0
Original Research: COPD |

Short- and Medium-term Prognosis in Patients Hospitalized for COPD ExacerbationPrognosis in Patients Hospitalized for COPD: The CODEX Index FREE TO VIEW

Pedro Almagro, MD; Joan B. Soriano, MD; Francisco J. Cabrera, MD; Ramon Boixeda, MD; M. Belen Alonso-Ortiz, MD; Bienvenido Barreiro, MD; Jesus Diez-Manglano, MD; Cristina Murio, MD; Josep L. Heredia, MD; the Working Group on COPD, Spanish Society of Internal Medicine*
Author and Funding Information

From the Internal Medicine Service (Dr Almagro), and the Respiratory Service (Drs Barreiro and Heredia), Hospital Universitario Mutua De Terrassa, Universidad de Barcelona, Barcelona; Programa de Epidemiología e Investigación Clínica (Dr Soriano), Fundación Caubet-Cimera, Centro Internacional de Medicina Respiratoria Avanzada, Baleares; the Internal Medicine Service (Dr Cabrera), Hospital General Universitario Gregorio Marañón, Madrid; the Internal Medicine Service (Dr Boixeda), Hospital de Mataró, Barcelona; the Medical Department (Dr Murio), Laboratorios Chiesi, Barcelona; the Internal Medicine Service (Dr Alonso-Ortiz), Hospital Juan Negrín, Gran Canaria; and the Internal Medicine Service (Dr Diez-Manglano), Hospital Royo Villanova, Zaragoza, Spain.

Correspondence to: Pedro Almagro, MD, Internal Medicine Service, Hospital Universitario Mutua De Terrassa, Universidad de Barcelona, Plaza Dr Robert No. 5, 08221, Terrassa, Barcelona, Spain; e-mail: 19908pam@comb.cat


For editorial comment see page 934

Dr Soriano is currently at FISIB-IdISPA Hospital Universitari Son Espases (Palma de Mallorca, Spain).

*A complete list of study investigators is provided in e-Appendix 1.

Funding/Support: This work was supported by Chiesi España.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2014;145(5):972-980. doi:10.1378/chest.13-1328
Text Size: A A A
Published online

Background:  No valid tools exist for evaluating the prognosis in the short and medium term after hospital discharge of patients with COPD. Our hypothesis was that a new index based on the CODEX (comorbidity, obstruction, dyspnea, and previous severe exacerbations) index can accurately predict mortality, hospital readmission, and their combination for the period from 3 months to 1 year after discharge in patients hospitalized for COPD.

Methods:  A multicenter study of patients hospitalized for COPD exacerbations was used to develop the CODEX index, and a different patient cohort was used for validation. Comorbidity was measured using the age-adjusted Charlson index, whereas dyspnea, obstruction, and severe exacerbations were calculated according to BODEX (BMI, airflow obstruction, dyspnea, and previous severe exacerbations) thresholds. Information about mortality and readmissions for COPD or other causes was collected at 3 and 12 months after hospital discharge.

Results:  Two sets of 606 and 377 patients were included in the development and validation cohorts, respectively. The CODEX index was associated with mortality at 3 months (P < .0001; hazard ratio [HR], 1.5; 95% CI, 1.2-1.8) and 1 year (P < .0001; HR, 1.3; 95% CI, 1.2-1.5), hospital readmissions in the same periods, and their combination (all P < .0001). All CODEX C statistics were superior to those of the BODEX, DOSE (dyspnea, airflow obstruction, smoking status, and exacerbation frequency), and updated ADO (age, dyspnea, and airflow obstruction) indexes.

Conclusions:  The CODEX index was a useful predictor of survival and readmission at both 3 months and 1 year after hospital discharge for a COPD exacerbation, with a prognostic capacity superior to other previously published indexes.

Figures in this Article

COPD is one of the most prevalent diseases in adults and is associated with high morbidity and mortality worldwide.1 Although many variables are associated with increased mortality in patients with COPD, the most frequently recognized are postbronchodilator FEV1, dyspnea, and exacerbations.2,3 Of interest, all these variables have been incorporated into the newly released GOLD (Global Initiative for Chronic Obstructive Lung Disease) recommendations for grading COPD severity.4 The use of multidimensional indexes, such as the BODE (BMI, airflow obstruction, dyspnea, and exercise capacity), BODEX (BMI, airflow obstruction, dyspnea, and previous severe exacerbations), ADO (age, dyspnea, and airflow obstruction), DOSE (dyspnea, airflow obstruction, smoking status, and exacerbation frequency), and other indexes, have improved the prognostic capacity beyond individual variables.59 However, to our knowledge, none has demonstrated prognostic value in the short term after a severe acute exacerbation of COPD (AECOPD), such as after a hospital discharge, which suggests that their usefulness may vary depending on the length of the study observation and the severity of disease.10

Although studies suggest that posthospital mortality may have declined, it continues to remain too high.11 It is well known that hospitalizations due to COPD are still associated with poorer survival in the months following, and this is at least in part because many patients are older, with advanced respiratory disease and frequent comorbidities. These three characteristics are all well-recognized indicators of mortality in overall COPD populations.12,13

Interestingly, comorbidities measured with the Charlson index are related to global but not to respiratory mortality in the original BODE index cohort. However, the original BODE index cohort was recruited from among ambulatory patients without significant comorbidities.5 A more recent study performed by the same group concluded that selected comorbidities are associated with higher mortality from COPD- and non-COPD-related causes, although, again, in this work, comorbidity was not related to survival in the short or medium term.14 Conversely, two other studies performed in patients hospitalized for AECOPD related comorbidity to mortality and hospital readmission at 90 days after discharge.15,16 It is precisely in this group of patients, with higher mortality in the short-term follow-up, that prognosis reliability is most needed, although regrettably none of the current indexes has proved to be useful.

