Turning Subtypes into Disease Axes to Improve Prediction of COPD Progression

Junxiang Chen; Michael H Cho; Edwin K Silverman; John E Hokanson; Greg Kinney; James D Crapo; Stephen I Rennard; Jennifer Dy; Peter J Castaldi; COPDGene investigators

doi:10.1136/thoraxjnl-2018-213005

. Author manuscript; available in PMC: 2020 Sep 18.

Published in final edited form as: Thorax. 2019 Jun 12;74(9):906–909. doi: 10.1136/thoraxjnl-2018-213005

Turning Subtypes into Disease Axes to Improve Prediction of COPD Progression

Junxiang Chen ¹, Michael H Cho ^2,³, Edwin K Silverman ^2,³, John E Hokanson ⁴, Greg Kinney ⁴, James D Crapo ⁵, Stephen I Rennard ^6,⁷, Jennifer Dy ¹, Peter J Castaldi ^2,⁸; COPDGene investigators

¹Department of Electrical and Computer Engineering, Northeastern University, Boston, MA;

²Channing Division of Network Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA;

³Pulmonary and Critical Care Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA;

⁴Dept of Epidemiology, University of Colorado, Denver, Aurora, CO;

⁵Pulmonary Medicine, National Jewish Health, Denver, CO;

⁶Pulmonary and Critical Care Medicine, University of Nebraska Medical Center, Omaha, Nebraska;

⁷Early Clinical Development, IMED Biotech Unit, AstraZeneca. Cambridge, UK;

⁸General Medicine and Primary Care, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 02115, USA.

Author contributions

Study concept and design: Chen, Kinney, Hokanson, Castaldi

Acquisition, analysis, or interpretation of data: Chen, Cho, Silverman, Crapo, Rennard, Dy, Castaldi

Drafting of the manuscript: Chen, Castaldi

Critical revision of the manuscript for important intellectual content: All authors

Statistical analysis: Chen, Dy, Castaldi

Obtained funding: Silverman, Crapo, Castaldi

COPDGene® Investigators – Core Units

Administrative Center: James D. Crapo, MD (PI); Edwin K. Silverman, MD, PhD (PI); Barry J. Make, MD; Elizabeth A. Regan, MD, PhD

Genetic Analysis Center: Terri Beaty, PhD; Ferdouse Begum, PhD; Peter J. Castaldi, MD, MSc; Michael Cho, MD; Dawn L. DeMeo, MD, MPH; Adel R. Boueiz, MD; Marilyn G. Foreman, MD, MS; Eitan Halper-Stromberg; Lystra P. Hayden, MD, MMSc; Craig P. Hersh, MD, MPH; Jacqueline Hetmanski, MS, MPH; Brian D. Hobbs, MD; John E. Hokanson, MPH, PhD; Nan Laird, PhD; Christoph Lange, PhD; Sharon M. Lutz, PhD; Merry-Lynn McDonald, PhD; Margaret M. Parker, PhD; Dandi Qiao, PhD; Elizabeth A. Regan, MD, PhD; Edwin K. Silverman, MD, PhD; Emily S. Wan, MD; Sungho Won, Ph.D.; Phuwanat Sakornsakolpat, M.D.; Dmitry Prokopenko, Ph.D.

Imaging Center: Mustafa Al Qaisi, MD; Harvey O. Coxson, PhD; Teresa Gray; MeiLan K. Han, MD, MS; Eric A. Hoffman, PhD; Stephen Humphries, PhD; Francine L. Jacobson, MD, MPH; Philip F. Judy, PhD; Ella A. Kazerooni, MD; Alex Kluiber; David A. Lynch, MB; John D. Newell, Jr., MD; Elizabeth A. Regan, MD, PhD; James C. Ross, PhD; Raul San Jose Estepar, PhD; Joyce Schroeder, MD; Jered Sieren; Douglas Stinson; Berend C. Stoel, PhD; Juerg Tschirren, PhD; Edwin Van Beek, MD, PhD; Bram van Ginneken, PhD; Eva van Rikxoort, PhD; George Washko, MD; Carla G. Wilson, MS;

PFT QA Center, Salt Lake City, UT: Robert Jensen, PhD

Data Coordinating Center and Biostatistics, National Jewish Health, Denver, CO: Douglas Everett, PhD; Jim Crooks, PhD; Camille Moore, PhD; Matt Strand, PhD; Carla G. Wilson, MS

