Definitions of Exacerbations: Does It Really Matter in Clinical Trials on COPD?

doi:10.1378/chest.08-1680

Chest

Volume 136, Issue 3, September 2009, Pages 918-923

https://doi.org/10.1378/chest.08-1680 Get rights and content

Many definitions of COPD exacerbations are reported. The choice for a definition determines the number of exacerbations observed. However, the effect of different definitions on the effect sizes of randomized controlled trials is unclear. This article provides an overview of the large variation of definitions of COPD exacerbations from the literature. Furthermore, the effect of using different definitions on effect sizes (relative risk and hazard ratio) was investigated in a randomized controlled discontinuation trial of inhaled corticosteroids. The following definitions were applied: (1) unscheduled medical attention, (2) a course of oral corticosteroids/antibiotics, (3) deterioration in two major or one major and one minor symptom according to Anthonisen (referenced later), (4) a change in one or more symptoms, (5) a change in two or more symptoms, and (6) a combination of numbers 2 and 4. Relative risks for the exacerbation rate ranged from 1.19 to 1.49, and hazard ratios for time to first exacerbation ranged from 1.36 to 1.84 for the various definitions, varying from nonsignificant to significant. Because the definition of a COPD exacerbation has an impact on the effect size of interventions, there is an urgent need for concerted attempts to reach agreement on a definition of an exacerbation. Also, the exact definition to be used in a study should be specified in the protocol.

Section snippets

How Have Exacerbations Been Defined in Study Reports?

A MEDLINE search (January 1987 to May 2007) identified 131 articles: COPD [tw] OR pulmonary disease, chronic obstructive [MeSH] OR COPD AND exacerbation *[tw] AND Randomized Controlled Trial [pt] AND English [la] NOT review [pt] in which only journals with a citation index of three or more in 2006 were included (Fig 1). Only 53% of these studies provided a definition of an exacerbation. In most of the studies in which no definition of an exacerbation was given, the variable “exacerbation” was

What Is the Impact of Different Definitions of Exacerbation on Effect Size Estimation in Randomized Controlled Trials?

Using data from the COPD study at the Department of Pulmonology in Enschede (COPE; a randomized, double-blind, placebo-controlled trial on discontinuation of inhaled corticosteroids [ICSs]),¹³ we were able to assess the potential impact of exacerbation definition on size and significance of treatment effect. Therefore, six definitions of exacerbations (Table 2) and two effect parameters that are commonly used in the literature (“exacerbation rate [=number of exacerbations per year]” and “time

Implications for Randomized Controlled Trials

We demonstrated that different definitions of an exacerbation led to different effect sizes and hence different conclusions about the effect of withdrawing ICS in patients with COPD. Thus, it really matters which definition of an exacerbation is chosen. Considerably larger RRs and HRs were found with the “Anthonisen definition” that specifically identifies major and minor symptoms (aimed at bacterial exacerbations), but also with the definition incorporating “a short course of oral

Conclusions

We have shown that many definitions of COPD exacerbation exist and the definition and analysis technique chosen to assess the treatment effect has an impact. This could potentially lead to opportunistic presentation of results from clinical trials. These results stress the need for concerted attempts to reach consensus about definitions and analysis methods of COPD exacerbations to be used in future randomized controlled trials.

Acknowledgments

Author contributions: All authors contributed to data interpretation and the writing of the article. Dr. Effing contributed to the conception of the topic, current reanalysis, literature search, and reanalysis of the data. Dr. Kerstjens contributed to the conception of the topic, current reanalysis, and reanalysis of the data. Drs. Monninkhof and van der Valk contributed to the collection of the COPE data, conception of the topic, and current reanalysis. Drs. Wouters and Postma contributed to

References (0)

Cited by (56)

