Robert D. Gibbons

The University of Tennessee Medical Center at Knoxville, Knoxville, Tennessee, United States

Are you Robert D. Gibbons?

Claim your profile

Publications (243)1179.76 Total impact

  • Robert D. Gibbons · David J. Weiss · Ellen Frank · David Kupfer
    [Show abstract] [Hide abstract]
    ABSTRACT: In this review we explore recent developments in computerized adaptive diagnostic screening and computerized adaptive testing for the presence and severity of mental health disorders such as depression, anxiety, and mania. The statistical methodology is unique in that it is based on multidimensional item response theory (severity) and random forests (diagnosis) instead of traditional mental health measurement based on classical test theory (a simple total score) or unidimensional item response theory. We show that the information contained in large item banks consisting of hundreds of symptom items can be efficiently calibrated using multidimensional item response theory, and the information contained in these large item banks can be precisely extracted using adaptive administration of a small set of items for each individual. In terms of diagnosis, computerized adaptive diagnostic screening can accurately track an hour-long face-to-face clinician diagnostic interview for major depressive disorder (as an example) in less than a minute using an average of four questions with unprecedented high sensitivity and specificity. Directions for future research and applications are discussed. Expected final online publication date for the Annual Review of Clinical Psychology Volume 12 is March 28, 2016. Please see for revised estimates.
    No preview · Article · Apr 2016 · Annual Review of Clinical Psychology
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Hospital-acquired pressure ulcers (HAPUs) are costly to treat and can result in Medicare reimbursement penalties. Statistical models can identify patients at greatest HAPU risk and improve prevention. Objectives To use electronic health record (EHR) data to predict HAPUs among hospitalized patients. Methods EHR data were obtained from an academic medical center that included hospitalized patients with at least 1 skin examination between 2011–2014. These data contained encounter-level demographic variables, diagnoses, prescription drugs and provider orders. HAPUs were defined by stages III, IV or unstageable pressure ulcers not present-on-admission as a secondary diagnosis, and excluded diagnosis of paraplegia/quadriplegia. Random forests and k-means clustering were applied to reduce the dimensionality of the large dataset. A 2-level mixed-effects logistic regression of patient-encounters evaluated associations between covariates and HAPU incidence (Equation 1). Results The approach produced a sample population of 23,054 patients with 1,549 HAPUs. The mixed-effects model predicted HAPUs with exceptional (99%) accuracy for a rare event (table 1). The greatest odds ratio (OR) of HAPU incidence was among patients diagnosed with spinal cord injury (ICD-9 907.2: OR=247.4; P<0.001). Other high ORs included osteomyelitis (ICD-9 730: OR=27.7, P<0.001), bed confinement (ICD-9 V49.84: OR=31.7, P<0.001), and prescribed topical/subcutaneous enzymes (OR=5.7, P<0.001). Conclusions Early detection of HAPUs is feasible and the results of these statistical predictions can allow providers to better target prevention to specific patients. This model also implicates spinal cord injury as a potential risk-factor for unavoidable HAPUs. Providers may be missing opportunities to co-diagnose spinal cord injury with paraplegia/quadriplegia which could improve hospital performance measures. Equation 1. Mixed-effects Logistic Regression Model Level-1: Encounter-level Fixed Effects Level-2: Patient/Cluster-level Random Effect Where··· i: Patient j: Encounter