The main objective of our study was to develop and validate a new index to predict short-term mortality and readmissions in patients hospitalized for a severe acute COPD exacerbation. For this purpose, we selected five variables with previously recognized usefulness in the prognostic evaluation of these patients (comorbidity, age, obstruction, dyspnea, and exacerbations), using previously validated cutoff points whenever possible.

Data from the EPOC en los Servicios de Medicina Interna (ESMI) study were used for the development of the CODEX (comorbidity, obstruction, dyspnea, and previous severe exacerbations) index. The ESMI study is a longitudinal, observational, multicenter study conducted in 70 EDs and internal medicine services in Spain. Investigators included the first 10 consecutive patients seen for a severe COPD exacerbation during a 1-year period (October 2009-October 2010). Full methodology is available elsewhere and is summarized in e-Appendix 1.16,17 During admission, all patients were evaluated using a standardized questionnaire. Comorbidity was documented using the previously validated Charlson index, a standard scale with 15 chronic diseases graded for severity of disease, in which one point is added to the total score for each decade of life over the age of 50 years.18 Dyspnea was measured using the modified Medical Research Council (mMRC) scale. The vital status and number of hospital readmissions for COPD or other causes at 3 months and 1 year after hospital discharge were collected at outpatient visits and, when this was not possible, by reviewing medical records or through telephone calls to the patient or family. Respiratory and other causes of death were recorded by the responsible investigator. Dyspnea, obstruction measured with postbronchodilator FEV1 expressed as a percentage, and the number of hospitalizations for AECOPDs during the previous year were stratified according to previously published BODE and BODEX thresholds.5,6 Finally, the original Charlson index was stratified in tertiles (0-4 = 0; 5-7 = 1; ≥ 8 = 2). As a result, possible scores on the newly proposed CODEX index range from 0 to 10 points (Table 1).

Table Graphic Jump Location
Table 1 —Variables and Thresholds to Estimate the CODEX Index

CODEX = comorbidity, obstruction, dyspnea, and previous severe exacerbations; mMRC = modified Medical Research Council.

a 

Charlson index: one point is added to the total score for each decade of life over the age of 50 y.

b 

Severe exacerbations of COPD during the previous year (hospitalization or ED consultations).

The validation cohort analysis was performed by pooling data from three previously published cohorts of patients hospitalized at University Hospital Mutua de Terrassa, all with the same inclusion and exclusion criteria as in the ESMI study.11,16,19 Additionally, we evaluated the usefulness of the CODEX index in predicting the combined end point of mortality or readmission for any cause, both at 3 and 12 months, and compared its usefulness with the BODEX, DOSE, and updated ADO indexes to predict mortality, readmissions, and their combination in both time periods.

The ESMI study and the studies used for the validation cohort were approved by the Clinical Investigation Ethics Committee of the Hospital Mútua de Terrassa (EO/0922_0709). All patients participated voluntarily and signed written informed consent forms.

Statistical Analysis

Qualitative variables were expressed as absolute frequencies and percentages, and quantitative variables were summarized as mean and SD in the case of normally distributed, or as median otherwise. Comparisons among means were made using the Student t test or Mann-Whitney U test for continuous variables and the χ2 or Fisher exact test for categorical variables. Time-dependent variables from hospital discharge were analyzed with Cox logistic regression and Kaplan-Meier statistics. To assess the ability of the CODEX index to predict all-cause mortality or readmissions, and to make comparisons with other mortality indexes, we plotted receiver operating characteristic (ROC) curves, and their areas under the curve (AUCs) were compared with C statistics. Comparative significance was assessed with the DeLong method.20 In all comparisons, statistical significance was identified for P values < .05.

Development Cohort

Overall, 679 patients were screened, and 606 were included in the initial ESMI study. Fifteen patients were excluded for having an incomplete minimum dataset, and in another 58 cases, spirometry was unable to be performed or did not fulfill the spirometric criteria for COPD (Fig 1). No differences existed between included and excluded patients with respect to age, sex, smoking history, or dyspnea measured by the mMRC scale (data not shown). Our study included patients hospitalized for exacerbations of COPD, regardless of whether they required mechanical ventilation, either invasive or not.

Figure Jump LinkFigure 1. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) flowchart of participants and causes of exclusion. ESMI = EPOC en los Servicios de Medicina Interna.Grahic Jump Location

The mean age of the ESMI study population was 72.6 years (SD, 9.9 years), and 89.8% were men. Seventy patients (11.5%) were discharged from the ED without requiring conventional hospitalization. Patients discharged directly from the ED were younger, with a higher FEV1 and fewer comorbidities (all P < .01). For this reason, we initially performed a subanalysis that showed no differences between the CODEX index and the mortality rate at 3 months, after excluding patients discharged from the ED. The rest of the analysis was conducted with the whole ESMI cohort.

Validation Cohort

The validation cohort consisted of 377 patients hospitalized in the University Hospital Mútua de Terrassa during three different time periods (1997-1998, 1999-2000, and 2003-2004). The average age of these patients was 71.9 years (SD, 9.7 years), with a predominance of men (94%). The main characteristics of the ESMI and validation cohorts are presented in Table 2.

Table Graphic Jump Location
Table 2 —Demographic and Clinical Characteristics of the ESMI and Validation Cohorts

Data are presented as mean (SD) unless indicated otherwise. ESMI = EPOC en los Servicios de Medicina Interna; ns = not significant. See Table 1 legend for expansion of other abbreviation.

a 

Dyspnea measured with mMRC scale.

b 

No. hospitalizations for COPD during previous year.

c 

Postbronchodilator test.

