Adam Ciarleglio

Wavelet-based scalar-on-function finite mixture regression models

Classical finite mixture regression is useful for modeling the relationship between scalar predictors and scalar responses arising from subpopulations defined by the di ering associations between those predictors and responses. The classical finite mixture regression model is extended to incorporate functional predictors by taking a wavelet-based approach in which both the functional predictors and the component-specific coefficient functions are represented in terms of an appropriate wavelet basis. By using the wavelet representation of the model, the coefficients corresponding to the functional covariates become the predictors. In this setting, there are typically many more predictors than observations. Hence a lasso-type penalization is employed to simultaneously perform feature selection and estimation. Specification of the model is discussed and a fitting algorithm is provided. The wavelet-based approach is evaluated on synthetic data as well as applied to a real data set from a study of the relationship between cognitive ability and di usion tensor imaging measures in subjects with multiple sclerosis.

Statistical analysis plan for stage 1 EMBARC (Establishing Moderators and Biosignatures of Antidepressant Response for Clinical Care) study

Antidepressant medications are commonly used to treat depression, but only about 30% of patients reach remission with any single first-step antidepressant. If the first-step treatment fails, response and remission rates at subsequent steps are even more limited. The literature on biomarkers for treatment response is largely based on secondary analyses of studies designed to answer primary questions of efficacy, rather than on a planned systematic evaluation of biomarkers for treatment decision. The lack of evidence-based knowledge to guide treatment decisions for patients with depression has lead to the recognition that specially designed studies with the primary objective being to discover biosignatures for optimizing treatment decisions are necessary. Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) is one such discovery study. Stage 1 of EMBARC is a randomized placebo controlled clinical trial of 8 week duration. A wide array of patient characteristics is collected at baseline, including assessments of brain structure, function and connectivity along with electrophysiological, biological, behavioral and clinical features. This paper reports on the data analytic strategy for discovering biosignatures for treatment response based on Stage 1 of EMBARC.

The Norwood Operation: Relative Effects of Surgeon and Institutional Volumes on Outcomes and Resource Utilization

BACKGROUND Hypoplastic left heart syndrome is the most expensive birth defect managed in the United States, with a 5-year survival rate below 70%. Increasing evidence suggests that hospital volumes are inversely associated with mortality for infants with single ventricles undergoing stage 1 surgical palliation. Our aim was to examine the relative effects of surgeon and institutional volumes on outcomes and resource utilisation for these children. METHODS A retrospective study was conducted using the Pediatric Health Information System database to examine the effects of the number of procedures performed per surgeon and per centre on mortality, costs, and post-operative length of stay for infants undergoing Risk Adjustment for Congenital Heart Surgery risk category six operations at tertiary-care paediatric hospitals, from 1 January, 2004 to 31 December, 2013. Multivariable modelling was used, adjusting for patient and institutional characteristics. Gaussian kernel densities were constructed to show the relative distributions of the effects of individual institutions and surgeons, before and after adjusting for the number of cases performed. RESULTS A total of 2880 infants from 35 institutions met the inclusion criteria. Mortality was 15.0%. Median post-operative length of stay was 24 days (IQR 14-41). Median standardized inpatient hospital costs were

Flexible functional regression methods for estimating individualized treatment rules

156,000 (IQR

Hippocampal volume and integrity as predictors of cognitive decline in intact elderly

108,000-

Asthma and subjective sleep disordered breathing in a large cohort of urban adolescents

248,000) in 2013 dollars. In the multivariable analyses, higher institutional volume was inversely associated with mortality (p=0.001), post-operative length of stay (p=0.004), and costs (p=0.001). Surgeon volume was associated with none of the measured outcomes. Neither institutional nor surgeon volumes explained much of the wide variation in outcomes and resource utilization observed between institutions and between surgeons. CONCLUSIONS Increased institutional - but not surgeon - volumes are associated with reduced mortality, post-operative length of stay, and costs for infants undergoing stage 1 palliation.

