Jichun Xie

Positions:

Associate Professor of Biostatistics & Bioinformatics

Integrative Genomics
School of Medicine

Member of the Duke Cancer Institute

Duke Cancer Institute
School of Medicine

Education:

Ph.D. 2011

University of Pennsylvania

Grants:

Duke CTSA (UL1)

Administered By
Institutes and Centers
Awarded By
National Institutes of Health
Role
Biostatistician Investigator
Start Date
End Date

Bioinformatics and Computational Biology Training Program

Administered By
Basic Science Departments
Awarded By
National Institutes of Health
Role
Mentor
Start Date
End Date

A hands-on, integrative next-generation sequencing course: design, experiment, and analysis

Administered By
Integrative Genomics
Awarded By
National Institutes of Health
Role
Training Faculty
Start Date
End Date

Race-Related Alternative Splicing: Novel Targets in Prostate Cancer

Administered By
Medicine, Medical Oncology
Awarded By
National Institutes of Health
Role
Biostatistician
Start Date
End Date

Statistical/Computational Methods for Pharmacogenomics and Individualized Therapy

Administered By
Integrative Genomics
Awarded By
University of North Carolina - Chapel Hill
Role
Co Investigator
Start Date
End Date

Publications:

Increasing Rates of Imaging in Failed Back Surgery Syndrome Patients: Implications for Spinal Cord Stimulation.

BACKGROUND: Failed back surgery syndrome (FBSS) has a high incidence following spinal surgery, is notoriously refractory to treatment, and results in high health care utilization. Spinal cord stimulation (SCS) is a well-accepted modality for pain relief in this population; however, until recently magnetic resonance imaging (MRI) was prohibited due to risk of heat conduction through the device. OBJECTIVES: We examined trends in imaging use over the past decade in patients with FBSS to determine its impact on health care utilization and implications for patients receiving SCS. STUDY DESIGN: Retrospective. SETTING: Inpatient and outpatient sample. METHODS: We identified patients from 2000 to 2012 using the Truven MarketScan database. Annual imaging rates (episodes per 1000 patient months) were determined for MRI, computed tomography (CT) scan, x-ray, and ultrasound. A multivariate Poisson regression model was used to determine imaging trends over time, and to compare imaging in SCS and non-SCS populations. RESULTS: A total of 311,730 patients with FBSS were identified, of which 5.17% underwent SCS implantation (n = 16,118). The median (IQR) age was 58.0 (49.0 - 67.0) years. Significant increases in imaging rate ratios were found in all years for each of the modalities. Increases were seen in the use of CT scans (rate ratio [RR] = 3.03; 95% confidence interval [CI]: 2.79 - 3.29; P < 0.0001), MRI (RR = 1.73; 95% CI: 1.61 - 1.85; P < 0.0001), ultrasound (RR = 2.00; 95% CI: 1.84 - 2.18; P < 0.0001), and x-ray (RR = 1.10; 95% CI: 1.05 - 1.15; P < 0.0001). Despite rates of MRI in SCS patients being half that in the non-SCS group, these patients underwent 19% more imaging procedures overall (P < 0.0001). SCS patients had increased rates of x-ray (RR = 1.27; 95% CI: 1.25 - 1.29), CT scans (RR = 1.32; 95% CI: 1.30 - 1.35), and ultrasound (RR = 1.10; 95% CI: 1.07 - 1.13) (all P < 0.0001). LIMITATIONS: This study is limited by a lack of clinical and historical variables including the complexity of prior surgeries and pain symptomatology. Miscoding cannot be precluded, as this sample is taken from a large nationwide database. CONCLUSIONS: We found a significant trend for increased use of advanced imaging modalities between the years 2000 and 2012 in FBSS patients. Those patients treated with SCS were 50% less likely to receive an MRI (as expected, given prior incompatibility of neuromodulation devices), yet 32% and 27% more likely to receive CT and x-ray, respectively. Despite the decrease in the use of MRI in those patients treated with SCS, their overall imaging rate increased by 19% compared to patients without SCS. This underscores the utility of MR-conditional SCS systems. These findings demonstrate that imaging plays a significant role in driving health care expenditures. This is the largest analysis examining the role of imaging in the FBSS population and the impact of SCS procedures. Further studies are needed to assess the impact of MRI-conditional SCS systems on future trends in imaging in FBSS patients receiving neuromodulation therapies. Key words: Failed back surgery syndrome, spinal cord stimulation, imaging, health care utilization, MRI, chronic pain, back pain, neuromodulation.
Authors
Farber, SH; Han, JL; Petraglia Iii, FW; Gramer, R; Yang, S; Pagadala, P; Parente, B; Xie, J; Petrella, JR; Lad, SP
MLA Citation
Farber, S. Harrison, et al. “Increasing Rates of Imaging in Failed Back Surgery Syndrome Patients: Implications for Spinal Cord Stimulation..” Pain Physician, vol. 20, no. 6, Sept. 2017, pp. E969–77.
URI
https://scholars.duke.edu/individual/pub1276265
PMID
28934801
Source
pubmed
Published In
Pain Physician
Volume
20
Published Date
Start Page
E969
End Page
E977

