Elizabeth DeLong

Overview:

Developing methods for assessing, comparing, and validating statistical models of short and long term outcomes.
Developing and testing methods for evaluating comparative effectiveness in observational data. 

Positions:

Professor Emeritus of Biostatistics and Bioinformatics

Biostatistics & Bioinformatics
School of Medicine

Chair, Department of Biostatistics and Bioinformatics

Biostatistics & Bioinformatics
School of Medicine

Member in the Duke Clinical Research Institute

Duke Clinical Research Institute
School of Medicine

Member of the Duke Cancer Institute

Duke Cancer Institute
School of Medicine

Education:

M.A. 1970

The University of Maine, Orono

Ph.D. 1979

University of North Carolina - Chapel Hill

Grants:

Putting Patients at the Center of Kidney Care Transitions

Administered By
Medicine, General Internal Medicine
Awarded By
Patient Centered Outcomes Research Institute
Role
Co Investigator
Start Date
End Date

IPA - Andrzej Kosinski

Administered By
Biostatistics & Bioinformatics
Awarded By
Durham Veterans Affairs Medical Center
Role
Principal Investigator
Start Date
End Date

Making better use of randomized trials: assessing applicability & transporting causal effects

Administered By
Biostatistics & Bioinformatics
Awarded By
Brown University
Role
Principal Investigator
Start Date
End Date

Integrated Biostatistical Training for CVD Research

Administered By
Biostatistics & Bioinformatics
Awarded By
North Carolina State University
Role
Mentor
Start Date
End Date

Engaging, Inspiring, and Preparing the Next Generation of Biostatisticians

Administered By
Biostatistics & Bioinformatics
Awarded By
North Carolina State University
Role
Principal Investigator
Start Date
End Date

Publications:

Empirical use of causal inference methods to evaluate survival differences in a real-world registry vs those found in randomized clinical trials.

With heighted interest in causal inference based on real-world evidence, this empirical study sought to understand differences between the results of observational analyses and long-term randomized clinical trials. We hypothesized that patients deemed "eligible" for clinical trials would follow a different survival trajectory from those deemed "ineligible" and that this factor could partially explain results. In a large observational registry dataset, we estimated separate survival trajectories for hypothetically trial-eligible vs ineligible patients under both coronary artery bypass surgery (CABG) and percutaneous coronary intervention (PCI). We also explored whether results would depend on the causal inference method (inverse probability of treatment weighting vs optimal full propensity matching) or the approach to combine propensity scores from multiple imputations (the "across" vs "within" approaches). We found that, in this registry population of PCI/CABG multivessel patients, 32.5% would have been eligible for contemporaneous RCTs, suggesting that RCTs enroll selected populations. Additionally, we found treatment selection bias with different distributions of propensity scores between PCI and CABG patients. The different methodological approaches did not result in different conclusions. Overall, trial-eligible patients appeared to demonstrate at least marginally better survival than ineligible patients. Treatment comparisons by eligibility depended on disease severity. Among trial-eligible three-vessel diseased and trial-ineligible two-vessel diseased patients, CABG appeared to have at least a slight advantage with no treatment difference otherwise. In conclusion, our analyses suggest that RCTs enroll highly selected populations, and our findings are generally consistent with RCTs but less pronounced than major registry findings.
Authors
Lee, H-J; Wong, JB; Jia, B; Qi, X; DeLong, ER
URI
https://scholars.duke.edu/individual/pub1450949
PMID
32643219
Source
pubmed
Published In
Stat Med
Published Date
DOI
10.1002/sim.8581

Angiographic validation of the American College of Cardiology Foundation-the Society of Thoracic Surgeons Collaboration on the Comparative Effectiveness of Revascularization Strategies study.

BACKGROUND: The goal of this study was to compare angiographic interpretation of coronary arteriograms by sites in community practice versus those made by a centralized angiographic core laboratory. METHODS AND RESULTS: The study population consisted of 2013 American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR) records with 2- and 3- vessel coronary disease from 54 sites in 2004 to 2007. The primary analysis compared Registry (NCDR)-defined 2- and 3-vessel disease versus those from an angiographic core laboratory analysis. Vessel-level kappa coefficients suggested moderate agreement between NCDR and core laboratory analysis, ranging from kappa=0.39 (95% confidence intervals, 0.32-0.45) for the left anterior descending artery to kappa=0.59 (95% confidence intervals, 0.55-0.64) for the right coronary artery. Overall, 6.3% (n=127 out of 2013) of those patients identified with multivessel disease at NCDR sites had had 0- or 1-vessel disease by core laboratory reading. There was no directional bias with regard to overcall, that is, 12.3% of cases read as 3-vessel disease by the sites were read as <3-vessel disease by the core laboratory, and 13.9% of core laboratory 3-vessel cases were read as <3-vessel by the sites. For a subset of patients with left main coronary disease, registry overcall was not linked to increased rates of mortality or myocardial infarction. CONCLUSIONS: There was only modest agreement between angiographic readings in clinical practice and those from an independent core laboratory. Further study will be needed because the implications for patient management are uncertain.
Authors
Chakrabarti, AK; Grau-Sepulveda, MV; O'Brien, S; Abueg, C; Ponirakis, A; Delong, E; Peterson, E; Klein, LW; Garratt, KN; Weintraub, WS; Gibson, CM
MLA Citation
Chakrabarti, Anjan K., et al. “Angiographic validation of the American College of Cardiology Foundation-the Society of Thoracic Surgeons Collaboration on the Comparative Effectiveness of Revascularization Strategies study.Circ Cardiovasc Interv, vol. 7, no. 1, Feb. 2014, pp. 11–18. Pubmed, doi:10.1161/CIRCINTERVENTIONS.113.000679.
URI
https://scholars.duke.edu/individual/pub1006727
PMID
24496239
Source
pubmed
Published In
Circulation. Cardiovascular Interventions
Volume
7
Published Date
Start Page
11
End Page
18
DOI
10.1161/CIRCINTERVENTIONS.113.000679

