONCLUSIONS: In an independent validation cohort, the Mehran risk model demonstrates greater discriminatory power than the WBH model in predicting the incidence of CIN. Mortality was significantly higher in patients who developed CIN after PCI.
Assuntos
Meios de Contraste/efeitos adversos , Doença das Coronárias , Nefropatias , Intervenção Coronária Percutânea , Medição de Risco/métodos , Idoso , Idoso de 80 Anos ou mais , Doença das Coronárias/diagnóstico , Doença das Coronárias/epidemiologia , Doença das Coronárias/terapia , Creatinina/análise , Feminino , Humanos , Nefropatias/induzido quimicamente , Nefropatias/diagnóstico , Nefropatias/mortalidade , Masculino , Pessoa de Meia-Idade , Modelos Teóricos , Avaliação de Processos e Resultados em Cuidados de Saúde , Intervenção Coronária Percutânea/efeitos adversos , Intervenção Coronária Percutânea/métodos , Valor Preditivo dos Testes , Prognóstico , Curva ROC , Fatores de Risco , Estados Unidos/epidemiologiaRESUMO
In survival analysis, frailty models are potential choices for modeling unexplained heterogeneity in a population. This tutorial presents an overview and general framework of frailty modeling and estimation for multiplicative hazards models in the context of biomedical and genetic studies. Other topics in frailty models, such as diagnostic methods for model adequacy and inference in frailty models, are also discussed. Examples of analyses using multivariate frailty models in a non-parametric hazards setting on biomedical datasets are provided, and the implications of choosing to use frailty and relevance to genetic applications are discussed.
Assuntos
Modelos Estatísticos , Modelos de Riscos Proporcionais , Humanos , Modelos Genéticos , Dinâmica Populacional , Análise de SobrevidaRESUMO
OBJECTIVE: To evaluate in vivo cross-sectional conformational changes of ascending aortic wall excursion in patients undergoing resection for aortic aneurysm with those undergoing elective coronary artery bypass grafting (CABG) using epi-aortic echocardiography. DESIGN: A prospective observational investigation. SETTING: A single tertiary care university hospital. PARTICIPANTS: Thirty-four patients undergoing elective ascending aorta resection and 23 elective CABG patients. INTERVENTION: In an open-chest model and with use of an epi-aortic echocardiographic probe, measurements of aortic wall excursion were made on the ascending aortic aneurysms. Control measurements were made on the transitional neck portions of the aneurysmal aortas (internal control) and CABG aortas (external control). MEASUREMENTS AND MAIN RESULTS: The aortic aneurysm measurements exhibited no difference (2.8%, p < 0.62) between the excursion of the anterior and posterior walls. In contrast, under similar hemodynamic conditions, the anterior wall of the aneurysm neck moved 48.2% (p < 0.0004) more than the posterior wall. Similarly, in the CABG control group, the anterior wall moved 24% (p < 0.027) more than the posterior wall. CONCLUSION: This in vivo study documented a lack of asymmetric aortic wall motion in ascending aortic aneurysms. In contrast, both the internal and external control groups (aneurysm neck and CABG) demonstrated asymmetric wall motion. The lack of asymmetric wall motion may be an important aspect of aneurysm pathophysiology and key to the development of management strategies for timing of surgical intervention.