A sparse structured shrinkage estimator for nonparametric varying-coefficient model with an application in genomics

Many problems in genomics are related to variable selection where high-dimensional genomic data are treated as covariates. Such genomic covariates often have certain structures and can be represented as vertices of an undirected graph. Biological processes also vary as functions depending upon some biological state, such as time. High-dimensional variable selection where covariates are graph-structured and underlying model is nonparametric presents an important but largely unaddressed statistical challenge. Motivated by the problem of regression-based motif discovery, we consider the problem of variable selection for high-dimensional nonparametric varying-coefficient models and introduce a sparse structured shrinkage (SSS) estimator based on basis function expansions and a novel smoothed penalty function. We present an efficient algorithm for computing the SSS estimator. Results on model selection consistency and estimation bounds are derived. Moreover, finite-sample performances are studied via simulations, and the effects of high-dimensionality and structural information of the covariates are especially highlighted. We apply our method to motif finding problem using a yeast cell-cycle gene expression dataset and word counts in genes' promoter sequences. Our results demonstrate that the proposed method can result in better variable selection and prediction for high-dimensional regression when the underlying model is nonparametric and covariates are structured. Supplemental materials for the article are available online. © 2012 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America.
Authors
John Daye, Z; Xie, J; Li, H
MLA Citation
John Daye, Z., et al. “A sparse structured shrinkage estimator for nonparametric varying-coefficient model with an application in genomics.” Journal of Computational and Graphical Statistics, vol. 21, no. 1, Apr. 2012, pp. 110–33. Scopus, doi:10.1198/jcgs.2011.10102.
URI
https://scholars.duke.edu/individual/pub1099913
Source
scopus
Published In
Journal of Computational and Graphical Statistics : a Joint Publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America
Volume
21
Published Date
Start Page
110
End Page
133
DOI
10.1198/jcgs.2011.10102

Prevalence and Cost Analysis of Chronic Pain After Hernia Repair: A Potential Alternative Approach With Neurostimulation.

OBJECTIVES: Chronic pain (CP) affects a significant number of patients following hernia repair, ranging from 11 to 54% in the literature. The aim of this study was to assess the prevalence, overall costs, and health care utilization associated with CP after hernia repair. MATERIALS AND METHODS: A retrospective longitudinal study was performed using the Truven MarketScan® data base to identify patients who develop chronic neuropathic posthernia repair pain from 2001 to 2012. Patients were grouped into CP and No Chronic Pain (No CP) cohorts. Patients were excluded if they 1) were under 18 years of age; 2) had a previous pain diagnosis; 3) had CP diagnosed <90 days after the index hernia repair; 4) had less than one year of follow-up; or 5) had less than one-year baseline record before hernia repair. Patients were grouped into the CP cohort if their CP diagnosis was made within the two years following index hernia repair. Total, outpatient, and pain prescription costs were collected in the period of five years prehernia to nine years posthernia repair. A longitudinal multivariate analysis was used to model the effects of chronic neuropathic posthernia repair pain on total inpatient/outpatient and pain prescription costs. RESULTS: We identified 76,173 patients who underwent hernia repair and met inclusion criteria (CP: n = 14,919, No CP: n = 61,254). There was a trend for increased total inpatient/outpatient and pain prescription costs one-year posthernia repair, when compared to baseline costs for both cohorts. In both cohorts, total inpatient/outpatient costs remained elevated from baseline through nine years posthernia repair, with the CP cohort experiencing significantly higher cumulative median costs (CP: $51,334, No CP: $37,388). The CP diagnosis year was associated with a 1.75-fold increase (p < 0.001) in total inpatient/outpatient costs and a 2.26-fold increase (p < 0.001) in pain prescription costs versus all other years. In the longitudinal analysis, the CP cohort had a 1.14-fold increase (p < 0.001) in total inpatient/outpatient costs and 2.00-fold increase (p < 0.001) in pain prescription costs. CONCLUSIONS: Our study demonstrates the prevalence of CP after hernia surgery to be nearly 20%, with significantly increased costs and healthcare resource utilization. While current treatment paradigms are effective for many, there remains a large number of patients that could benefit from an overall approach that includes nonopioid treatments, such as potentially incorporating neurostimulation, for CP that presents posthernia repair.
Authors
Elsamadicy, AA; Ashraf, B; Ren, X; Sergesketter, AR; Charalambous, L; Kemeny, H; Ejikeme, T; Yang, S; Pagadala, P; Parente, B; Xie, J; Pappas, TN; Lad, SP
MLA Citation
Elsamadicy, Aladine A., et al. “Prevalence and Cost Analysis of Chronic Pain After Hernia Repair: A Potential Alternative Approach With Neurostimulation..” Neuromodulation, vol. 22, no. 8, Dec. 2019, pp. 960–69. Pubmed, doi:10.1111/ner.12871.
URI
https://scholars.duke.edu/individual/pub1355141
PMID
30320933
Source
pubmed
Published In
Neuromodulation
Volume
22
Published Date
Start Page
960
End Page
969
DOI
10.1111/ner.12871

Long-term Cost Utility of Spinal Cord Stimulation in Patients with Failed Back Surgery Syndrome.