Comparison of machine learning algorithms for the identification of acute exacerbations in chronic obstructive pulmonary disease
2020, Computer Methods and Programs in Biomedicine
Citation Excerpt :
To date, the definition of AECOPDs can be categorized broadly on the basis of patient symptoms (e.g., worsening of dyspnea and increases in sputum volume or purulence) or definitive events (e.g., requiring hospitalization, taking antibiotics) [18,19]. Furthermore, according to the differences in basic conditions, causes and complications, the acute exacerbations between patients, and within the same patient may be very different [20-22]. These issues have caused great difficulties in the identification of AECOPDs.
Identifying acute exacerbations in chronic obstructive pulmonary disease (AECOPDs) is of utmost importance for reducing the associated mortality and financial burden. In this research, the authors aimed to develop identification models for AECOPDs and to compare the relative performance of different modeling paradigms to find the best model for this task.
Data were extracted from electronic medical records (EMRs) of patients with chronic obstructive pulmonary disease who admitted to the China-Japan Friendship Hospital between February 2011 and March 2017. Five machine learning algorithms (random forest, support vector machine, logistic regression, K-nearest neighbor and naïve Bayes) were used to develop the AECOPDs identification models. Feature selection was performed to find an optimal feature subset. 10-folds cross-validation was used to find the best hyperparameters for each model. The following metrics: area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value were used to evaluate the performance of these models.
A total of 303 EMRs (AECOPDs patients:135; None AECOPDs patients: 168) were included in the study. The SVM model obtained the best performance (sensitivity: 0.80, specificity: 0.83, positive predictive value:0.81, negative predictive value:0.85 and area under the receiver operating characteristic curve: 0.90) after performing feature selection.
Our research confirms that the proposed model based on the support vector machine is a powerful tool to identify AECOPDs patients, and it is promising to provide decision support for clinicians when they are struggling to give a confirmed clinical diagnosis.
Impaired Sleep Quality in COPD Is Associated With Exacerbations: The CanCOLD Cohort Study
2019, Chest
Citation Excerpt :
However, analyses excluding self-reported asthma were consistent with our main results. Second, because definitions of COPD exacerbation vary among studies, comparison may be limited,26,27 but CanCOLD used a standard definition, as recommended by GOLD.28 Our assessment of exacerbations was based on participant recall, which may be subject to recall bias, but minimized by the relatively frequent telephone assessments (ie, every 3 months).
COPD increases susceptibility to sleep disturbances, which may in turn predispose to increased respiratory symptoms. The objective of this study was to evaluate, in a population-based sample, the relationship between subjective sleep quality and risk of COPD exacerbations.
Data were obtained from the Canadian Cohort Obstructive Lung Disease (CanCOLD) study. Participants with COPD who had completed 18 months of follow-up were included. Sleep quality was measured with the Pittsburgh Sleep Quality Index (PSQI) and a three-factor analysis. Symptom-based (dyspnea or sputum change ≥ 48 h) and event-based (symptoms plus medication or unscheduled health services use) exacerbations were assessed. Association of PSQI with exacerbation rate was assessed by using negative binomial regression. Exacerbation-free survival was also assessed.
A total of 480 participants with COPD were studied, including 185 with one or more exacerbations during follow-up and 203 with poor baseline sleep quality (PSQI score > 5). Participants with subsequent symptom-based exacerbations had higher median baseline PSQI scores than those without (6.0 [interquartile range, 3.0-8.0] vs 5.0 [interquartile range, 2.0-7.0]; P = .01), and they were more likely to have baseline PSQI scores > 5 (50.3% vs 37.3%; P = .01). Higher PSQI scores were associated with increased symptom-based exacerbation risk (adjusted rate ratio, 1.09; 95% CI, 1.01-1.18; P = .02) and event-based exacerbation risk (adjusted rate ratio, 1.10; 95% CI, 1.00-1.21; P = .048). The association occurred mainly in those with undiagnosed COPD. Strongest associations were with Factor 3 (sleep disturbances and daytime dysfunction). Time to symptom-based exacerbation was shorter in participants with poor sleep quality (adjusted hazard ratio, 1.49; 95% CI, 1.09-2.03).
Higher baseline PSQI scores were associated with increased risk of COPD exacerbation over 18 months’ prospective follow-up.
A comparison between discrete and continuous time Bayesian networks in learning from clinical time series data with irregularity
2019, Artificial Intelligence in Medicine
Citation Excerpt :
It is defined either in terms of specific worsening symptoms for consecutive days or if there is evidence of a patient's hospital admission due to an exacerbation. Unfortunately, clinicians have so far not been able to agree on a clinical definition of an exacerbation [3,8]. Rather than focusing our research on automatically deciding on the presence or absence of an exacerbation, using multiple definitions, we aim at trying to understand the dynamic behavior of the symptoms of COPD.
Recently, mobile devices, such as smartphones, have been introduced into healthcare research to substitute paper diaries as data-collection tools in the home environment. Such devices support collecting patient data at different time points over a long period, resulting in clinical time-series data with high temporal complexity, such as time irregularities. Analysis of such time series poses new challenges for machine-learning techniques. The clinical context for the research discussed in this paper is home monitoring in chronic obstructive pulmonary disease (COPD).
The goal of the present research is to find out which properties of temporal Bayesian network models allow to cope best with irregularly spaced multivariate clinical time-series data.