Contemporary mortality risk prediction for percutaneous coronary intervention: results from 588,398 procedures in the National Cardiovascular Data Registry.

OBJECTIVES: We sought to create contemporary models for predicting mortality risk following percutaneous coronary intervention (PCI). BACKGROUND: There is a need to identify PCI risk factors and accurately quantify procedural risks to facilitate comparative effectiveness research, provider comparisons, and informed patient decision making. METHODS: Data from 181,775 procedures performed from January 2004 to March 2006 were used to develop risk models based on pre-procedural and/or angiographic factors using logistic regression. These models were independently evaluated in 2 validation cohorts: contemporary (n = 121,183, January 2004 to March 2006) and prospective (n = 285,440, March 2006 to March 2007). RESULTS: Overall, PCI in-hospital mortality was 1.27%, ranging from 0.65% in elective PCI to 4.81% in ST-segment elevation myocardial infarction patients. Multiple pre-procedural clinical factors were significantly associated with in-hospital mortality. Angiographic variables provided only modest incremental information to pre-procedural risk assessments. The overall National Cardiovascular Data Registry (NCDR) model, as well as a simplified NCDR risk score (based on 8 key pre-procedure factors), had excellent discrimination (c-index: 0.93 and 0.91, respectively). Discrimination and calibration of both risk tools were retained among specific patient subgroups, in the validation samples, and when used to estimate 30-day mortality rates among Medicare patients. CONCLUSIONS: Risks for early mortality following PCI can be accurately predicted in contemporary practice. Incorporation of such risk tools should facilitate research, clinical decisions, and policy applications.
Authors
Peterson, ED; Dai, D; DeLong, ER; Brennan, JM; Singh, M; Rao, SV; Shaw, RE; Roe, MT; Ho, KKL; Klein, LW; Krone, RJ; Weintraub, WS; Brindis, RG; Rumsfeld, JS; Spertus, JA; NCDR Registry Participants,
MLA Citation
Peterson, Eric D., et al. “Contemporary mortality risk prediction for percutaneous coronary intervention: results from 588,398 procedures in the National Cardiovascular Data Registry.J Am Coll Cardiol, vol. 55, no. 18, May 2010, pp. 1923–32. Pubmed, doi:10.1016/j.jacc.2010.02.005.
URI
https://scholars.duke.edu/individual/pub743052
PMID
20430263
Source
pubmed
Published In
J Am Coll Cardiol
Volume
55
Published Date
Start Page
1923
End Page
1932
DOI
10.1016/j.jacc.2010.02.005

Factors Associated with Longer Delays to Hospital Presentation for Patients with Non-ST-Elevation Myocardial Infarction

Authors
Ting, HH; Chen, AY; Roe, MT; Chan, PS; Spertus, JA; Nallamothu, BK; Sullivan, MD; DeLong, ER; Krumholz, HM; Peterson, ED
MLA Citation
Ting, Henry H., et al. “Factors Associated with Longer Delays to Hospital Presentation for Patients with Non-ST-Elevation Myocardial Infarction.” Circulation, vol. 118, no. 18, LIPPINCOTT WILLIAMS & WILKINS, Oct. 2008, pp. S1164–S1164.
URI
https://scholars.duke.edu/individual/pub874513
Source
wos
Published In
Circulation
Volume
118
Published Date
Start Page
S1164
End Page
S1164

Influence of provider volume and case-mix on risk-adjusted performance rankings

Authors
O'Brien, SM; DeLong, ER; Edwards, FH; Peterson, ED
MLA Citation
O’Brien, Sean M., et al. “Influence of provider volume and case-mix on risk-adjusted performance rankings.” Circulation, vol. 113, no. 21, LIPPINCOTT WILLIAMS & WILKINS, 2006, pp. E792–E792.
URI
https://scholars.duke.edu/individual/pub874514
Source
wos
Published In
Circulation
Volume
113
Published Date
Start Page
E792
End Page
E792

Research Areas:

Case-Control Studies
Confounding Factors (Epidemiology)
Coronary Artery Bypass
Coronary Artery Disease
Data Interpretation, Statistical
Decision Making
Diagnostic Errors
Follow-Up Studies
Health Policy
Health Services Research
Heart Diseases
Laboratories
Linear Models
Logistic Models
Multivariate Analysis
Outcome Assessment (Health Care)
Predictive Value of Tests
Prospective Studies
ROC Curve
Registries
Regression Analysis
Research Design
Risk Assessment
Rural Population
Survival Analysis