Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website

RCS 05-196 – HSR Study

RCS 05-196
Research Career Scientist Award
Xiao-Hua Andrew Zhou, PhD MSc
VA Puget Sound Health Care System Seattle Division, Seattle, WA
Seattle, WA
Funding Period: October 2007 - September 2018
1. Methods in Personalized Medicine:
Background: An important area of medical research is personalized medicine, whose goal is to identify the best treatment to achieve the optimal clinical outcome based on a patient's individual profile, including both genetical and clinical information.

2. Modeling of Health Care Costs of Veterans with Chronic Diseases:
Background: New cost models are needed to allow for the accurate identification of disease-attributable costs and investigation of modifiable patient and system factors associated with variation within important care areas.

3. A Joint Evaluation of Surgery-Related Outcomes and Costs across VAMCs:
Background: VHA has been strongly committed to reviewing and improving the quality of care provided to veterans. One of the best examples of this activity has been the National Surgical Quality Improvement Program (NSQIP). However, 30-day mortality focused in the NSQIP only provides a limited picture of surgical quality.

4. Preventing Vein Graft Stenosis in Peripheral Vascular Surgery:
Background: Vein graft stenosis and failure is a serious clinical problem in peripheral arterial disease (PAD) that requires additional research into the key contributing biological pathways.

5. Re-identification Risk of VA De-Identified Patient Health Data:
Background: Privacy, security, and confidentiality of patient level health care data are cornerstones of VA health care. Re-identification is a technique for removing identifying information from a data set before it is released for research or public use. The potential to re-identify individual Veterans in de-identified datasets will become an increasing concern as the size and complexity of electronic health data across the nation.

6. INtegrated Care After Exacerbation of COPD (InCaseE):
Background: Chronic obstructive pulmonary disease (COPD) exacerbations are common among Veterans admitted to hospital, lead to decrements in health-related quality of life, and are important drivers of health care expenditures. An intervention to improve COPD care is needed, not only to treat patients for COPD and their accompanying comorbidities, but also to redesign the care delivery system, such as specialties treating patients within Patient Aligned Care Teams (PACT). Determining how to deploy existing specialties using a PACT-Veteran-centric approach is important to improve access, timeliness, and quality of care.

7. Semi-parametric Statistical Methods for Predicting High-cost VA Patients Using High-Dimensional Covariates:
Background: The rising demands and health care costs make it urgent to develop new statistical
methods to accurately predict high-costs VA patients and important risk factors associated with high costs. Health care cost data are characterized by a high level of skewness and heteroscedastic variances. The large number of variables collected in the VA database provides rich information, but at the same time, imposes great challenges for statistical analysis and computation. The administrative and electronic medical record data from VA databases often contain missing data. The new statistical procedure we propose aims to take advantage of the rich databases in VA for analyzing costs data. It employs and develops state-of-art high-dimensional semiparametric statistical procedures to handle the complexity of VA data sets.

1. Methods in Personalized Medicine:
Objectives: Develop new statistical methods, called covariate specific treatment effect (CSTE) curve to select the best treatment for individual patients based on patients' marker values.

2. Modeling of Health Care Costs of Veterans with Chronic Diseases:
Objectives: Develop robust non-parametric cost models that allow for detailed evaluation of costs within the VA.

3. A Joint Evaluation of Surgery-Related Outcomes and Costs across VAMCs:
Objectives: Develop new statistical methods for evaluating short- and long-term costs and develop new statistical methods for assessing surgical care efficiency via jointly modeling surgical morbidity, mortality, and costs.

4. Preventing Vein Graft Stenosis in Peripheral Vascular Surgery:
Objectives: Identify the key biological pathways that contribute to a pathological thrombo-inflammatory response and "at risk" patients by phenotype.