BACKGROUND: Failed back surgery syndrome (FBSS) is a cause of significant morbidity for up to 40% of patients following spine surgery, and is estimated to cost almost $20 billion. Treatment options for these patients currently include conventional medical management (CMM), repeat operation, or spinal cord stimulation (SCS). Much of the published data regarding cost effectiveness of SCS comprise smaller scale randomized controlled trials (RCTs) rather than large databases capturing practices throughout the US. SCS has been shown to have superior outcomes to CMM or repeat spinal operation in several landmark studies, yet there are few large studies examining its long-term economic impact. OBJECTIVES: This study compares health care utilization for SCS compared to other management in patients with FBSS. STUDY DESIGN: Retrospective. SETTING: Inpatient and outpatient sample. METHODS: Patients with a history of FBSS from 2000 to 2012 were selected. We compared those who received SCS to those who underwent conventional management. A longitudinal analysis was used to model the value of log(cost) in each one year interval using a generalized estimating equations (GEE) model to account for the correlation of the same patient's cost in multiple years. Similarly, a Poisson GEE model with the log link was applied to correlated count outcomes. RESULTS: We identified 122,827 FBSS patients. Of these, 5,328 underwent SCS implantation (4.34%) and 117,499 underwent conventional management. Total annual costs decreased over time following implantation of the SCS system, with follow-up analysis at 1, 3, 6, and 9 years. The longitudinal GEE model demonstrated that placement of an SCS system was associated with an initial increase in total costs at the time of implantation (cost ratio [CR]: 1.74; 95% confidence interval [CI]: 1.41, 2.15, P < 0.001), however there was a significant and sustained 68% decrease in cost in the year following SCS placement (CR: 0.32; 95% CI: 0.24, 0.42, P < 0.001) compared to CMM. There was also an aggregate time trend that for each additional year after SCS, cost decreased on average 40% percent annually (CR: 0.60; 95% CI: 0.55, 0.65, P < 0.001), with follow-up up to 1, 3, 6, and 9 years post-procedure. LIMITATIONS: Costs are not correlated with patient outcomes, patients are not stratified in terms of complexity of prior back surgery, as well as inherent limitations of a retrospective analysis. CONCLUSIONS: We found that from 2000 to 2012, only 4.3% of patients across the United States with FBSS were treated with SCS. Long-term total annual costs for these patients were significantly reduced compared to patients with conventional management. Although implantation of an SCS system results in a short-term increase in costs at one year, the subsequent annual cumulative costs were significantly decreased long-term in the following 9 years after implantation. This study combines the largest group of FBSS patients studied to date along with the longest follow-up interval ever analyzed. Since SCS has repeatedly been shown to have superior efficacy to CMM in randomized clinical trials, the current study demonstrating improved long-term health economics at 1, 3, 6, and 9 years supports the long-term cost utility of SCS in the treatment of FBSS patients. Key words: Failed back surgery syndrome, spinal cord stimulation, back pain, leg pain, neuromodulation, FBSS, SCS.
Authors
Farber, SH; Han, JL; Elsamadicy, AA; Hussaini, Q; Yang, S; Pagadala, P; Parente, B; Xie, J; Lad, SP
MLA Citation
Farber, S. Harrison, et al. “Long-term Cost Utility of Spinal Cord Stimulation in Patients with Failed Back Surgery Syndrome..” Pain Physician, vol. 20, no. 6, Sept. 2017, pp. E797–805.
URI
https://scholars.duke.edu/individual/pub1276266
PMID
28934786
Source
pubmed
Published In
Pain Physician
Volume
20
Published Date
Start Page
E797
End Page
E805

Sample size and power analysis for sparse signal recovery in genome-wide association studies.

Genome-wide association studies have successfully identified hundreds of novel genetic variants associated with many complex human diseases. However, there is a lack of rigorous work on evaluating the statistical power for identifying these variants. In this paper, we consider sparse signal identification in genome-wide association studies and present two analytical frameworks for detailed analysis of the statistical power for detecting and identifying the disease-associated variants. We present an explicit sample size formula for achieving a given false non-discovery rate while controlling the false discovery rate based on an optimal procedure. Sparse genetic variant recovery is also considered and a boundary condition is established in terms of sparsity and signal strength for almost exact recovery of both disease-associated variants and nondisease-associated variants. A data-adaptive procedure is proposed to achieve this bound. The analytical results are illustrated with a genome-wide association study of neuroblastoma.
Authors
Xie, J; Cai, TT; Li, H
MLA Citation
Xie, Jichun, et al. “Sample size and power analysis for sparse signal recovery in genome-wide association studies..” Biometrika, vol. 98, no. 2, June 2011, pp. 273–90. Pubmed, doi:10.1093/biomet/asr003.
URI
https://scholars.duke.edu/individual/pub1099898
PMID
23049128
Source
pubmed
Published In
Biometrika
Volume
98
Published Date
Start Page
273
End Page
290
DOI
10.1093/biomet/asr003