The mean value for the CODEX index in hospitalized patients in the ESMI study was 4.9 (2.4), decreasing slightly to 4.7 (2.4) when patients discharged from the ED were incorporated (P = not significant), with a median of 5 points both in the ESMI study and in the validation cohort. Figure 2 shows the frequency point distribution of the CODEX index in the ESMI and validation cohorts (P = not significant).

Figure Jump LinkFigure 2. Distribution of the CODEX index in the ESMI and validation cohorts. CODEX = comorbidity, obstruction, dyspnea, and previous severe exacerbations; n.s. = not significant. See Figure 1 legend for expansion of other abbreviations.Grahic Jump Location
Mortality

Vital status at 3 months was collected for 557 case patients (91.9%) in the ESMI study, 27 of whom died within 3 months after discharge (4.8%). Mortality at 3 months was associated with age, number of hospital admissions for COPD or other causes in the previous year, greater dyspnea (mMRC score), chronic home oxygen therapy use, more functional dependence (Katz index), and Charlson index both with and without age stratification, but was not associated with BMI (P = .2) (e-Table 1). The CODEX index was statistically associated with mortality (P < .0001; hazard ratio [HR] 1.5; 95% CI, 1.2-1.8). Mortality in the validation cohort was 7%, and the CODEX index showed a similar association (P = .006; HR, 1.4; 95% CI, 1.1-1.7). The AUC was 0.73 for the ESMI cohort, 0.71 for the validation cohort, and 0.72 for the combination of the two.

Mortality at 30 days was 2.1% in the ESMI study and 1.6% in the validation cohort. Vital status was assessed at 1 year in 932 patients of the ESMI study and validation cohort (95.1%), and of these, 126 (13.5%) had died. No differences were found between patients with and without complete follow-up in terms of age, sex, length of hospitalization, dyspnea, FEV1, or score on the CODEX index (data not shown). Mortality causes were respiratory (58%), cardiovascular (16%), and neoplasm (5%); other or unknown causes accounted for 21%. The CODEX index was significantly related to mortality at 1 year in both the ESMI study (P < .0001; HR, 1.3; 95% CI, 1.2-1.5) and the validation cohort. (P < .0001; HR, 1.4; 95% CI, 1.2-1.5). The CODEX index was also a significant predictor of mortality for respiratory causes at 1 year after discharge (P < .0001; HR, 1.28; 95% CI, 1.16-1.4).

The AUC at 1 year was 0.67 for the ESMI study and 0.68 for the validation cohort. Kaplan-Meier curves for survival at 3 months and 1 year for the global population are represented graphically in Figure 3.

Figure Jump LinkFigure 3. Kaplan-Meier curves for mortality. H.R. = hazard ratio.Grahic Jump Location
Combined Variable

The CODEX index was significantly related to hospital readmission at both 3 months (5.7 [2.2] vs 3.5 [3.3], P < .0001) and 1 year (5.5 [2] vs 4.7 [2.2], P < .0001). In assessing the combined variable (death or readmission), 34.2% of patients had died or had been readmitted for COPD at 3 months, increasing to 53.8% at 1 year. In both periods, the CODEX index was significantly related to the presence of the composite variable: 3 months (6.1 [2.2] vs 4.6 [2.2]) and 1 year (5.6 [2.1] vs 4.6 [2.1]) (P < .0001 for both). Figure 4 shows the relationship between the different outcome variables and the scores obtained on the CODEX index.

Figure Jump LinkFigure 4. Scores on the CODEX index and mortality or combined variable. See Figure 2 legend for expansion of abbreviation.Grahic Jump Location
Comparison With Other Indexes

The updated ADO, DOSE, and BODEX indexes were significantly related to the probability of death at 3 months and at 1 year after hospital discharge (Table 3). However, comparison of the ROC curves (C statistics) up to 3 months after discharge showed a better predictive capacity for the CODEX index compared with the BODEX (P < .005), DOSE (P < .01), and updated ADO (P < .05) indexes. This difference remained significant at 1 year for the BODEX (P < .02) and DOSE indexes (P < .02), but not for the updated ADO index (Fig 5).

Table Graphic Jump Location
Table 3 —Prognostic Validity of Mortality (HR and AUC) at 3 Mo and 1 Y After Hospital Discharge of CODEX Index and All Other COPD Multivariate Indexes (ADO, BODEX, BODE, and DOSE) in the ESMI Study

ADO = age, dyspnea, and airflow obstruction; AUC = area under the curve; BODE = BMI, airflow obstruction, dyspnea, and exercise capacity; BODEX = BMI, airflow obstruction, dyspnea, and previous severe exacerbations; DOSE = dyspnea, airflow obstruction, smoking status, and exacerbation frequency; HR = hazard ratio. See Table 1 and 2 legends for expansion of other abbreviations.

Figure Jump LinkFigure 5. ROC curves for the CODEX, BODEX, DOSE, and updated ADO_u indexes. ROC = receiver operating characteristic. ADO_u = age, dyspnea, and airflow obstruction; BODEX = BMI, airflow obstruction, dyspnea, and previous severe exacerbations; DOSE = dyspnea, airflow obstruction, smoking status, and exacerbation frequency. See Figure 2 legend for expansion of other abbreviations.Grahic Jump Location

The possibility of readmission at 3 months reached statistical significance for the BODEX (4.8 [2.5] vs 4.2 [2.1], P = .008) and DOSE (4.3 [1.7] vs 3.6 [1.8], P > .01) indexes, but not for the updated ADO index. These results are similar to those observed during the year after discharge: BODEX index (P = .002), DOSE index (P < .001), and updated ADO index (P = .1).