    Full-text · Article · Nov 2015 · BMJ quality & safety
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: Computerized adaptive testing (CAT) provides an alternative to fixed-length assessments. The study validated a suite of computerized adaptive tests for mental health (CAT-MH) in a community psychiatric sample. Methods: A total of 145 adults from a community outpatient clinic, including 19 with no history of a mental disorder (control group), were prospectively evaluated with CAT for depression (CAD-MDD and CAT-DI), mania (CAT-MANIA), and anxiety symptoms (CAT-ANX). Ratings were compared with gold-standard psychiatric assessments, including the Structured Clinical Interview for DSM-IV-TR (SCID), Hamilton Rating Scale for Depression (HAM-D-25), Patient Health Questionnaire (PHQ-9), Center for Epidemiologic Studies Depression Scale (CES-D), and Global Assessment of Functioning (GAF). Results: Sensitivity and specificity for CAD-MDD were .96 and .64, respectively (.96 and 1.00 for major depression versus the control group). CAT for depression severity (CAT-DI) correlated well with the HAM-D-25 (r=.79), PHQ-9 (r=.90), and CES-D (r=.90) and had an odds ratio (OR) of 27.88 across its range for current SCID major depressive disorder. CAT-ANX correlated with the HAM-D-25 (r=.73), PHQ-9 (r=.78), and CES-D (r=.81) and had an OR of 11.52 across its range for current SCID generalized anxiety disorder. CAT-MANIA did not correlate well with the HAM-D-25 (r=.31), PHQ-9 (r=.37), and CES-D (r=.39), but it had an OR of 11.56 across its range for a current SCID bipolar diagnosis. Participants found the CAT-MH acceptable and easy to use, averaging 51.7 items and 9.4 minutes to complete the full battery. Conclusions: Compared with gold-standard diagnostic and assessment measures, CAT-MH provided an effective, rapidly administered assessment of psychiatric symptoms.
    No preview · Article · Jun 2015 · Psychiatric Services
  • [Show abstract] [Hide abstract]
    ABSTRACT: There are many advantages to individual participant data meta-analysis for combining data from multiple studies. These advantages include greater power to detect effects, increased sample heterogeneity, and the ability to perform more sophisticated analyses than meta-analyses that rely on published results. However, a fundamental challenge is that it is unlikely that variables of interest are measured the same way in all of the studies to be combined. We propose that this situation can be viewed as a missing data problem in which some outcomes are entirely missing within some trials and use multiple imputation to fill in missing measurements. We apply our method to five longitudinal adolescent depression trials where four studies used one depression measure and the fifth study used a different depression measure. None of the five studies contained both depression measures. We describe a multiple imputation approach for filling in missing depression measures that makes use of external calibration studies in which both depression measures were used. We discuss some practical issues in developing the imputation model including taking into account treatment group and study. We present diagnostics for checking the fit of the imputation model and investigate whether external information is appropriately incorporated into the imputed values. Copyright © 2015 John Wiley & Sons, Ltd. Copyright © 2015 John Wiley & Sons, Ltd.
    No preview · Article · Jun 2015 · Statistics in Medicine
  • Source
    W. Padula · R.D. Gibbons · R.J. Valuck · M.B. Makic · D. Meltzer