Assuntos
Aorta/diagnóstico por imagem , Aneurisma Aórtico/diagnóstico por imagem , Ecocardiografia/métodos , Idoso , Aorta/cirurgia , Aneurisma Aórtico/cirurgia , Ponte de Artéria Coronária/métodos , Ecocardiografia/instrumentação , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Estudos ProspectivosRESUMO
OBJECTIVE: To describe development of the Positive Affect, Promoting Positive Engagement, and Adherence for Life (APPEAL) program. METHOD: APPEAL is intended to increase HIV medication adherence through promotion of positive affect, and was developed through an iterative process involving 6 focus groups (N = 34) that elicited feedback on intervention content, followed by an individually administered prepilot of the entire intervention (N = 7). RESULTS: Participants provided feedback on important potential moderator variables, including depression, on mode of intervention administration, and on anticipated barriers and benefits to participation. Insights gained were used to finalize study procedures in preparation for a feasibility trial. For the feasibility trial, a total of 80 participants who, in the past 6 months have had at least one plasma HIV RNA >200 copies/mL, will be randomized to receive APPEAL or standard of care (N = 40 per group). Intervention group participants will receive 3 monthly, individually administered sessions, and all participants will have their medication adherence monitored and complete structured interviews at baseline and at 3 and 6 months. CONCLUSION: The APPEAL program is innovative in that it focuses on promoting self-regulation of positive emotions, an understudied approach to promoting chronic disease self-management behaviors such as HIV medication adherence. Findings from the feasibility trial will gauge suitability of the APPEAL intervention and evaluation methods for subsequent testing in a confirmatory trial and will examine changes in positive affect, the primary mechanism of change targeted in the intervention. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
Assuntos
Adesão à Medicação/estatística & dados numéricos , Adulto , Estudos de Viabilidade , Feminino , Grupos Focais , Humanos , Masculino , Adulto JovemRESUMO
OBJECTIVE: The purpose of our study was to investigate the efficacy of cross-table lateral knee radiography in the diagnosis of knee effusions compared with an MRI reference standard, to evaluate reader experience in effusion assessment, and to establish a new threshold for suprapatellar pouch measurement for the diagnosis of effusion. MATERIALS AND METHODS: First- and third-year radiology residents and an attending musculoskeletal radiologist retrospectively assessed 108 cross-table lateral knee radiographs for qualitative grading of joint fluid and quantitative measurement of the suprapatellar pouch. Qualitative and quantitative evaluation of ipsilateral knee MRI examinations performed within 1 week of radiography was performed by two attending musculoskeletal radiologists as a reference standard. RESULTS: Qualitative visual grading of cross-table lateral radiographs had a sensitivity of 90-92%, specificity of 39-54%, and accuracy of 69-76% for joint effusion. Extrapolating from previous work showing 4 mL of fluid distends the suprapatellar pouch to 4 mm on midline sagittal MRI, the corresponding measurement on cross-table lateral radiographs was predicted to be 7 mm. Using this new criterion of effusion, sensitivity, specificity, and accuracy compared with an MR midline sagittal reference standard were 76%, 83%, and 81%, respectively. Historical data for overhead lateral radiographs had a sensitivity of 78%, specificity of 80%, and accuracy of 79%. CONCLUSION: Qualitative visual assessment of cross-table lateral knee radiographs is highly sensitive for the detection of joint effusion. By performing quantitative evaluation with a new 7-mm criterion for suprapatellar pouch measurement, sensitivity, specificity, and accuracy are equivalent to that of overhead lateral radiography.
Assuntos
Exsudatos e Transudatos/citologia , Aumento da Imagem/métodos , Artropatias/diagnóstico , Articulação do Joelho/patologia , Imageamento por Ressonância Magnética/métodos , Decúbito Dorsal , Feminino , Humanos , Masculino , Variações Dependentes do Observador , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
OBJECTIVES: The goal of this study was to assess the frequency and predictors of vascular closure device (VCD) deployment failure, and its association with vascular complications of 3 commonly used VCDs. BACKGROUND: VCDs are commonly used following percutaneous coronary intervention on the basis of studies demonstrating reduced time to ambulation, increased patient comfort, and possible reduction in vascular complications as compared with manual compression. However, limited data are available on the frequency and predictors of VCD failure, and the association of deployment failure with vascular complications. METHODS: From a de-identified dataset provided by Massachusetts Department of Health, 23,813 consecutive interventional coronary procedures that used either a collagen plug-based (n = 18,533), a nitinol clip-based (n = 2,284), or a suture-based (n = 2,996) VCD between June 2005 and December 2007 were identified. The authors defined VCD failure as unsuccessful deployment or failure to achieve immediate access site hemostasis. RESULTS: Among 23,813 procedures, the VCD failed in 781 (3.3%) procedures (2.1% of collagen plug-based, 6.1% of suture-based, 9.5% of nitinol clip-based VCDs). Patients with VCD failure had an excess risk of "any" (7.7% vs. 2.8%; p < 0.001), major (3.3% vs. 0.8%; p < 0.001), or minor (5.8% vs. 2.1%; p < 0.001) vascular complications compared with successful VCD deployment. In a propensity score-adjusted analysis, when compared with collagen plug-based VCD (reference odds ratio [OR] = 1.0), nitinol clip-based VCD had 2-fold increased risk (OR: 2.0, 95% confidence interval [CI]: 1.8 to 2.3, p < 0.001) and suture-based VCD had 1.25-fold increased risk (OR: 1.25, 95% CI: 1.2 to 1.3, p < 0.001) for VCD failure. VCD failure was a significant predictor of subsequent vascular complications for both collagen plug-based VCD and nitinol clip-based VCD, but not for suture-based VCD. CONCLUSIONS: VCD failure rates vary depending upon the type of VCD used and are associated with significantly higher vascular complications as compared with deployment successes.