With respect to the combined variable (death and/or rehospitalization), all indexes reached statistical significance at 3 months and at 1 year. New ROC curves showed a better predictive capacity at 90 days from hospital discharge for the CODEX index compared with the BODEX (P < .005) and ADO (P < .0001) indexes, although no statistical difference was observed with the DOSE index. The 1-year analysis showed similar results for the comparison of the AUC in the ROC curves between the CODEX index and the ADO index (P < .0001), the BODEX index (P < .001), and the DOSE index (P < .05). Full comparison analysis among the several indexes and outcomes is detailed in e-Tables 2 to 5.

Our study, performed in patients hospitalized for AECOPD, has three main findings. First, we demonstrated that a simple index can be useful in predicting survival in both the short and medium term after hospital discharge in these patients. Second, we reported the usefulness of this CODEX index in evaluating the risk of readmission, as well as the composite end point (readmission and/or mortality). Third, we had the opportunity to compare the prognostic validity with other previously validated indexes in the general population with COPD, and to compare their efficacy with the new index in patients with more severe disease, such as those requiring hospital admission for AECOPD. To our knowledge, no previous multicomponent, simple index has been shown to be useful in predicting mortality in a global population of patients admitted for AECOPD in periods shorter than 1 year after hospital discharge.10

Few studies have examined the short-term prognosis of patients hospitalized for COPD, and these were performed in selected populations, such as patients admitted to ICUs or those with hypercapnia at admission. In addition, the proposed prognostic indexes are excessively complex and are not applicable to routine clinical practice.2123 More recently, a new index was developed for predicting in-hospital mortality in patients with AECOPD, and it uses a different approach from that of our study. Of note, in this study, patients with comorbidities who were expected to have survival limited to < 12 months were excluded.24

The most widely recognized variables associated with decreased survival in COPD populations are lung function, measured with FEV1 after bronchodilator test; dyspnea; and exacerbations. The first two are the basis of multicomponent indexes. It is well established that FEV1 is a good predictor of mortality in the general population, and historically it has been considered the most important variable for evaluating severity in COPD. Furthermore, the decline of FEV1 over time has been used to indicate progression of the disease, although considerable individual variation exists.25 The second variable, dyspnea, is the most frequent and limiting symptom in COPD, and its relationship with mortality is even more important than that observed for pulmonary function.2 In contrast, the role of severe exacerbations in the natural history of COPD, especially those requiring hospitalization or an ED visit, has been recognized relatively recently.3,13 An increased frequency of exacerbations indicates clinical activity of the disease and is associated with a deterioration of comorbidities and worsening health status. The number of these severe exacerbations increases with the progression of the disease, especially in the advanced stages of COPD, and they are associated with a marked increase in mortality in the weeks following the exacerbations.26

Multicomponent scales have been developed to improve prognosis prediction in COPD, and they have proved to be better predictors of survival than any isolated variable. The most recognized and widely used is the BODE index.5 This index has shown not only prognostic reliability in several studies, but also sensitivity to clinical changes such as the impact of exacerbations and improvement with rehabilitation.27 However, the BODE index was developed and validated in outpatients without significant comorbidities,5 and, to our knowledge, only one study performed in a limited number of patients has proven its usefulness as a predictor of long-term mortality in patients hospitalized for COPD exacerbations.19 Alternatively, the BODEX index replaced the 6-min walk test component for exacerbation frequency a year ago and is more easily performed in this population, with a preserved power to predict mortality risk.6 Another recent multicomponent scale, the DOSE index, is associated with an increased risk of exacerbation in the following year, although it has been shown to be related to mortality at 5 years.9,28

Finally, the ADO index, composed of age, FEV1, and dyspnea stratified into different thresholds, has demonstrated its usefulness as a mortality predictor and has been updated to a larger cohort.7,8 However, the influence of age on the updated ADO index may appear clinically excessive (eg, being between 60 and 69 years of age has the same prognostic implication as presenting an FEV1 < 35% and more than having the maximal degree of dyspnea in the mMRC scale). Obviously, age is a predictor of mortality in our study as well (e-Table 1). However, our intention was to build on already validated cutoffs (FEV1, dyspnea, and exacerbations from the BODE/BODEX) and not to seek new cut points based exclusively on the criterion of statistical power. In this sense, to measure comorbidity, we used the most widely accepted comorbidity index (Charlson index) in its original version (ie, with correction for age, adding one point for each decade from age 40 years). The Charlson index has shown its usefulness in survival prediction and health-care resource use in COPD and other chronic diseases.29,30 Of note, as reflected in e-Table 1, the prognostic signification of the Charlson index without age correction is very similar (P < .0001; HR, 1.35; 95% CI, 1.18-1.57) to the age-stratified Charlson index that was used in the CODEX index (P < .0001; HR, 1.26; 95% CI, 1.15-1.39). Additionally, the CODEX index is also related to respiratory mortality.

This newly proposed CODEX index is essentially an evolution of the BODE and BODEX indexes, retaining their cutoffs for dyspnea, obstruction, and previous exacerbations, but replacing BMI with comorbidity measured using the original Charlson index modified by age. Although it is well known that a low BMI is associated with a poor survival even in patients hospitalized for AECOPD,31 in our study, the Charlson index showed a better prognostic capacity. In addition, the prognostic reliability of the CODEX index does not improve by adding the BMI or functional dependence measured with the Katz index (data not shown).