Epidemiology Core, University of Colorado Anschutz Medical Campus, Aurora, CO: John E. Hokanson, MPH, PhD; John Hughes, PhD; Gregory Kinney, MPH, PhD; Sharon M. Lutz, PhD; Katherine Pratte, MSPH; Kendra A. Young, PhD

Mortality Adjudication Core: Surya Bhatt, MD; Jessica Bon, MD; MeiLan K. Han, MD, MS; Barry Make, MD; Carlos Martinez, MD, MS; Susan Murray, ScD; Elizabeth Regan, MD; Xavier Soler, MD; Carla G. Wilson, MS

Biomarker Core: Russell P. Bowler, MD, PhD; Katerina Kechris, PhD; Farnoush Banaei-Kashani, Ph.D

COPDGene® Investigators – Clinical Centers

Ann Arbor VA: Jeffrey L. Curtis, MD; Carlos H. Martinez, MD, MPH; Perry G. Pernicano, MD

Baylor College of Medicine, Houston, TX: Nicola Hanania, MD, MS; Philip Alapat, MD; Mustafa Atik, MD; Venkata Bandi, MD; Aladin Boriek, PhD; Kalpatha Guntupalli, MD; Elizabeth Guy, MD; Arun Nachiappan, MD; Amit Parulekar, MD;

Brigham and Women’s Hospital, Boston, MA: Dawn L. DeMeo, MD, MPH; Craig Hersh, MD, MPH; Francine L. Jacobson, MD, MPH; George Washko, MD

Columbia University, New York, NY: R. Graham Barr, MD, DrPH; John Austin, MD; Belinda D’Souza, MD; Gregory D.N. Pearson, MD; Anna Rozenshtein, MD, MPH, FACR; Byron Thomashow, MD

Duke University Medical Center, Durham, NC: Neil MacIntyre, Jr., MD; H. Page McAdams, MD; Lacey Washington, MD

HealthPartners Research Institute, Minneapolis, MN: Charlene McEvoy, MD, MPH; Joseph Tashjian, MD

Johns Hopkins University, Baltimore, MD: Robert Wise, MD; Robert Brown, MD; Nadia N. Hansel, MD, MPH; Karen Horton, MD; Allison Lambert, MD, MHS; Nirupama Putcha, MD, MHS

Los Angeles Biomedical Research Institute at Harbor UCLA Medical Center, Torrance, CA: Richard Casaburi, PhD, MD; Alessandra Adami, PhD; Matthew Budoff, MD; Hans Fischer, MD; Janos Porszasz, MD, PhD; Harry Rossiter, PhD; William Stringer, MD

Michael E. DeBakey VAMC, Houston, TX: Amir Sharafkhaneh, MD, PhD; Charlie Lan, DO

Minneapolis VA: Christine Wendt, MD; Brian Bell, MD

Morehouse School of Medicine, Atlanta, GA: Marilyn G. Foreman, MD, MS; Eugene Berkowitz, MD, PhD; Gloria Westney, MD, MS

National Jewish Health, Denver, CO: Russell Bowler, MD, PhD; David A. Lynch, MB

Reliant Medical Group, Worcester, MA: Richard Rosiello, MD; David Pace, MD

Temple University, Philadelphia, PA: Gerard Criner, MD; David Ciccolella, MD; Francis Cordova, MD; Chandra Dass, MD; Gilbert D’Alonzo, DO; Parag Desai, MD; Michael Jacobs, PharmD; Steven Kelsen, MD, PhD; Victor Kim, MD; A. James Mamary, MD; Nathaniel Marchetti, DO; Aditi Satti, MD; Kartik Shenoy, MD; Robert M. Steiner, MD; Alex Swift, MD; Irene Swift, MD; Maria Elena Vega-Sanchez, MD

University of Alabama, Birmingham, AL: Mark Dransfield, MD; William Bailey, MD; Surya Bhatt, MD; Anand Iyer, MD; Hrudaya Nath, MD; J. Michael Wells, MD

University of California, San Diego, CA: Joe Ramsdell, MD; Paul Friedman, MD; Xavier Soler, MD, PhD; Andrew Yen, MD

University of Iowa, Iowa City, IA: Alejandro P. Comellas, MD; Karin F. Hoth, PhD; John Newell, Jr., MD; Brad Thompson, MD

University of Michigan, Ann Arbor, MI: MeiLan K. Han, MD, MS; Ella Kazerooni, MD; Carlos H. Martinez, MD, MPH