    Full-text · Article · May 2015 · Value in Health
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Classrooms are unique and complex work settings in which teachers and students both participate in and contribute to classroom processes. This article describes the measurement phase of a study that examined the social ecology of urban classrooms. Informed by the dimensions and items of an established measure of organizational climate, we designed the Student Climate Survey (n = 53 items) to assess student psychological climate in third through eighth grades. We administered the survey to 621 students at three time points within one school year in 69 classrooms within eight urban schools. A multidimensional item response theory (IRT) analysis based on a full-information item bifactor model revealed 18 items that loaded on a primary factor and demonstrated good criterion and predictive validity. Opportunities for the Student Climate Survey to advance our contextual understanding of urban classrooms and inform intervention are discussed.
    Full-text · Article · Feb 2015 · The Journal of Early Adolescence
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: The purpose of this study was to determine whether polymorphisms in the serotonin transporter (SLC6A4) and serotonin-2Areceptor (HTR2A) genes are associated with response to escitalopram in patients with autism spectrumdisorder (ASD). Methods: Forty-four participants withASDwere enrolled in a 6 week, forced titration, open label examination of the selective serotonin reuptake inhibitor (SSRI) escitalopram. Doses increased at weekly intervals starting at 2.5mg daily with a maximum possible dose of 20 mg daily achieved by the end of the study. If adverse events were experienced, participants subsequently received the previously tolerated dose for the duration of study. SLC6A4 (5-HTTLPR) and HTR2A (rs7997012) genotype groups were assessed in relation to treatment outcomes and drug doses. Results: Insistence on sameness and irritability symptoms significantly improved over the course of the 6 week treatment period ( p < 0.0001) in this open-label trial. There were no significant differences observed in the rate of symptom improvement over time across genotype groups. Similarly, dosing trajectory was not significantly associated with genotype groups. Conclusions: Previous studies have identified SLC6A4 and HTR2A associations with SSRI response in patients with depression and 5-HTTLPR (SLC6A4) associations with escitalopram response in ASD. We did not observe evidence for similar relationships in this ASD study.
    Full-text · Article · Jan 2015 · Journal of Child and Adolescent Psychopharmacology
  • Source
    Robert D Gibbons · Marcelo Coca Perraillon · Jong Bae Kim
    [Show abstract] [Hide abstract]
    ABSTRACT: The need to harmonize different outcome metrics is a common problem in research synthesis and economic evaluation of health interventions and technology. The purpose of this paper is to describe the use of multidimensional item response theory (IRT) to equate different scales which purport to measure the same construct at the item level. We provide an overview of multidimensional item response theory in general and the bi-factor model which is particularly relevant for applications in this area. We show how both the underlying true scores of two or more scales that are intended to measure the same latent variable can be equated and how the item responses from one scale can be used to predict the item responses for a scale that was not administered but are necessary for the purpose of economic evaluations. As an example, we show that a multidimensional IRT model predicts well both the EQ-5D descriptive system and the EQ-5D preference index from SF-12 data which cannot be directly used to perform an economic evaluation. Results based on multidimensional IRT performed well compared to traditional regression methods in this area. A general framework for harmonization of research instruments based on multidimensional IRT is described.
    Preview · Article · Dec 2014 · Health Services and Outcomes Research Methodology
  • [Show abstract] [Hide abstract]
    ABSTRACT: PurposeIn the 2004, FDA placed a black box warning on antidepressants for risk of suicidal thoughts and behavior in children and adolescents. The purpose of this paper is to examine the risk of suicide attempt and self-inflicted injury in depressed children ages 5–17 treated with antidepressants in two large observational datasets taking account time-varying confounding.METHODS We analyzed two large US medical claims databases (MarketScan and LifeLink) containing 221,028 youth (ages 5–17) with new episodes of depression, with and without antidepressant treatment during the period of 2004–2009. Subjects were followed for up to 180 days. Marginal structural models were used to adjust for time-dependent confounding.ResultsFor both datasets, significantly increased risk of suicide attempts and self-inflicted injury were seen during antidepressant treatment episodes in the unadjusted and simple covariate adjusted analyses. Marginal structural models revealed that the majority of the association is produced by dynamic confounding in the treatment selection process; estimated odds ratios were close to 1.0 consistent with the unadjusted and simple covariate adjusted association being a product of chance alone.Conclusions Our analysis suggests antidepressant treatment selection is a product of both static and dynamic patient characteristics. Lack of adjustment for treatment selection based on dynamic patient characteristics can lead to the appearance of an association between antidepressant treatment and suicide attempts and self-inflicted injury among youths in unadjusted and simple covariate adjusted analyses. Marginal structural models can be used to adjust for static and dynamic treatment selection processes such as that likely encountered in observational studies of associations between antidepressant treatment selection, suicide and related behaviors in youth. Copyright © 2014 John Wiley & Sons, Ltd.