The importance of comorbidities in the prognosis of patients with COPD is well established.4 Even in studies in which patients with severe comorbidities were excluded by design, a significant percentage of patients died of cardiovascular disease or cancer during follow-up.5,32,33 Studies performed in hospitalized patients with COPD have shown a strong independent association with comorbidity and medium- to long-term mortality after discharge.15,16,34 However, to date, none of the COPD indexes has included comorbidity as a prognostic variable. Divo et al14 reported the usefulness of a comorbidity scale in COPD in improving the prognostic capacity of the BODE index. Our results confirm that compared with other previous COPD multicomponent indexes, the CODEX index is the most useful in predicting survival, hospital readmissions, and a combination of the two in the short and medium term in patients hospitalized for AECOPD. Only the updated ADO index in the assessment of mortality at 1 year and the DOSE index in the combined variable at 3 months show a similar relationship between sensitivity and specificity.

Our study has several limitations. First, comparison with the BODE index was not possible because the 6-min walk test was not included in the ESMI study protocol, because performing the 6-min walk test can be difficult in this population. Second, our study had a clear predominance of men. This finding is similar to all studies of COPD performed in our area and is probably related to the fact that fewer Spanish women than men smoked in the 1940s and 1950s.35 Finally, the validation cohort comes from the same institution in different time periods, but with the same inclusion and exclusion criteria of the ESMI study.

In conclusion, the CODEX index is a simple, valid tool for predicting survival and risk of COPD readmission or their combination in patients hospitalized for AECOPD during the year after hospital discharge, and it can be used to assess the severity of COPD in this population. Determining its usefulness in other populations requires further study.

Author contributions: Dr Almagro had full access to all the data in the study and takes responsibility for its integrity and the accuracy of the data analysis.

Dr Almagro: contributed to the study design, analysis and interpretation of the data, and writing of the manuscript.

Dr Soriano: contributed to the study design, analysis and interpretation of the data, and writing of the manuscript.

Dr Cabrera: contributed to the data collection and review of the manuscript.

Dr Boixeda: contributed to the data collection and review of the manuscript.

Dr Alonso-Ortiz: contributed to the data collection and review of the manuscript.

Dr Barreiro: contributed to the data collection and review of the manuscript.

Dr Diez-Manglano: contributed to the data collection and review of the manuscript.

Dr Murio: contributed to the study design and writing of the manuscript.

Dr Heredia: contributed to the data collection and review of the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Almagro participated in speaking activities, industry advisory committees, and other related activities sponsored by Takeda Nycomed; Almirall, SA; Boehringer-Ingelheim GmbH; Pfizer Inc; Chiesi Pharmaceuticals; GlaxoSmithKline; The Menarini Group; Merck, Sharp & Dohme; and Novartis. Drs Soriano, Cabrera, Boixeda, Alonso-Ortiz, Barreiro, Diez-Manglano, Murio, and Heredia have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or in the preparation of the manuscript.

Other contributions: We thank Tom Yohannan, medical copyeditor, for editorial assistance.

Additional information: The e-Appendix and e-Tables can be found in the “Supplemental Materials” area of the online article.

ADO

age, dyspnea, and airflow obstruction

AECOPD

acute exacerbation of COPD

AUC

area under the curve

BODE

BMI, airflow obstruction, dyspnea, and exercise capacity

BODEX

BMI, airflow obstruction, dyspnea, and previous severe exacerbations

CODEX

comorbidity, obstruction, dyspnea, and previous severe exacerbations

DOSE

dyspnea, airflow obstruction, smoking status, and exacerbation frequency

ESMI

EPOC en los Servicios de Medicina Interna

HR

hazard ratio

mMRC

modified Medical Research Council

ROC

receiver operating characteristic

Vos T, Flaxman AD, Naghavi M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010 [published correction appears inLancet. 2013;381(9867):628]. Lancet. 2012;380(9859):2163-2196. [CrossRef] [PubMed]
 
Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea is a better predictor of 5-year survival than airway obstruction in patients with COPD. Chest. 2002;121(5):1434-1440. [CrossRef] [PubMed]
 
Soler-Cataluña JJ, Martínez-García MA, Román Sánchez P, Salcedo E, Navarro M, Ochando R. Severe acute exacerbations and mortality in patients with chronic obstructive pulmonary disease. Thorax. 2005;60(11):925-931. [CrossRef] [PubMed]
 
Vestbo J, Hurd SS, Agustí AG, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med. 2013;187(4):347-365. [CrossRef] [PubMed]
 
Celli BR, Cote CG, Marin JM, et al. The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease. N Engl J Med. 2004;350(10):1005-1012. [CrossRef] [PubMed]
 
Soler-Cataluña JJ, Martínez-García MA, Sánchez LS, Tordera MP, Sánchez PR. Severe exacerbations and BODE index: two independent risk factors for death in male COPD patients. Respir Med. 2009;103(5):692-699. [CrossRef] [PubMed]
 
Puhan MA, Garcia-Aymerich J, Frey M, et al. Expansion of the prognostic assessment of patients with chronic obstructive pulmonary disease: the updated BODE index and the ADO index. Lancet. 2009;374(9691):704-711. [CrossRef] [PubMed]
 
Puhan MA, Hansel NN, Sobradillo P, et al; International COPD Cohorts Collaboration Working Group. Large-scale international validation of the ADO index in subjects with COPD: an individual subject data analysis of 10 cohorts. BMJ Open. 2012;2(6). pii: e002152.
 