Assuntos
Falha de Equipamento/estatística & dados numéricos , Intervenção Coronária Percutânea , Procedimentos Cirúrgicos Vasculares/instrumentação , Técnicas de Fechamento de Ferimentos/instrumentação , Idoso , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Estudos RetrospectivosRESUMO
We examined the behavior of alternative smoothing methods for modeling environmental epidemiology data. Model fit can only be examined when the true exposure-response curve is known and so we used simulation studies to examine the performance of penalized splines (P-splines), restricted cubic splines (RCS), natural splines (NS), and fractional polynomials (FP). Survival data were generated under six plausible exposure-response scenarios with a right skewed exposure distribution, typical of environmental exposures. Cox models with each spline or FP were fit to simulated datasets. The best models, e.g. degrees of freedom, were selected using default criteria for each method. The root mean-square error (rMSE) and area difference were computed to assess model fit and bias (difference between the observed and true curves). The test for linearity was a measure of sensitivity and the test of the null was an assessment of statistical power. No one method performed best according to all four measures of performance, however, all methods performed reasonably well. The model fit was best for P-splines for almost all true positive scenarios, although fractional polynomials and RCS were least biased, on average.
Assuntos
Simulação por Computador , Interpretação Estatística de Dados , Exposição Ambiental , Estudos Epidemiológicos , Modelos de Riscos Proporcionais , HumanosRESUMO
To allow for non-linear exposure-response relationships, we applied flexible non-parametric smoothing techniques to models of time to lung cancer mortality in two occupational cohorts with skewed exposure distributions. We focused on three different smoothing techniques in Cox models: penalized splines, restricted cubic splines, and fractional polynomials. We compared standard software implementations of these three methods based on their visual representation and criterion for model selection. We propose a measure of the difference between a pair of curves based on the area between them, standardized by the average of the areas under the pair of curves. To capture the variation in the difference over the range of exposure, the area between curves was also calculated at percentiles of exposure and expressed as a percentage of the total difference. The dose-response curves from the three methods were similar in both studies over the denser portion of the exposure range, with the difference between curves up to the 50th percentile less than 1 per cent of the total difference. A comparison of inverse variance weighted areas applied to the data set with a more skewed exposure distribution allowed us to estimate area differences with more precision by reducing the proportion attributed to the upper 1 per cent tail region. Overall, the penalized spline and the restricted cubic spline were closer to each other than either was to the fractional polynomial.
Assuntos
Interpretação Estatística de Dados , Neoplasias Pulmonares/mortalidade , Exposição Ocupacional/estatística & dados numéricos , Modelos de Riscos Proporcionais , Adulto , Estudos de Coortes , Seguimentos , Humanos , Masculino , Mineração , New Mexico/epidemiologia , Software , Fatores de TempoRESUMO
OBJECTIVES: Genotyping errors can induce biases in frequency estimates for haplotypes of single nucleotide polymorphisms (SNPs). Here, we considered the impact of SNP allele misclassification on haplotype odds ratio estimates from case-control studies of unrelated individuals. METHODS: We calculated bias analytically, using the haplotype counts expected in cases and controls under genotype misclassification. We evaluated the bias due to allele misclassification across a range of haplotype distributions using empirical haplotype frequencies within blocks of limited haplotype diversity. We also considered simple two- and three-locus haplotype distributions to understand the impact of haplotype frequency and number of SNPs on misclassification bias. RESULTS: We found that for common haplotypes (>5% frequency), realistic genotyping error rates (0.1-1% chance of miscalling an allele), and moderate relative risks (2-4), the bias was always towards the null and increases in magnitude with increasing error rate, increasing odds ratio. For common haplotypes, bias generally increased with increasing haplotype frequency, while for rare haplotypes, bias generally increased with decreasing frequency. When the chance of miscalling an allele is 0.5%, the median bias in haplotype-specific odds ratios for common haplotypes was generally small (<4% on the log odds ratio scale), but the bias for some individual haplotypes was larger (10-20%). Bias towards the null leads to a loss in power; the relative efficiency using a test statistic based upon misclassified haplotype data compared to a test based on the unobserved true haplotypes ranged from roughly 60% to 80%, and worsened with increasing haplotype frequency. CONCLUSIONS: The cumulative effect of small allele-calling errors across multiple loci can induce noticeable bias and reduce power in realistic scenarios. This has implications for the design of candidate gene association studies that utilize multi-marker haplotypes.