5. Re-identification Risk of VA De-Identified Patient Health Data:
Objectives: (1) Obtain a comprehensive understanding of issues related to the re-identification risk of VA patient health data that meets HIPAA de-identification criteria;
(2) develop standardized statistical procedures to evaluate the re-identification risk of specific HIPAA deidentified VA patient health databases.

6. INtegrated Care After Exacerbation of COPD (InCaseE):
Objectives: Test a novel intervention that is aligned with VA operational goals and seeks to improve the quality of care among patients with COPD, improve their quality of life, and reduce their hospital re-admissions and mortality.

7. Semi-parametric Statistical Methods for Predicting High-cost VA Patients Using High-Dimensional Covariates:
Objectives: Develop a High Costs Prediction (HCP) system, which employs novel high-dimensional semiparametric statistical methods and algorithms to analyze large VA database with missing values and occurrence of censoring. Combine the HCP system with the existing Care Assessment Needs Scoring (CAN) system, in order to make important progress toward the goal of building a data-driven decision support system.

1. Methods in Personalized Medicine:
Methods: We developed a new statistical method, called conditional average and quantile treatment effect (CSTE) curves, to select the optimal treatment based the patient's marker value when the clinical outcome of the patient is binary and continuous. We also used B-spine data to estimate these curves.

2. Modeling of Health Care Costs of Veterans with Chronic Diseases:
Methods: We utilized semi-parametric and non-parametric models to evaluate costs within the VA.

3. A Joint Evaluation of Surgery-Related Outcomes and Costs across VAMCs:
Methods: We received data from the National Surgical Quality Improvement Program (VASQIP) and used it to analyze facility variability in complication rates and costs for the seven common surgical procedures in the VHA.

4. Preventing Vein Graft Stenosis in Peripheral Vascular Surgery:
Methods: We conducted a prospective, longitudinal study of PAD patients who underwent leg bypass surgery with autogenous vein.

5. Re-identification Risk of VA De-Identified Patient Health Data:
Methods: We performed a literature review on existing statistical concepts and methods for re-identification risk. We also applied the existing methods to a de-identified VA Rheumatoid Arthritis Registry database to assess the de-identification risk of releasing it and made recommendations to VACO about the release of future VA datasets.

6. INtegrated Care After Exacerbation of COPD (InCaseE):
Methods: We are using a randomized stepped-wedge design to evaluate a multifaceted intervention that seeks to improve quality-of-life and decrease rate of hospital readmission and mortality among patients with COPD.

7. Semi-parametric Statistical Methods for Predicting High-cost VA Patients Using High-Dimensional Covariates:
Methods: We are developing a novel semiparametric procedure for predicting high costs patients. The approach we propose incorporates high-dimensional covariates and nonlinear covariate effects and addresses the challenge of censoring by death, which improves accuracy and increases the flexibility of modeling. We will link data from the Managerial Cost Accounting System (MCA, formerly Decision Support System or DSS) with three VA databases including: the VA Patient Treatment File (PTF); the VA Outpatient Clinic File (OCF); and the VA Beneficiary Identification and Records Locator Subsystem death file. We will compare the newly proposed methods with existing methods using both the VA data and simulated data.

Not yet available.

1. Methods in Personalized Medicine:
Impact: The results contribute novel and critical clinical information for the welfare of veterans and address treatment of veterans with colorectal cancer. Our statistical methods contribute to statistical literature in personalize medicine, which is one of most active and important research areas in medicine.

2. Modeling of Health Care Costs of Veterans with Chronic Diseases:
Impact: The knowledge gained provides researchers with tools for examining the VA medical care costs with increased precision, to provide more efficient care.

3. A Joint Evaluation of Surgery-Related Outcomes and Costs across VAMCs:
Impact: The results from this project provide objective comparisons on VA surgical care across VA facilities and add an additional economic dimension to current assessment. They provide the knowledge to better reward efficient care that improves health. They also improve our understanding of how high quality care can be provided at low cost.