    No preview · Article · Sep 2014 · Pharmacoepidemiology and Drug Safety
  • [Show abstract] [Hide abstract]
    ABSTRACT: Rationale: Most ward risk scores were created using subjective opinion in individual hospitals and only utilize vital signs. Objectives: To develop and validate a risk score using commonly collected electronic health record (EHR) data. Methods: All patients hospitalized on the wards in five hospitals were included in this observational cohort study. Discrete-time survival analysis was used to predict the combined outcome of cardiac arrest (CA), intensive care unit (ICU) transfer, or death on the wards. Laboratory results, vital signs, and demographics were utilized as predictor variables. The model was developed in the first 60% of the data at each hospital and then validated in the remaining 40%. The final model was compared to the Modified Early Warning Score (MEWS) using the area under the receiver operating characteristic curve (AUC). Measurements and Main Results: A total of 269,999 patient admissions were included, with 424 CAs, 13,188 ICU transfers, and 2,840 deaths occurring during the study period. The derived model was more accurate than the MEWS in the validation dataset for all outcomes (AUC 0.83 vs. 0.74 for CA, 0.79 vs. 0.73 for ICU transfer, 0.94 vs. 0.90 for death, and 0.82 vs. 0.76 for the combined outcome; p-value<0.01 for all comparisons). This accuracy improvement was seen across all hospitals. Conclusions: We developed an accurate ward risk stratification tool using commonly collected EHR variables in a large multicenter dataset. Further study is needed to determine whether implementation in real-time would improve patient outcomes.
    No preview · Article · Aug 2014 · American Journal of Respiratory and Critical Care Medicine
  • [Show abstract] [Hide abstract]
    ABSTRACT: Abstract Meta-analysis has been used extensively for evaluation of efficacy and safety of medical interventions. Its advantages and utilities are well known. However, recent studies have raised questions about the accuracy of the commonly used moment-based meta-analytic methods in general and for rare binary outcomes in particular. The issue is further complicated for studies with heterogeneous effect sizes. Likelihood-based mixed-effects modeling provides an alternative to moment-based methods such as inverse-variance weighted fixed- and random-effects estimators. In this paper, we compare and contrast different mixed-effect modeling strategies in the context of meta-analysis. Their performance in estimation and testing of overall effect and heterogeneity are evaluated when combining results from studies with a binary outcome. Models that allow heterogeneity in both baseline rate and treatment effect across studies have low type I and type II error rates, and their estimates are the least biased among the models considered.
    No preview · Article · Jun 2014 · Journal of Biopharmaceutical Statistics
  • [Show abstract] [Hide abstract]
    ABSTRACT: Landfill functional stability provides a target that supports no environmental threat at the relevant point of exposure in the absence of active control systems. With respect to leachate management, this study investigates "gateway" indicators for functional stability in terms of the predictability of leachate characteristics, and thus potential threat to water quality posed by leachate emissions. Historical studies conducted on changes in municipal solid waste (MSW) leachate concentrations over time (longitudinal analysis) have concentrated on indicator compounds, primarily chemical oxygen demand (COD) and biochemical oxygen demand (BOD). However, validation of these studies using an expanded database and larger constituent sets has not been performed. This study evaluated leachate data using a mixed-effects regression model to determine the extent to which leachate constituent degradation can be predicted based on waste age or operational practices. The final dataset analyzed consisted of a total of 1402 samples from 101MSW landfills. Results from the study indicated that all leachate constituents exhibit a decreasing trend with time in the post-closure period, with 16 of the 25 target analytes and aggregate classes exhibiting a statistically significant trend consistent with well-studied indicators such as BOD. Decreasing trends in BOD concentration after landfill closure can thus be considered representative of trends for many leachate constituents of concern.
    No preview · Article · Jun 2014 · Waste Management
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The introduction of shape parameters into statistical distributions provided flexible models that produced better fit to experimental data. The Weibull and gamma families are prime examples wherein shape parameters produce more reliable statistical models than standard exponential models in lifetime studies. In the presence of many independent gamma populations, one may test equality (or homogeneity) of shape parameters. In this paper, we develop two tests for testing shape parameters of gamma distributions using chi-square distributions, stochastic majorization and Schur convexity. The first one tests hypotheses on the shape parameter of a single gamma distribution. We numerically examine the performance of this test and find that it controls Type I error rate for small samples. To compare shape parameters of a set of independent gamma populations, we develop a test that is unbiased in the sense of Schur convexity. These tests are motivated by the need to have simple, easy to use tests and accurate procedures in case of small samples. We illustrate the new tests using three real datasets taken from engineering and environmental science. In addition, we investigate the Bayes Factor in this context and conclude that for small samples, the frequentist approach performs better than the Bayesian approach.
    Full-text · Article · May 2014 · Communication in Statistics- Simulation and Computation
  • David A Brent · Robert Gibbons