Jones RC, Donaldson GC, Chavannes NH, et al. Derivation and validation of a composite index of severity in chronic obstructive pulmonary disease: the DOSE Index. Am J Respir Crit Care Med. 2009;180(12):1189-1195. [CrossRef] [PubMed]
 
Marin JM, Alfageme I, Almagro P, et al. Multicomponent indices to predict survival in COPD: the COCOMICS study. Eur Respir J. 2013;42(2):323-332. [CrossRef] [PubMed]
 
Almagro P, Salvadó M, Garcia-Vidal C, et al. Recent improvement in long-term survival after a COPD hospitalisation. Thorax. 2010;65(4):298-302. [CrossRef] [PubMed]
 
Erbas B, Ullah S, Hyndman RJ, Scollo M, Abramson M. Forecasts of COPD mortality in Australia: 2006-2025. BMC Med Res Methodol. 2012;12(1):17. [CrossRef] [PubMed]
 
Almagro P, Calbo E, Ochoa de Echagüen A, et al. Mortality after hospitalization for COPD. Chest. 2002;121(5):1441-1448. [CrossRef] [PubMed]
 
Divo M, Cote C, de Torres JP, et al; BODE Collaborative Group. Comorbidities and risk of mortality in patients with COPD. Am J Respir Crit Care Med. 2012;186(2):155-161. [CrossRef] [PubMed]
 
Roberts CM, Stone RA, Lowe D, Pursey NA, Buckingham RJ. Co-morbidities and 90-day outcomes in hospitalized COPD exacerbations. COPD. 2011;8(5):354-361. [CrossRef] [PubMed]
 
Almagro P, Cabrera FJ, Diez J, et al; Working Group on COPD, Spanish Society of Internal Medicine. Comorbidities and short-term prognosis in patients hospitalized for acute exacerbation of COPD: the EPOC en Servicios de medicina interna (ESMI) study. Chest. 2012;142(5):1126-1133. [CrossRef] [PubMed]
 
Almagro P, López F, Cabrera FJ, et al; Grupos de trabajo de EPOC y Paciente Pluripatológico y Edad Avanzada de la Sociedad Española de Medicina Interna. Comorbidities in patients hospitalized due to chronic obstructive pulmonary disease. A comparative analysis of the ECCO and ESMI studies [in Spanish]. Rev Clin Esp. 2012;212(6):281-286. [CrossRef] [PubMed]
 
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383. [CrossRef] [PubMed]
 
Sanjaume M, Almagro P, Rodríguez-Carballeira M, Barreiro B, Heredia JL, Garau J. Post-hospital mortality in patients re-admitted due to COPD. Utility of BODE index [in Spanish]. Rev Clin Esp. 2009;209(8):364-370. [CrossRef] [PubMed]
 
DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837-845. [CrossRef] [PubMed]
 
Seneff MG, Wagner DP, Wagner RP, Zimmerman JE, Knaus WA. Hospital and 1-year survival of patients admitted to intensive care units with acute exacerbation of chronic obstructive pulmonary disease. JAMA. 1995;274(23):1852-1857. [CrossRef] [PubMed]
 
Wildman MJ, Sanderson C, Groves J, et al. Predicting mortality for patients with exacerbations of COPD and Asthma in the COPD and Asthma Outcome Study (CAOS). QJM. 2009;102(6):389-399. [CrossRef] [PubMed]
 
Connors AF Jr, Dawson NV, Thomas C, et al. Outcomes following acute exacerbation of severe chronic obstructive lung disease. The SUPPORT investigators (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments). Am J Respir Crit Care Med. 1996;154(4 pt 1):959-967. [PubMed]
 
Steer J, Gibson J, Bourke SC. The DECAF score: predicting hospital mortality in exacerbations of chronic obstructive pulmonary disease. Thorax. 2012;67(11):970-976. [CrossRef] [PubMed]
 
Vestbo J, Edwards LD, Scanlon PD, et al; ECLIPSE Investigators. Changes in forced expiratory volume in 1 second over time in COPD. N Engl J Med. 2011;365(13):1184-1192. [CrossRef] [PubMed]
 
Suissa S, Dell’Aniello S, Ernst P. Long-term natural history of chronic obstructive pulmonary disease: severe exacerbations and mortality. Thorax. 2012;67(11):957-963. [CrossRef] [PubMed]
 
Cote CG, Celli BR. Pulmonary rehabilitation and the BODE index in COPD. Eur Respir J. 2005;26(4):630-636. [CrossRef] [PubMed]
 
Sundh J, Janson C, Lisspers K, Ställberg B, Montgomery S. The Dyspnoea, Obstruction, Smoking, Exacerbation (DOSE) index is predictive of mortality in COPD. Prim Care Respir J. 2012;21(3):295-301. [CrossRef] [PubMed]
 
Austin PC, Stanbrook MB, Anderson GM, Newman A, Gershon AS. Comparative ability of comorbidity classification methods for administrative data to predict outcomes in patients with chronic obstructive pulmonary disease. Ann Epidemiol. 2012;22(12):881-887. [CrossRef] [PubMed]
 
Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245-1251. [CrossRef] [PubMed]
 
Lainscak M, von Haehling S, Doehner W, et al. Body mass index and prognosis in patients hospitalized with acute exacerbation of chronic obstructive pulmonary disease. J Cachexia Sarcopenia Muscle. 2011;2(2):81-86. [CrossRef] [PubMed]
 
Simon-Tuval T, Scharf SM, Maimon N, Bernhard-Scharf BJ, Reuveni H, Tarasiuk A. Determinants of elevated healthcare utilization in patients with COPD. Respir Res. 2011;12:7 10.1186/1465-9921-12-7. [CrossRef] [PubMed]
 
Calverley PM, Anderson JA, Celli B, et al; TORCH investigators. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med. 2007;356(8):775-789. [CrossRef] [PubMed]
 