University of Minnesota, Minneapolis, MN: Joanne Billings, MD; Abbie Begnaud, MD; Tadashi Allen, MD

University of Pittsburgh, Pittsburgh, PA: Frank Sciurba, MD; Jessica Bon, MD; Divay Chandra, MD, MSc; Carl Fuhrman, MD; Joel Weissfeld, MD, MPH

University of Texas Health Science Center at San Antonio, San Antonio, TX: Antonio Anzueto, MD; Sandra Adams, MD; Diego Maselli-Caceres, MD; Mario E. Ruiz, MD

^✉

Corresponding Author: Peter J. Castaldi, MD, MSc, Channing Division of Network Medicine, Brigham and Women’s Hospital, 181 Longwood Avenue, Boston, MA, 02115, peter.castaldi@channing.harvard.edu

PMCID: PMC7500804 NIHMSID: NIHMS1605222 PMID: 31189730

Abstract

COPD is an umbrella definition encompassing multiple disease processes. COPD heterogeneity has been described as distinct subgroups of individuals (subtypes), or as continuous measures of COPD variability (disease axes). There is little consensus on whether subtypes or disease axes are preferred, and the relative value of disease axes and subtypes for predicting COPD progression is unknown. Using a propensity score approach to learn disease axes from pairs of subtypes, we demonstrate that these disease axes predict prospective FEV₁ decline and emphysema progression more accurately than the subtype pairs from which they were derived.

Introduction

The heterogeneity of chronic obstructive pulmonary disease (COPD) obscures our understanding of its natural history and molecular mechanisms. COPD heterogeneity is often represented as distinct subgroups of subjects (subtypes), but it can also be represented as continuous axes of variability, i.e. disease axes[1]. A multi-cohort study demonstrated that subtypes identified by clustering were not reproducible across cohorts, whereas disease axes from the same cohorts were more consistent[2]. There is currently no consensus on the best approach to characterize COPD heterogeneity.

We define a subtype as a single subgroup of subjects and a COPD disease axis as any continuous representation of COPD heterogeneity. We describe a method, similar in concept to propensity scores[3], where a pair of COPD subtypes can be used to define a single disease axis by using the subtype pair as the response in a logistic regression model that predicts the likelihood of subtype membership. These predictions constitute a subtype-defined disease axis. For example, in the case of chronic bronchitis, the chronic bronchitis subtype is a binary yes/no classification based on patient symptoms. Conversely, the chronic bronchitis disease axis is a continuous measure derived from a predictive model that describes the propensity of each subject to have chronic bronchitis. Using longitudinal data from the Genetic Epidemiology of COPD (COPDGene) Study, we demonstrate that subtype-defined disease axes provide better prediction of prospective COPD progression than the original subtype pairs from which they were derived.

Methods

Subjects in COPDGene with complete five-year follow-up data were analyzed (n=4,726). Four general subtype classes were selected for study: Chronic Bronchitis (CB) per the ATS-DLD definition[4], the pink puffer (PP)/blue bloater (BB) subtype[5], frequent exacerbators (>=2 COPD exacerbations over the previous 12 months)[6], and upper/lower lobe emphysema predominant subjects with a log U/L ratio >1.5 for upper lobe or <−1.5 for lower lobe predominance. We refer to a subtype pair as two subtypes that are conceptually related and therefore used to construct a disease axis. For example, the CB subtype class yields a single subtype pair (chronic bronchitis present versus absent), whereas the PP/BB subtype class yields two pairs (PP/neither and BB/neither).

For each subtype pair, we used weighted logistic regression to identify a linear combination of predictors that provide optimal classification for that pair. The beta-coefficients of this regression were used to calculate the disease axis value for each analyzed subject. This software is available at https://github.com/Chen-Jxiang/SODA. We selected the baseline values of 27 variables to serve as the predictors in the regression models (see Supplemental Materials for variables used). Disease axes were generated only from visit 1 data.

For the analyses of COPD progression, separate regression models were used to relate subtypes or disease axis scores to either five-year change in FEV₁ % of predicted or change in emphysema. To formally test for whether disease axes provide incremental improvement in prediction beyond that provided by a subtype pair, we constructed nested regression models in which a disease axis was added to a base model containing the original subtype pair. Additional information is included in the Supplement.