4. Preventing Vein Graft Stenosis in Peripheral Vascular Surgery:
Impact: By identifying biological pathways and phenotypes for a pathological thrombo, we can target those patients in greatest need of treatment, while sparing low risk patients from the adverse effects of treatment.

5. Re-identification Risk of VA De-Identified Patient Health Data:
Impact: The project adds to the methodology and science of data protection and informs VA policy on release of VA data. This benefits veterans by reducing the risk that veteran data will be compromised, as well as allowing datasets to be more safely and effectively shared, facilitating important future research and policies serving veterans.

6. INtegrated Care After Exacerbation of COPD (InCaseE):
Impact: The intervention aims to help veterans with COPD to achieve better clinical outcomes by providing evidence about how VA may expand the responsibilities of specialists to better support patients during high risk periods.

7. Semi-parametric Statistical Methods for Predicting High-cost VA Patients Using High-Dimensional Covariates:
Impact: The proposed High Costs Prediction (HCP) system will improve care allocation by identifying patients who are at high-risk of incurring high costs within a subsequent one-year period. Targeting care to these patients can reduce avoidable use of health care services and have a positive impact on reducing costs. The HCP system also allows us to identify disease areas that contribute significantly to high health care costs which policymakers can target by future intervention.

External Links for this Project

NIH Reporter

Grant Number: IK6RX002991-01

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


Journal Articles

  1. Lin H, Liu D, Zhou XH. A correlated random-effects model for normal longitudinal data with nonignorable missingness. Statistics in medicine. 2010 Jan 30; 29(2):236-47. [view]
  2. Lin H, Zhou XA, Li G. A direct semiparametric receiver operating characteristic curve regression with unknown link and baseline functions. Statistica Sinica. 2012 Oct 15; 22(4):1427-1456. [view]
  3. Tang L, Zhou XH. A general framework of marker design with optimal allocation to assess clinical utility. Statistics in medicine. 2013 Feb 20; 32(4):620-30. [view]
  4. Chen B, Zhou XH. A latent-variable marginal method for multi-level incomplete binary data. Statistics in medicine. 2012 Nov 20; 31(26):3211-22. [view]
  5. Liu D, Zhou XH. A model for adjusting for nonignorable verification bias in estimation of the ROC curve and its area with likelihood-based approach. Biometrics. 2010 Dec 1; 66(4):1119-28. [view]
  6. Li JL, Zhou XA, Fine JP. A regression approach to ROC surface, with applications to Alzheimer's disease. Science China Mathematics. 2012 Aug 15; 55(8):1583-95. [view]
  7. Lin H, Zhou XH. A semiparametric 2-part mixed-effects heteroscedastic transformation model for correlated right-skewed semicontinuous data. Biostatistics. 2009 Oct 1; 10(4):640-58. [view]
  8. Tang LL, Zhou XH. A semiparametric separation curve approach for comparing correlated ROC data from multiple markers. Journal of computational and graphical statistics : a joint publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America. 2012 Aug 16; 21(3):662-676. [view]
  9. Hedrick SC, Guihan M, Chapko MK, Sullivan J, Zhou XH, Manheim LM, Forsberg CW, Mambourg FJ. Assisted living pilot program: health outcomes. Journal of aging and health. 2009 Feb 1; 21(1):190-207. [view]
  10. Chapko MK, Manheim LM, Guihan M, Sullivan JH, Zhou XH, Wang L, Mambourg FJ, Hedrick SC, Hedrick SC. Assisted living pilot program: utilization and cost findings. Journal of aging and health. 2009 Feb 1; 21(1):208-25. [view]
  11. Zhou XA, Ma Y. BATE curve in assessment of clinical utility of predictive biomarkers. Science China Mathematics. 2012 Aug 15; 55(8):1529-1552. [view]
  12. Zhou XH, Li SL, Tian F, Cai BJ, Xie YM, Pei Y, Kang S, Fan M, Li JP. Building a disease risk model of osteoporosis based on traditional Chinese medicine symptoms and western medicine risk factors. Statistics in medicine. 2012 Mar 30; 31(7):643-52. [view]
  13. Duan XG, Zhou XA. Composite quantile regression for the receiver operating characteristic curve. Biometrika. 2013 Jul 29; 100(4):889-900. [view]
  14. Chen B, Zhou XH. Doubly Robust Estimates for Binary Longitudinal Data Analysis with Missing Response and Missing Covariates. Biometrics. 2011 Sep 1. [view]
  15. Wang Z, Zhou XH, Wang M. Evaluation of diagnostic accuracy in detecting ordered symptom statuses without a gold standard. Biostatistics. 2011 Jul 1; 12(3):567-81. [view]