    No preview · Article · Apr 2014 · JAMA Internal Medicine
  • [Show abstract] [Hide abstract]
    ABSTRACT: Limited translation of research into practice has prompted study of diffusion and implementation, and development of effective methods of encouraging adoption, dissemination and implementation. Mixed methods techniques offer approaches for assessing and addressing processes affecting implementation of evidence-based interventions. We describe common mixed methods approaches used in dissemination and implementation research, discuss strengths and limitations of mixed methods approaches to data collection, and suggest promising methods not yet widely used in implementation research. We review qualitative, quantitative, and hybrid approaches to mixed methods dissemination and implementation studies, and describe methods for integrating multiple methods to increase depth of understanding while improving reliability and validity of findings.
    No preview · Article · Apr 2014 · Administration and Policy in Mental Health and Mental Health Services Research
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Many factors affect the risks for neurodevelopmental maladies such as autism spectrum disorders (ASD) and intellectual disability (ID). To compare environmental, phenotypic, socioeconomic and state-policy factors in a unified geospatial framework, we analyzed the spatial incidence patterns of ASD and ID using an insurance claims dataset covering nearly one third of the US population. Following epidemiologic evidence, we used the rate of congenital malformations of the reproductive system as a surrogate for environmental exposure of parents to unmeasured developmental risk factors, including toxins. Adjusted for gender, ethnic, socioeconomic, and geopolitical factors, the ASD incidence rates were strongly linked to population-normalized rates of congenital malformations of the reproductive system in males (an increase in ASD incidence by 283% for every percent increase in incidence of malformations, 95% CI: [91%, 576%], p<6×10-5). Such congenital malformations were barely significant for ID (94% increase, 95% CI: [1%, 250%], p = 0.0384). Other congenital malformations in males (excluding those affecting the reproductive system) appeared to significantly affect both phenotypes: 31.8% ASD rate increase (CI: [12%, 52%], p<6×10-5), and 43% ID rate increase (CI: [23%, 67%], p<6×10-5). Furthermore, the state-mandated rigor of diagnosis of ASD by a pediatrician or clinician for consideration in the special education system was predictive of a considerable decrease in ASD and ID incidence rates (98.6%, CI: [28%, 99.99%], p = 0.02475 and 99% CI: [68%, 99.99%], p = 0.00637 respectively). Thus, the observed spatial variability of both ID and ASD rates is associated with environmental and state-level regulatory factors; the magnitude of influence of compound environmental predictors was approximately three times greater than that of state-level incentives. The estimated county-level random effects exhibited marked spatial clustering, strongly indicating existence of as yet unidentified localized factors driving apparent disease incidence. Finally, we found that the rates of ASD and ID at the county level were weakly but significantly correlated (Pearson product-moment correlation 0.0589, p = 0.00101), while for females the correlation was much stronger (0.197, p<2.26×10-16).
    Full-text · Article · Mar 2014 · PLoS Computational Biology
  • Subhash Aryal · Dulal K. Bhaumik · Thomas Mathew · Robert D. Gibbons
    [Show abstract] [Hide abstract]
    ABSTRACT: In this article we derive an optimal test for testing the significance of covariance matrices of random-effects of two multivariate mixed-effects linear models. We compute the power of this newly derived test via simulation for various alternative hypotheses in a bivariate set up for unbalanced designs and observe that power responds sharply when sample size and alternative hypotheses are changed. For some balanced designs we compare power of the optimal test to that of the likelihood ratio test via simulation, and find that the proposed test has greater power than the likelihood ratio test. The results are illustrated using real data on human growth. Other relevant applications of the model are highlighted.
    No preview · Article · Feb 2014 · Journal of Multivariate Analysis
  • Source