Results

A conceptual overview of our approach is shown in Figure 1. Subtype definitions and characteristics of the subjects are shown in ST 1 and 2. One disease axis was identified for each subtype pair, resulting in a total of six disease axes (one each for frequent/non-frequent exacerbators and presence/absence of chronic bronchitis, and two each for the Pink Puffer/Blue Bloater and Upper/Lower Emphysema subtypes). To determine how well the disease axes could correctly classify their original subtypes, we examined the discrimination performance which was excellent for the pink puffer/blue bloater and upper/lower emphysema subgroups (AUC > 0.98), and reasonable for the frequent exacerbator (AUC = 0.79) and chronic bronchitis subgroups (AUC = 0.67). The predictors and beta coefficients from these models are shown in ST 3 and 4.

We then studied how well each subtype pair and its respective disease axis could predict two measures of COPD progression, change in FEV₁ and quantitative CT emphysema progression. For both outcomes, we observed that regression models containing disease axes as predictors explained a greater proportion of the variance of COPD progression than similar models containing the subtype pair, with particularly marked improvement noted for emphysema progression (Table 1). To formally test for significant improvement in prediction from disease axes, for each subtype-containing model we added the corresponding disease axes and compared the two models (Table 2).

Table 1.

Regression models using either subtypes or disease axes to predict change in emphysema or change in FEV₁ % of predicted.

Progression Measure	Subtype Class	Subtype Pair	Subtype Models			Disease Axis Models
Progression Measure	Subtype Class	Subtype Pair	Beta (SE)	P	% Variance Explained	Beta (SE)	P	% Variance Explained
Δ emphysema (Perc15)	Chronic Bronchitis	Chronic Bronchitis (No versus Yes)	−1.1 (0.5)	0.05	7.5	−8.0 (0.5)	<0.001	12.8
	Frequent Exacerbator	Frequent Exacerbators (No versus Yes)	0.3 (0.8)	0.68	7.5	−2.1(0.4)	<0.001	8.0
	Pink Puffer/Blue Bloater	PP/BB (Neither versus PP)	−2.7 (2.0)	0.17	7.6	−1.2 (0.2)	<0.001	8.5
	Pink Puffer/Blue Bloater	PP/BB (Neither versus BB)	−7.8 (3.2)	0.01	7.6	−0.3 (0.2)	0.07	8.5
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus LLE)	−6.9 (2.1)	0.001	8.3	6.7 (0.4)	<0.001	12.8
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus ULE)	−4.7 (0.8)	<0.001	8.3	−5.1 (0.3)	<0.001	12.8
Δ FEV1 % of predicted	Chronic Bronchitis	Chronic Bronchitis (No versus Yes)	−1.9 (0.4)	<0.001	6.0	−2.5 (0.4)	<0.001	6.4
	Frequent Exacerbator	Frequent Exacerbators (No versus Yes)	−1.9 (0.6)	0.002	5.8	−2.2 (0.3)	<0.001	6.5
	Pink Puffer/Blue Bloater	PP/BB (Neither versus PP)	−3.5 (1.5)	0.03	5.7	−0.7 (0.1)	<0.001	6.4
	Pink Puffer/Blue Bloater	PP/BB (Neither versus BB)	−3.2 (2.5)	0.19	5.7	−0.6 (0.1)	<0.001	6.4
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus LLE)	−0.7 (1.7)	0.68	5.7	−0.6 (0.3)	0.09	5.8
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus ULE)	−2.3 (0.6)	<0.001	5.7	0.3 (0.2)	0.19	5.8

Open in a new tab

For each outcome and subtype class, two regression models were constructed including either the subtype(s) or disease axes as predictors. The COPD progression outcomes were either change in FEV1 or change in emphysema between visit 1 and 2. Values in the subtype model columns are from models that include categorical subtype assignment as a predictor, and values in the disease axis columns are from models that include the corresponding disease axes as a predictor. All models also include baseline FEV1 % predicted, baseline emphysema, and current smoking status at visit 1 and 2. PP/BB and Upper/Lower emphysema subtypes have three categories and thus 2 contrasts are included in the same model for each subtype group, respectively.

Perc15 – 15^th percentile of the lung density histogram.

PP – Pink Puffer; BB – Blue Bloater; LLE = lower lobe emphysema predominant subtype; ULE – upper lobe emphysema predominant subtype.

Table 2.

Regression models using both subtypes and disease axes to predict change in emphysema or change in FEV1.