  16. Feng P, Zhou XH, Zou QM, Fan MY, Li XS. Generalized propensity score for estimating the average treatment effect of multiple treatments. Statistics in medicine. 2012 Mar 30; 31(7):681-97. [view]
  17. Tang LL, Liu A, Schisterman EF, Zhou XH, Liu CC. Homogeneity tests of clustered diagnostic markers with applications to the BioCycle Study. Statistics in medicine. 2012 Dec 10; 31(28):3638-48. [view]
  18. Chen H, Geng Z, Zhou XH. Identifiability and estimation of causal effects in randomized trials with noncompliance and completely nonignorable missing data. Biometrics. 2009 Sep 1; 65(3):675-82. [view]
  19. Hsieh HN, Su HY, Zhou XH. Interval estimation for the difference in paired areas under the ROC curves in the absence of a gold standard test. Statistics in medicine. 2009 Nov 10; 28(25):3108-23. [view]
  20. Shuai P, Zhou XH, Lao L, Li X. Issues of design and statistical analysis in controlled clinical acupuncture trials: an analysis of English-language reports from Western journals. Statistics in medicine. 2012 Mar 30; 31(7):606-18. [view]
  21. Liu W, Zhang B, Zhang Z, Zhou XH. Joint modeling of transitional patterns of Alzheimer's disease. PLoS ONE. 2013 Sep 20; 8(9):e75487. [view]
  22. Sobel M, Moreno K, Yagi M, Kohler T, Tang G, Clowes A, Zhou XA, Eugenio E. Low levels of a natural IgM antibody are associated with vein graft stenosis and failure. Journal of Vascular Surgery. 2013 Jul 15; 58(4):997-1005. [view]
  23. Taylor L, Zhou XH. Methods for clustered encouragement design studies with noncompliance and missing data. Biostatistics. 2011 Apr 1; 12(2):313-26. [view]
  24. Chen B, Zhou XH. Non-homogeneous Markov process models with informative observations with an application to Alzheimer's disease. Biometrical journal. Biometrische Zeitschrift. 2011 May 1. [view]
  25. Song X, Zhou XH, Ma S. Nonparametric receiver operating characteristic-based evaluation for survival outcomes. Statistics in medicine. 2012 Oct 15; 31(23):2660-75. [view]
  26. Estes A, Munson J, Dawson G, Koehler E, Zhou XH, Abbott R. Parenting stress and psychological functioning among mothers of preschool children with autism and developmental delay. Autism : The International Journal of Research and Practice. 2009 Jul 1; 13(4):375-87. [view]
  27. Sachs MC, Zhou XH. Partial summary measures of the predictiveness curve. Biometrical journal. Biometrische Zeitschrift. 2013 Jul 1; 55(4):589-602. [view]
  28. Hernandez S, McClendon MJ, Zhou XH, Sachs M, Lerner AJ. Pharmacological treatment of Alzheimer's disease: effect of race and demographic variables. Journal of Alzheimer's Disease : Jad. 2010 Jan 1; 19(2):665-72. [view]
  29. Wang H, Zhou XA. Quantile regression for estimating conditional means of health care costs. Biometrika. 2010 Feb 10; 97(1):147-158. [view]
  30. Wang Z, Zhou XH. Random effects models for assessing diagnostic accuracy of traditional Chinese doctors in absence of a gold standard. Statistics in medicine. 2012 Mar 30; 31(7):661-71. [view]
  31. Taylor L, Zhou XH. Relaxing latent ignorability in the ITT analysis of randomized studies with missing data and noncompliance. AACN clinical issues. 2009 Apr 1; 19(2):749-764. [view]
  32. Liu D, Zhou XH. ROC analysis in biomarker combination with covariate adjustment. Academic Radiology. 2013 Jul 1; 20(7):874-82. [view]
  33. Lin H, Zhou L, Peng H, Zhou XA. Selection and combination of biomarkers using the optimal ROC curve for disease classification and prediction. The Canadian journal of statistics = Revue canadienne de statistique. 2011 Jun 1; 39(2):324-343. [view]
  34. Rajan KB, Zhou XH. Semi-parametric area under the curve regression method for diagnostic studies with ordinal data. Biometrical journal. Biometrische Zeitschrift. 2012 Jan 1; 54(1):143-56. [view]
  35. Liu D, Zhou XH. Semiparametric estimation of the covariate-specific ROC curve in presence of ignorable verification bias. Biometrics. 2011 Sep 1; 67(3):906-16. [view]
  36. Skaron A, Li K, Zhou XH. Statistical methods for MRMC ROC studies. Academic Radiology. 2012 Dec 1; 19(12):1499-507. [view]
  37. Zhou XH, Hu N, Hu G, Root M. Synthesis analysis of regression models with a continuous outcome. Statistics in medicine. 2009 May 15; 28(11):1620-35. [view]
  38. Young KY, Laird A, Zhou XH. The efficiency of clinical trial designs for predictive biomarker validation. Clinical trials (London, England). 2010 Oct 1; 7(5):557-66. [view]
  39. Zhou XH, Johnson LL. The statistical challenges in clinical studies. Preface. Statistics in medicine. 2012 Mar 30; 31(7):601. [view]
  40. Zhou XH, Chen B, Xie YM, Tian F, Liu H, Liang X. Variable selection using the optimal ROC curve: an application to a traditional Chinese medicine study on osteoporosis disease. Statistics in medicine. 2012 Mar 30; 31(7):628-35. [view]