    Full-text · Article · Jan 2014 · The Journal of Clinical Psychiatry

  • No preview · Article · Nov 2013 · JAMA Psychiatry
  • [Show abstract] [Hide abstract]
    ABSTRACT: Over 200,000 in-hospital cardiac arrests occur in the United States each year and many of these events may be preventable. Current vital sign-based risk scores for ward patients have demonstrated limited accuracy, which leads to missed opportunities to identify those patients most likely to suffer cardiac arrest and inefficient resource utilization. We derived and validated a prediction model for cardiac arrest while treating ICU transfer as a competing risk using electronic health record data. A retrospective cohort study. An academic medical center in the United States with approximately 500 inpatient beds. Adult patients hospitalized from November 2008 until August 2011 who had documented ward vital signs. None. Vital sign, demographic, location, and laboratory data were extracted from the electronic health record and investigated as potential predictor variables. A person-time multinomial logistic regression model was used to simultaneously predict cardiac arrest and ICU transfer. The prediction model was compared to the VitalPAC Early Warning Score using the area under the receiver operating characteristic curve and was validated using three-fold cross-validation. A total of 56,649 controls, 109 cardiac arrest patients, and 2,543 ICU transfers were included. The derived model more accurately detected cardiac arrest (area under the receiver operating characteristic curve, 0.88 vs 0.78; p < 0.001) and ICU transfer (area under the receiver operating characteristic curve, 0.77 vs 0.73; p < 0.001) than the VitalPAC Early Warning Score, and accuracy was similar with cross-validation. At a specificity of 93%, our model had a higher sensitivity than the VitalPAC Early Warning Score for cardiac arrest patients (65% vs 41%). We developed and validated a prediction tool for ward patients that can simultaneously predict the risk of cardiac arrest and ICU transfer. Our model was more accurate than the VitalPAC Early Warning Score and could be implemented in the electronic health record to alert caregivers with real-time information regarding patient deterioration.
    No preview · Article · Nov 2013 · Critical care medicine

Publication Stats

12k Citations
1,179.76 Total Impact Points


  • 2015
    • The University of Tennessee Medical Center at Knoxville
      Knoxville, Tennessee, United States
  • 1983-2015
    • University of Chicago
      • • Department of Health Studies
      • • Department of Medicine
      • • School of Social Service Administration
      Chicago, Illinois, United States
  • 1984-2011
    • Harvard University
      • Department of Psychology
      Cambridge, MA, United States
  • 1982-2011
    • University of Illinois at Chicago
      • • Division of Epidemiology and Biostatistics
      • • Department of Psychiatry (Chicago)
      • • School of Public Health
      Chicago, Illinois, United States
  • 2008
    • Chicago Center For Family Health
      Chicago, Illinois, United States
  • 2007
    • Keio University
      • Department of Neuropsychiatry
      Edo, Tokyo, Japan
  • 2006-2007
    • Columbia University
      • Department of Psychiatry
      New York, New York, United States
  • 2001
    • University of Pittsburgh
      Pittsburgh, Pennsylvania, United States
  • 1999
    • Children's Memorial Hospital
      Chicago, Illinois, United States
  • 1990
    • IIT Research Institute (IITRI)
      Chicago, Illinois, United States
    • Rush Medical College
      Chicago, Illinois, United States
  • 1989
    • Universidad de Extremadura
      Ara Pacis Augustalis, Extremadura, Spain
  • 1987
    • Loyola University Chicago
      • Stritch School of Medicine
      Chicago, Illinois, United States
  • 1986
    • Illinois State University
      State College, Pennsylvania, United States