Slenter RH, Sprooten RT, Kotz D, Wesseling G, Wouters EF, Rohde GG. Predictors of 1-year mortality at hospital admission for acute exacerbations of chronic obstructive pulmonary disease. Respiration. 2013;85(1):15-26. [CrossRef] [PubMed]
 
Pozo-Rodríguez F, López-Campos JL, Alvarez-Martínez CJ, et al; AUDIPOC Study Group. Clinical audit of COPD patients requiring hospital admissions in Spain: AUDIPOC study. PLoS ONE. 2012;7(7):e42156. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) flowchart of participants and causes of exclusion. ESMI = EPOC en los Servicios de Medicina Interna.Grahic Jump Location
Figure Jump LinkFigure 2. Distribution of the CODEX index in the ESMI and validation cohorts. CODEX = comorbidity, obstruction, dyspnea, and previous severe exacerbations; n.s. = not significant. See Figure 1 legend for expansion of other abbreviations.Grahic Jump Location
Figure Jump LinkFigure 3. Kaplan-Meier curves for mortality. H.R. = hazard ratio.Grahic Jump Location
Figure Jump LinkFigure 4. Scores on the CODEX index and mortality or combined variable. See Figure 2 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 5. ROC curves for the CODEX, BODEX, DOSE, and updated ADO_u indexes. ROC = receiver operating characteristic. ADO_u = age, dyspnea, and airflow obstruction; BODEX = BMI, airflow obstruction, dyspnea, and previous severe exacerbations; DOSE = dyspnea, airflow obstruction, smoking status, and exacerbation frequency. See Figure 2 legend for expansion of other abbreviations.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1 —Variables and Thresholds to Estimate the CODEX Index

CODEX = comorbidity, obstruction, dyspnea, and previous severe exacerbations; mMRC = modified Medical Research Council.

a 

Charlson index: one point is added to the total score for each decade of life over the age of 50 y.

b 

Severe exacerbations of COPD during the previous year (hospitalization or ED consultations).

Table Graphic Jump Location
Table 2 —Demographic and Clinical Characteristics of the ESMI and Validation Cohorts

Data are presented as mean (SD) unless indicated otherwise. ESMI = EPOC en los Servicios de Medicina Interna; ns = not significant. See Table 1 legend for expansion of other abbreviation.

a 

Dyspnea measured with mMRC scale.

b 

No. hospitalizations for COPD during previous year.

c 

Postbronchodilator test.

Table Graphic Jump Location
Table 3 —Prognostic Validity of Mortality (HR and AUC) at 3 Mo and 1 Y After Hospital Discharge of CODEX Index and All Other COPD Multivariate Indexes (ADO, BODEX, BODE, and DOSE) in the ESMI Study

ADO = age, dyspnea, and airflow obstruction; AUC = area under the curve; BODE = BMI, airflow obstruction, dyspnea, and exercise capacity; BODEX = BMI, airflow obstruction, dyspnea, and previous severe exacerbations; DOSE = dyspnea, airflow obstruction, smoking status, and exacerbation frequency; HR = hazard ratio. See Table 1 and 2 legends for expansion of other abbreviations.

References

Vos T, Flaxman AD, Naghavi M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010 [published correction appears inLancet. 2013;381(9867):628]. Lancet. 2012;380(9859):2163-2196. [CrossRef] [PubMed]
 
Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea is a better predictor of 5-year survival than airway obstruction in patients with COPD. Chest. 2002;121(5):1434-1440. [CrossRef] [PubMed]
 
Soler-Cataluña JJ, Martínez-García MA, Román Sánchez P, Salcedo E, Navarro M, Ochando R. Severe acute exacerbations and mortality in patients with chronic obstructive pulmonary disease. Thorax. 2005;60(11):925-931. [CrossRef] [PubMed]
 
Vestbo J, Hurd SS, Agustí AG, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med. 2013;187(4):347-365. [CrossRef] [PubMed]
 
Celli BR, Cote CG, Marin JM, et al. The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease. N Engl J Med. 2004;350(10):1005-1012. [CrossRef] [PubMed]
 
Soler-Cataluña JJ, Martínez-García MA, Sánchez LS, Tordera MP, Sánchez PR. Severe exacerbations and BODE index: two independent risk factors for death in male COPD patients. Respir Med. 2009;103(5):692-699. [CrossRef] [PubMed]
 
Puhan MA, Garcia-Aymerich J, Frey M, et al. Expansion of the prognostic assessment of patients with chronic obstructive pulmonary disease: the updated BODE index and the ADO index. Lancet. 2009;374(9691):704-711. [CrossRef] [PubMed]
 
Puhan MA, Hansel NN, Sobradillo P, et al; International COPD Cohorts Collaboration Working Group. Large-scale international validation of the ADO index in subjects with COPD: an individual subject data analysis of 10 cohorts. BMJ Open. 2012;2(6). pii: e002152.
 