Progression Measure	Subtype Class	Subtype Pair	Subtypes		Disease Axes		% Variance Explained
Progression Measure	Subtype Class	Subtype Pair	Beta (SE)	P	Beta (SE)	P	% Variance Explained
Δ emphysema (Perc15)	Chronic Bronchitis	Chronic Bronchitis (No versus Yes)	−0.1 (0.5)	0.78	−8.0 (0.5)	<0.001	12.8
	Frequent Exacerbator	Frequent Exacerbators (No versus Yes)	0.9 (0.8)	0.27	−2.2(0.4)	<0.001	8.0
	Pink Puffer/Blue Bloater	PP/BB (Neither versus PP)	−0.2 (2.0)	0.92	−1.2 (0.2)	<0.001	8.6
	Pink Puffer/Blue Bloater	PP/BB (Neither versus BB)	−8.2 (3.2)	0.01	−0.2 (0.2)	0.21	8.6
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus LLE)	−1.1 (2.2)	0.60	−3.0 (0.2)	<0.001	13.9
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus ULE)	−0.20 (1.0)	0.84	−2.6 (0.2)	<0.001	13.9
Δ FEV1 % of predicted	Chronic Bronchitis	Chronic Bronchitis (No versus Yes)	−1.7 (0.4)	<0.001	−2.3 (0.4)	<0.001	6.8
	Frequent Exacerbator	Frequent Exacerbators (No versus Yes)	−1.3 (0.6)	0.02	−2.1 (0.3)	<0.001	6.6
	Pink Puffer/Blue Bloater	PP/BB (Neither versus PP)	−2.3 (1.6)	0.20	−0.7 (0.1)	<0.001	6.4
	Pink Puffer/Blue Bloater	PP/BB (Neither versus BB)	−1.4 (2.5)	0.59	−0.6 (0.1)	<0.001	6.4
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus LLE)	−0.8 (1.7)	0.63	0.2 (0.1)	0.20	5.9
	Upper/Lower Emphysema	Upper/Lower Emphysema (Neither versus ULE)	−2.8 (0.8)	<0.001	0.2 (0.1)	0.13	5.9

Open in a new tab

For each outcome and subtype class, one regression model was constructed for each subtype class. This model contained the relevant categorical subtypes as well as the corresponding disease axes. The COPD progression outcomes were either change in FEV1 or change in emphysema between visit 1 and 2. All models also include baseline FEV1 % predicted, baseline emphysema, and current smoking status at visit 1 and 2. PP/BB and Upper/Lower emphysema subtypes have three categories and thus 2 contrasts are included in the same model for each subtype group, respectively.

Perc15 – 15th percentile of the lung density histogram.

PP – Pink Puffer; BB – Blue Bloater; LLE = lower lobe emphysema predominant subtype; ULE – upper lobe emphysema predominant subtype.

We also examined how well baseline DA values predicted the consistency of subtype assignment over time, which is an important issue for the CB and frequent exacerbator subtypes. We classified subjects as persistent or intermittent members of these subtypes according to their status at both COPDGene Study visits, and we observed that persistent subjects had higher DA values than intermittent subjects (Supplemental Figures 1 and 2, p<0.001 for CB and p=0.007 for frequent exacerbators).

Discussion

Previous work has shown that COPD variability typically occurs along a continuum[2]. Thus, while subtypes may have intuitive appeal, disease axes are more accurate. The method presented here turns subtypes into disease axes, providing more accurate representations of COPD heterogeneity that represent a continuum defined by two COPD subtypes. These disease axes were more predictive of COPD progression than the subtypes from which they were derived; because 1) disease axes “expand” subtype information to all subjects in a dataset, and 2) disease axes extract subtype-related information from a large number of input variables and thus contain more COPD-related information than subtypes alone.

Since this method uses pre-defined subtypes to guide data-driven analysis, the strengths of this approach are the interpretability of the disease axes and the improved prediction of disease progression. However, when the sole goal is prediction, purely data-driven methods may yield superior performance. These disease axes were generated in a single cohort, so independent assessment of their generalizability is needed. These results provide proof of concept that subtype-defined disease axes provide more powerful prediction of COPD progression. In the future, it would be useful to define disease axes that can be produced from readily available variables, which would allow disease axes to be generated in a larger set of COPD studies.

In summary, relative to subtypes, disease axes provide more accurate clinical predictions, and in the future disease axes may improve our clinical characterization of COPD and enable more powerful biological discovery.