Two mainstream temporal Bayesian network models of multivariate clinical time series are studied: dynamic Bayesian networks, where the system is described as a snapshot at discrete time points, and continuous time Bayesian networks, where transitions between states are modeled in continuous time. Their capability of learning from clinical time series that vary in nature are extensively studied. In order to compare the two temporal Bayesian network types for regularly and irregularly spaced time-series data, three typical ways of observing time-series data were investigated: (1) regularly spaced in time with a fixed rate; (2) irregularly spaced and missing completely at random at discrete time points; (3) irregularly spaced and missing at random at discrete time points. In addition, similar experiments were carried out using real-world COPD patient data where observations are unevenly spaced.
For regularly spaced time series, the dynamic Bayesian network models outperform the continuous time Bayesian networks. Similarly, if the data is missing completely at random, discrete-time models outperform continuous time models in most situations. For more realistic settings where data is not missing completely at random, the situation is more complicated. In simulation experiments, both models perform similarly if there is strong prior knowledge available about the missing data distribution. Otherwise, continuous time Bayesian networks perform better. In experiments with unevenly spaced real-world data, we surprisingly found that a dynamic Bayesian network where time is ignored performs similar to a continuous time Bayesian network.
The results confirm conventional wisdom that discrete-time Bayesian networks are appropriate when learning from regularly spaced clinical time series. Similarly, we found that time series where the missingness occurs completely at random, dynamic Bayesian networks are an appropriate choice. However, for complex clinical time-series data that motivated this research, the continuous-time models are at least competitive and sometimes better than their discrete-time counterparts. Furthermore, continuous-time models provide additional benefits of being able to provide more fine-grained predictions than discrete-time models, which will be of practical relevance in clinical applications.
Chronic Obstructive Pulmonary Disease Exacerbations: A Need for Action
2018, American Journal of Medicine
Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality, with exacerbations contributing strongly to the overall severity and burden of disease at the individual level. Many patients with COPD are managed predominantly in the primary care setting; therefore, primary care physicians (PCPs) must be aware of and understand the causes, effects, and management of COPD exacerbations. This review offers practical information about how exacerbations are defined in the treatment setting, pathogenic and environmental causes, options for exacerbation treatment and prevention, and suggestions for improving care in the clinic. Exacerbations have a strong negative effect on patients and are associated with an increased risk of further exacerbations and decreases in lung function. Thus, it is important that exacerbations not requiring hospitalization be identified and confidently managed in the clinical setting. Understanding treatment options for acute exacerbations and maintenance treatment to prevent future exacerbations may help PCPs better understand what they can do to support their patients. This review aims to provide useful guidance for PCPs to identify exacerbations and adjust their practice for optimal management of exacerbations in patients with COPD. It also aims to encourage PCPs that they have an important task in integrating effective COPD management into the primary care setting.
What are the definitions of acute COPD exacerbations and COPD decompensations?
2017, Revue des Maladies Respiratoires
The Inaccuracy of Patient Recall for COPD Exacerbation Rate Estimation and Its Implications: Results from Central Adjudication
2016, Chest
Citation Excerpt :
A limitation of our study is that even though we put great effort in accurate assessment and adjudication of exacerbations, we still may have missed or misclassified some. No standardized definition and measurement methods exist, and different definitions may lead to different results.30,31 We used an event-based definition, the most commonly used definition in drug development and clinical trials used by the US Food and Drug Administration and European Medicines Agency to approve new therapies.
COPD exacerbation incidence rates are often ascertained retrospectively through patient recall and self-reports. We compared exacerbation ascertainment through patient self-reports and single-physician chart review to central adjudication by a committee and explored determinants and consequences of misclassification.
Self-reported exacerbations (event-based definition) in 409 primary care patients with COPD participating in the International Collaborative Effort on Chronic Obstructive Lung Disease: Exacerbation Risk Index Cohorts (ICE COLD ERIC) cohort were ascertained every 6 months over 3 years. Exacerbations were adjudicated by single experienced physicians and an adjudication committee who had information from patient charts. We assessed the accuracy (sensitivities and specificities) of self-reports and single-physician chart review against a central adjudication committee (AC) (reference standard). We used multinomial logistic regression and bootstrap stability analyses to explore determinants of misclassifications.
The AC identified 648 exacerbations, corresponding to an incidence rate of 0.60 ± 0.83 exacerbations/patient-year and a cumulative incidence proportion of 58.9%. Patients self-reported 841 exacerbations (incidence rate, 0.75 ± 1.01; incidence proportion, 59.7%). The sensitivity and specificity of self-reports were 84% and 76%, respectively, those of single-physician chart review were between 89% and 96% and 87% and 99%, respectively. The multinomial regression model and bootstrap selection showed that having experienced more exacerbations was the only factor consistently associated with underreporting and overreporting of exacerbations (underreporters: relative risk ratio [RRR], 2.16; 95% CI, 1.76-2.65 and overreporters: RRR, 1.67; 95% CI, 1.39-2.00).
Patient 6-month recall of exacerbation events are inaccurate. This may lead to inaccurate estimates of incidence measures and underestimation of treatment effects. The use of multiple data sources combined with event adjudication could substantially reduce sample size requirements and possibly cost of studies.
www.ClinicalTrials.gov, NCT00706602

View all citing articles on Scopus

Funding/Support: This research was supported by grants from the Dutch Asthma Foundation, the Netherlands.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal.org/site/misc/reprints.xhtml).

View full text