Elsevier

The Lancet

Volume 374, Issue 9691, 29 August–4 September 2009, Pages 704-711
The Lancet

Articles
Expansion of the prognostic assessment of patients with chronic obstructive pulmonary disease: the updated BODE index and the ADO index

https://doi.org/10.1016/S0140-6736(09)61301-5Get rights and content

Summary

Background

The BODE index (including body-mass index, airflow obstruction, dyspnoea, and exercise capacity) was an important contribution to the prognostic assessment of patients with chronic obstructive pulmonary disease (COPD). However, no study has assessed whether the risk of mortality predicted by the BODE index matches the observed mortality in different populations. We assessed the calibration of the BODE index, updated it to improve its calibration, and developed and validated a simplified index for use in primary-care settings.

Methods

We included 232 patients from the Swiss Barmelweid cohort with longstanding and severe COPD and 342 patients from the Spanish Phenotype and Course of COPD cohort study who had had their first hospital admission due to moderate-to-severe COPD. In both cohorts we compared the observed 3-year risk of all-cause mortality with the risk predicted by the BODE index. We then updated the BODE index and developed a simplified ADO index (including age, dyspnoea, and airflow obstruction) from the Swiss cohort, and validated both in the Spanish cohort.

Findings

Calibration of the BODE index was poor, with relative underprediction of 3-year risk of mortality by 36% in the Swiss cohort (median predicted risk 21·7% [IQR 12·7–31·7] vs 34·1% observed risk; p=0·013) and relative overprediction by 39% in the Spanish cohort (16·7% [12·7–31·7] vs 12·0%; p=0·035). The 3-year risk of mortality predicted by both the updated BODE (median 10·7% [8·1–13·8]) and ADO indices (11·8% [9·1–14·3]) matched the observed mortality in the Spanish cohort well (p=0·99 and p=0.98, respectively).

Interpretation

Both the updated BODE and ADO indices could lend support to the prognostic assessment of patients with COPD in specialised and primary-care settings. Such assessment enhances the targeting of treatments to individual patients.

Funding

Swiss National Science Foundation; Klinik Barmelweid; Fondo de Investigación Sanitaria Ministry of Health, Spain; Agència d'Avaluació de Tecnologia i Recerca Mèdiques, Catalonia Government; Spanish Society of Pneumology and Thoracic Surgery; Catalan Foundation of Pneumology; Red RESPIRA; Red RCESP; Fondo de Investigación Sanitaria; Fondo de Investigación Sanitaria; Fundació La Marató de TV3; Novartis Farmacèutica, Spain.

Introduction

The BODE index was an important contribution to prognostic research in chronic obstructive pulmonary disease (COPD).1 It combines information about several predictors including body-mass index (BMI), airflow obstruction (forced expiratory volume in 1 s [FEV1]), dyspnoea (Medical Research Council [MRC] dyspnoea scale), and exercise capacity (6-min walk distance) in a score ranging from 0 to 10. This prognostic index predicts mortality significantly better than does lung function—the traditional prognostic COPD indicator—alone. The BODE index contributed to the acceptance that prognostic assessment in patients with COPD should go beyond lung function.2, 3

Findings from several studies4, 5, 6 have confirmed that the BODE index has better discriminative properties than does lung function. Discrimination refers to the ability of the prognostic index to distinguish between patients who will or will not die over a specific period of time. However, discrimination is not the only property that is relevant for prognostic indices. To be useful in practice, prognostic instruments should accurately predict the absolute risk of an event in individual patients.7 Guided by these predicted risks, clinicians and patients might decide on more or less comprehensive treatment to modify that risk. The absolute risks as predicted by risk scores should be compared with the observed risks in at least one other population (so-called calibration).8, 9, 10 Without any assessment of calibration, clinicians should be very cautious in applying such scores in practice because treatment selection could be inadequate if the risk is overestimated or underestimated. Unlike widely used risk scores such as the Framingham risk score and the APACHE (acute physiology and chronic health evaluation) scores, the BODE index does not provide absolute risks of mortality and its calibration has never been assessed. As a consequence, the BODE index seems not yet ready for use as a prognostic instrument in patients with COPD.

We aimed to assess the calibration of the BODE index in two different COPD populations, to explore how its prediction could be improved, and to develop a simplified risk index that is also applicable in primary-care settings.

Section snippets

Study design and patients

We included all patients with COPD in the Swiss Barmelweid cohort and the Spanish Phenotype and Course of COPD (PAC-COPD) cohort study.11 Patients in the Swiss cohort had longstanding and, on average, severe COPD (according to criteria from the Global initiative for chronic Obstructive Lung Disease [GOLD]), whereas those in the Spanish cohort study were enrolled after they had had their first hospital admission due to an exacerbation of moderate-to-severe COPD.11

In the Swiss cohort we included

Results

232 patients were included in the Swiss cohort and 342 in the Spanish cohort. The Swiss cohort included elderly COPD patients with moderate-to-severe chronic airflow obstruction, a moderate degree of dyspnoea, a mean 6-min walk distance of 363 m, and a mean partial pressure of oxygen in arterial blood (PaO2) of 62·7 mm Hg (table 1). Cardiovascular comorbidity was common, and most patients received one or several inhaled drugs (table 1). Median follow-up was 34 months (range 3–50) and 3-year

Discussion

Our study showed that the original BODE index did not accurately predict mortality in two different COPD populations from Switzerland and Spain. The updated BODE index and the ADO index were similarly accurate in their risk prediction in an external validation cohort, and better than was the original BODE index. The simplified point systems for the updated BODE and ADO indices developed for patients with longstanding severe disease and patients after their first hospital admission due to an

References (40)

  • MA Puhan et al.

    COPD: more than respiratory

    Lancet

    (2008)
  • BR Celli et al.

    The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease

    N Engl J Med

    (2004)
  • CG Cote et al.

    The modified BODE index: validation with mortality in COPD

    Eur Respir J

    (2008)
  • FJ Martinez et al.

    Longitudinal change in the BODE index predicts mortality in severe emphysema

    Am J Respir Crit Care Med

    (2008)
  • KG Moons et al.

    Prognosis and prognostic research: what, why, and how?

    BMJ

    (2009)
  • DG Altman et al.

    What do we mean by validating a prognostic model?

    Stat Med

    (2000)
  • DG Altman et al.

    Prognosis and prognostic research: validating a prognostic model

    BMJ

    (2009)
  • MA Puhan et al.

    Interval versus continuous high-intensity exercise in chronic obstructive pulmonary disease: a randomized trial

    Ann Intern Med

    (2006)
  • BR Celli et al.

    Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper

    Eur Respir J

    (2004)
  • E Balcells et al.

    Characteristics of patients admitted for the first time for COPD exacerbation

    Respir Med

    (2009)
  • Cited by (0)

    These authors contributed equally

    View full text