  1. Zhou XA, McClish DK, Obuchowski DM. Statistical Methods in Diagnostic Medicine. 2nd ed. New York: Wiley & Sons; 2010. 576 p. [view]
Conference Presentations

  1. Zhou XH, Ding XB. A Nonparametric Heteroscedastic Transformation Model for VA Cost Data. Presented at: VA HSR&D / QUERI National Meeting; 2012 Jul 18; National Harbor, MD. [view]
  2. Zhou XA. Development of Biomarkers: From Exploration to Qualification of Clinical Utility for Therapeutics. Presented at: Federal Drug Administration Center for Drug Evaluation and Research Lecture; 2008 Nov 3; Silver Spring, MD. [view]
  3. Zhou XA, Atkins D, Taylor L. Mediation Analysis in Health Services Research. Presented at: VA HSR&D National Meeting; 2011 Feb 17; Baltimore, MD. [view]
  4. Zhou XA. New Advances in Comparative Effectiveness Research and Related Issues: Generalized partially linear models for incomplete longitudinal data in the presence of population-level information. Paper presented at: IMS-SWUFE China International Conference on Statistics and Probability; 2013 Jul 1; Chengdu, China. [view]
  5. Zhou XH, Kim HM. Statistical Challenges and Methods in Using VA Administrative Data for Comparative Studies. Presented at: VA HSR&D National Meeting; 2009 Feb 11; Washington, DC. [view]

DRA: Health Systems
DRE: none
Keywords: none
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.