Jones RC, Donaldson GC, Chavannes NH, et al. Derivation and validation of a composite index of severity in chronic obstructive pulmonary disease: the DOSE Index. Am J Respir Crit Care Med. 2009;180(12):1189-1195. [CrossRef] [PubMed]
 
Marin JM, Alfageme I, Almagro P, et al. Multicomponent indices to predict survival in COPD: the COCOMICS study. Eur Respir J. 2013;42(2):323-332. [CrossRef] [PubMed]
 
Almagro P, Salvadó M, Garcia-Vidal C, et al. Recent improvement in long-term survival after a COPD hospitalisation. Thorax. 2010;65(4):298-302. [CrossRef] [PubMed]
 
Erbas B, Ullah S, Hyndman RJ, Scollo M, Abramson M. Forecasts of COPD mortality in Australia: 2006-2025. BMC Med Res Methodol. 2012;12(1):17. [CrossRef] [PubMed]
 
Almagro P, Calbo E, Ochoa de Echagüen A, et al. Mortality after hospitalization for COPD. Chest. 2002;121(5):1441-1448. [CrossRef] [PubMed]
 
Divo M, Cote C, de Torres JP, et al; BODE Collaborative Group. Comorbidities and risk of mortality in patients with COPD. Am J Respir Crit Care Med. 2012;186(2):155-161. [CrossRef] [PubMed]
 
Roberts CM, Stone RA, Lowe D, Pursey NA, Buckingham RJ. Co-morbidities and 90-day outcomes in hospitalized COPD exacerbations. COPD. 2011;8(5):354-361. [CrossRef] [PubMed]
 
Almagro P, Cabrera FJ, Diez J, et al; Working Group on COPD, Spanish Society of Internal Medicine. Comorbidities and short-term prognosis in patients hospitalized for acute exacerbation of COPD: the EPOC en Servicios de medicina interna (ESMI) study. Chest. 2012;142(5):1126-1133. [CrossRef] [PubMed]
 
Almagro P, López F, Cabrera FJ, et al; Grupos de trabajo de EPOC y Paciente Pluripatológico y Edad Avanzada de la Sociedad Española de Medicina Interna. Comorbidities in patients hospitalized due to chronic obstructive pulmonary disease. A comparative analysis of the ECCO and ESMI studies [in Spanish]. Rev Clin Esp. 2012;212(6):281-286. [CrossRef] [PubMed]
 
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383. [CrossRef] [PubMed]
 
Sanjaume M, Almagro P, Rodríguez-Carballeira M, Barreiro B, Heredia JL, Garau J. Post-hospital mortality in patients re-admitted due to COPD. Utility of BODE index [in Spanish]. Rev Clin Esp. 2009;209(8):364-370. [CrossRef] [PubMed]
 
DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837-845. [CrossRef] [PubMed]
 
Seneff MG, Wagner DP, Wagner RP, Zimmerman JE, Knaus WA. Hospital and 1-year survival of patients admitted to intensive care units with acute exacerbation of chronic obstructive pulmonary disease. JAMA. 1995;274(23):1852-1857. [CrossRef] [PubMed]
 
Wildman MJ, Sanderson C, Groves J, et al. Predicting mortality for patients with exacerbations of COPD and Asthma in the COPD and Asthma Outcome Study (CAOS). QJM. 2009;102(6):389-399. [CrossRef] [PubMed]
 
Connors AF Jr, Dawson NV, Thomas C, et al. Outcomes following acute exacerbation of severe chronic obstructive lung disease. The SUPPORT investigators (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments). Am J Respir Crit Care Med. 1996;154(4 pt 1):959-967. [PubMed]
 
Steer J, Gibson J, Bourke SC. The DECAF score: predicting hospital mortality in exacerbations of chronic obstructive pulmonary disease. Thorax. 2012;67(11):970-976. [CrossRef] [PubMed]
 
Vestbo J, Edwards LD, Scanlon PD, et al; ECLIPSE Investigators. Changes in forced expiratory volume in 1 second over time in COPD. N Engl J Med. 2011;365(13):1184-1192. [CrossRef] [PubMed]
 
Suissa S, Dell’Aniello S, Ernst P. Long-term natural history of chronic obstructive pulmonary disease: severe exacerbations and mortality. Thorax. 2012;67(11):957-963. [CrossRef] [PubMed]
 
Cote CG, Celli BR. Pulmonary rehabilitation and the BODE index in COPD. Eur Respir J. 2005;26(4):630-636. [CrossRef] [PubMed]
 
Sundh J, Janson C, Lisspers K, Ställberg B, Montgomery S. The Dyspnoea, Obstruction, Smoking, Exacerbation (DOSE) index is predictive of mortality in COPD. Prim Care Respir J. 2012;21(3):295-301. [CrossRef] [PubMed]
 
Austin PC, Stanbrook MB, Anderson GM, Newman A, Gershon AS. Comparative ability of comorbidity classification methods for administrative data to predict outcomes in patients with chronic obstructive pulmonary disease. Ann Epidemiol. 2012;22(12):881-887. [CrossRef] [PubMed]
 
Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245-1251. [CrossRef] [PubMed]
 
Lainscak M, von Haehling S, Doehner W, et al. Body mass index and prognosis in patients hospitalized with acute exacerbation of chronic obstructive pulmonary disease. J Cachexia Sarcopenia Muscle. 2011;2(2):81-86. [CrossRef] [PubMed]
 
Simon-Tuval T, Scharf SM, Maimon N, Bernhard-Scharf BJ, Reuveni H, Tarasiuk A. Determinants of elevated healthcare utilization in patients with COPD. Respir Res. 2011;12:7 10.1186/1465-9921-12-7. [CrossRef] [PubMed]
 
Calverley PM, Anderson JA, Celli B, et al; TORCH investigators. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med. 2007;356(8):775-789. [CrossRef] [PubMed]
 
Slenter RH, Sprooten RT, Kotz D, Wesseling G, Wouters EF, Rohde GG. Predictors of 1-year mortality at hospital admission for acute exacerbations of chronic obstructive pulmonary disease. Respiration. 2013;85(1):15-26. [CrossRef] [PubMed]
 
Pozo-Rodríguez F, López-Campos JL, Alvarez-Martínez CJ, et al; AUDIPOC Study Group. Clinical audit of COPD patients requiring hospital admissions in Spain: AUDIPOC study. PLoS ONE. 2012;7(7):e42156. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Supporting Data

Online Supplement

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543