Neonatal Hypoplastic Left Heart Syndrome: Effects of Bloodstream Infections on Outcomes and Costs

A major focus of personalized medicine is on the development of individualized treatment rules. Good decision rules have the potential to significantly advance patient care and reduce the burden of a host of diseases. Statistical methods for developing such rules are progressing rapidly, but few methods have considered the use of pre-treatment functional data to guide in decision-making. Furthermore, those methods that do allow for the incorporation of functional pre-treatment covariates typically make strong assumptions about the relationships between the functional covariates and the response of interest. We propose two approaches for using functional data to select an optimal treatment that address some of the shortcomings of previously developed methods. Specifically, we combine the flexibility of functional additive regression models with Q-learning or A-learning in order to obtain treatment decision rules. Properties of the corresponding estimators are discussed. Our approaches are evaluated in several realistic settings using synthetic data and are applied to real data arising from a clinical trial comparing two treatments for major depressive disorder in which baseline imaging data are available for subjects who are subsequently treated.

A comparison of hippocampal volume and integrity: Which is the better predictor of cognitive decline?

The risk of Alzheimer’s disease can be predicted by volumetric analyses of MRI data in the medial temporal lobe. The present study compared a volumetric measurement of the hippocampus with a novel measure of hippocampal integrity (HI) derived from the ratio of parenchyma volume over total volume. Participants were cognitively intact and aged 60 years or older at baseline, and were tested twice, roughly 3 years apart. Participants had been recruited for a study on late-life major depression (LLMD) and were evenly split between depressed patients and controls. Linear regression models were applied to the data with a cognitive composite score as the outcome, and HI and volume, together or separately, as predictors. Subsequent cognitive performance was predicted well by models that included an interaction between HI and LLMD status, such that lower HI scores predicted more cognitive decline in depressed patients. More research is needed, but tentative results from this study appear to suggest that the newly introduced measure HI is an effective tool for the purpose of predicting future changes in general cognitive ability, and especially so in individuals with LLMD.

Earlier stage 1 palliation is associated with better clinical outcomes and lower costs for neonates with hypoplastic left heart syndrome

ABSTRACT Objective: Sleep disordered breathing (SDB) has not been well studied in urban adolescents with asthma in community settings. Nor has the association of SDB symptoms and asthma severity been studied. We characterized self-reported symptoms suggesting SDB and investigated the association of SDB symptoms, probable asthma, and asthma severity. Methods: 9,565 adolescents from 21 inner-city high schools were screened for an asthma intervention study. Students reported on symptoms suggesting SDB using questions from the 2007 NHANES, if they were ever diagnosed with asthma, and on asthma symptoms. Using generalized linear mixed models with logit link with school as a random intercept and adjusting for age, gender, and race/ethnicity, we examined associations of SDB symptoms, and demographic characteristics, probable asthma, and asthma severity. Results: 12% reported SDB symptoms. Older and bi-racial participants (compared to Caucasian) had higher odds of symptoms suggesting SDB (p <.001). Compared to those without probable asthma, adolescents with probable asthma had 2.63 greater odds of reporting SDB symptoms (p <.001). Among those with probable asthma, the odds of reporting SDB symptoms increased with asthma severity. When exploring daytime severity and severity due to night wakening separately, results were similar. All results remained significant when controlling for age, gender, and ethnicity. Conclusions: In a large urban community cohort of predominately ethnic minority adolescents, self-reported SDB symptoms were associated with probable asthma and increased asthma severity. This study highlights the importance of SDB as a modifiable co-morbidity of asthma.

Treatment Decisions Based on Scalar and Functional Baseline Covariates

BACKGROUND Hypoplastic left heart syndrome (HLHS) is not only a devastating disease, but also the most expensive birth defect managed in the US. Nosocomial bloodstream infections (NBIs) are common in neonates with HLHS. We examined the effects of NBIs on in-hospital mortality, length of stay, and costs for late preterm and term infants with HLHS undergoing stage 1 palliation, at both individual patient and hospital levels. METHODS We conducted a retrospective study of infants 35 weeks or greater gestation with HLHS, admitted to our institution January 1, 2003 to January 1, 2013. Children with other cardiac abnormalities, major comorbid conditions, or perinatal infections were excluded. Univariable and multivariable analyses were performed. To estimate the effects of reduced NBI incidence on resource utilization, predictive models were used. RESULTS One hundred forty-three children met inclusion criteria. In-hospital mortality was 9.1% (n = 13). Postoperative infection was observed in 12.6% (n = 18). Median length of stay was 23 days for survivors (IQR, 17 to 40; range, 9 to 132). Median costs were