Supplementary Material

Supplement

NIHMS1605222-supplement-Supplement.pdf^{(548.7KB, pdf)}

Acknowledgements:

COPDGene Phase 3

Grant Support and Disclaimer

The project described was supported by Award Number U01 HL089897 and Award Number U01 HL089856 from the National Heart, Lung, and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health.

COPD Foundation Funding

The COPDGene® project is also supported by the COPD Foundation through contributions made to an Industry Advisory Board comprised of AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, Novartis, Pfizer, Siemens and Sunovion.

Disclosures: PJC reports consulting fees and grant support from GSK and Novartis outside the submitted work. MHC has received grant support from GSK. In the past three years, Edwin K. Silverman received honoraria from Novartis for Continuing Medical Education Seminars and grant and travel support from GlaxoSmithKline.

References

1.Kinney GL, Santorico SA, Young KA, et al. Identification of Chronic Obstructive Pulmonary Disease Axes That Predict All-Cause Mortality: The COPDGene Study. American Journal of Epidemiology 2018;187:2109–16. doi: 10.1093/aje/kwy087 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Castaldi PJ, Benet M, Petersen H, et al. Do COPD subtypes really exist? COPD heterogeneity and clustering in 10 independent cohorts. Thorax 2017;72:998–1006. doi: 10.1136/thoraxjnl-2016-209846 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.ROSENBAUM PR RUBIN DB. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:41–55. doi: 10.1093/biomet/70.1.41 [DOI] [Google Scholar]
4.Ferris BG. Epidemiology Standardization Project (American Thoracic Society). Am Rev Respir Dis 1978;118:1–120. [PubMed] [Google Scholar]
5.Hersh CP, Make BJ, Lynch DA, et al. Non-emphysematous chronic obstructive pulmonary disease is associated with diabetes mellitus. BMC Pulm Med 2014;14:164. doi: 10.1186/1471-2466-14-164 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Hurst JR, Vestbo J, Anzueto A, et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010;363:1128–38. doi: 10.1056/NEJMoa0909883 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement

NIHMS1605222-supplement-Supplement.pdf^{(548.7KB, pdf)}

[R1] 1.Kinney GL, Santorico SA, Young KA, et al. Identification of Chronic Obstructive Pulmonary Disease Axes That Predict All-Cause Mortality: The COPDGene Study. American Journal of Epidemiology 2018;187:2109–16. doi: 10.1093/aje/kwy087 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Castaldi PJ, Benet M, Petersen H, et al. Do COPD subtypes really exist? COPD heterogeneity and clustering in 10 independent cohorts. Thorax 2017;72:998–1006. doi: 10.1136/thoraxjnl-2016-209846 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R3] 3.ROSENBAUM PR RUBIN DB. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:41–55. doi: 10.1093/biomet/70.1.41 [DOI] [Google Scholar]

[R4] 4.Ferris BG. Epidemiology Standardization Project (American Thoracic Society). Am Rev Respir Dis 1978;118:1–120. [PubMed] [Google Scholar]

[R5] 5.Hersh CP, Make BJ, Lynch DA, et al. Non-emphysematous chronic obstructive pulmonary disease is associated with diabetes mellitus. BMC Pulm Med 2014;14:164. doi: 10.1186/1471-2466-14-164 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Hurst JR, Vestbo J, Anzueto A, et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010;363:1128–38. doi: 10.1056/NEJMoa0909883 [DOI] [PubMed] [Google Scholar]

PERMALINK

Turning Subtypes into Disease Axes to Improve Prediction of COPD Progression

Junxiang Chen

Michael H Cho

Edwin K Silverman

John E Hokanson

Greg Kinney

James D Crapo

Stephen I Rennard

Jennifer Dy

Peter J Castaldi

Abstract

Introduction

Methods

Results

Figure 1. Overview of Subtype-Oriented Disease Axis Approach.

Table 1.

Table 2.

Discussion

Supplementary Material

Acknowledgements:

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Cite

Add to Collections

PERMALINK

Turning Subtypes into Disease Axes to Improve Prediction of COPD Progression

Junxiang Chen

Michael H Cho

Edwin K Silverman

John E Hokanson

Greg Kinney

James D Crapo

Stephen I Rennard

Jennifer Dy

Peter J Castaldi

Abstract

Introduction

Methods

Results

Figure 1. Overview of Subtype-Oriented Disease Axis Approach.

Table 1.

Table 2.

Discussion

Supplementary Material

Acknowledgements:

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases