American Journal of Respiratory and Critical Care Medicine

Exacerbations occur commonly in patients with moderate or severe chronic obstructive pulmonary disease (COPD) but factors affecting their severity and frequency or effects on quality of life are unknown. We measured daily peak expiratory flow rate (PEFR) and daily respiratory symptoms for 1 yr in 70 COPD patients (52 male, 18 female, mean age [ ± SD] 67.5 ± 8.3 yr, FEV1 1.06 ± 0.45 L, FVC 2.48  ± 0.82 L, FEV1/FVC 44 ± 15%, FEV1 reversibility 6.7 ± 9.1%, PaO2 8.8 ± 1.1 kPa). Quality of life was measured by the St. George's Respiratory Questionnaire (SGRQ). Exacerbations (E) were assessed at acute visit (reported exacerbation) or from diary card data each month (unreported exacerbation). In 61 (87%) patients there were 190 exacerbations (median 3; range, 1 to 8) of which 93 (51%) were reported. There were no differences in major symptoms (increase in dyspnea, sputum volume, or purulence) or physiological parameters between reported and unreported exacerbations. At exacerbation, median peak flow fell by an average of 6.6 L/min (p = 0.0003). Using the median number of exacerbations as the cutoff point, patients were classified as infrequent exacerbators (E = 0 to 2) or frequent exacerbators (E = 3 to 8). The SGRQ Total and component scores were significantly worse in the group that had frequent exacerbations: SGRQ Total score (mean difference = 14.8, p < 0.001), Symptoms (23.1, p < 0.001), Activities (12.2, p = 0.003), Impacts (13.9, p = 0.002). However there was no difference between frequent and infrequent exacerbators in the fall in peak flow at exacerbation. Factors predictive of frequent exacerbations were daily cough (p = 0.018), daily wheeze (p = 0.011), and daily cough and sputum (p = 0.009) and frequent exacerbations in the previous year (p  = 0.001). These findings suggest that patient quality of life is related to COPD exacerbation frequency.

Patients with moderate and severe chronic obstructive pulmonary disease (COPD) are prone to exacerbations; the frequency of these exacerbations increases with the severity of COPD (1). However, there is little information on the effect of exacerbation on quality of life. Some patients are prone to frequent exacerbations that may have considerable impact on activities of daily living and well-being, yet factors predicting development and severity of exacerbation have not been studied. Impairment of quality of life can be quantified with disease-specific quality-of-life measures, validated for use in chronic respiratory disease (2, 3). Changes in lung function occur with exacerbation, but these are variable, because of the relatively fixed airflow obstruction. However, quality-of-life scores show only a moderate relationship with symptoms and physiological parameters (4-7) and thus exacerbation may produce considerable distress to the patient, without change in lung function.

Previous studies of exacerbations have concentrated on predictive factors for hospital admission in patients with COPD (8-11). Poor quality of life is related to likelihood of hospital admission and to increased use of resources. However, most COPD exacerbations are treated at home and are not associated with hospital admission. There is no information as to the impact of such exacerbations and their differing severity on the daily life of COPD patients.

In this study, patients with moderate to severe COPD were followed prospectively over 12 mo. Patients recorded daily symptoms and daily peak flow on diary cards and reported exacerbations to the clinical team. We have investigated the effect of exacerbation frequency and severity on health-related quality of life and evaluated factors that predispose to exacerbation.


Eighty-four (57 male, 27 female) of 125 (81 male, 44 female) patients with COPD who attended the outpatient clinics were recruited in October 1995. Inclusion criteria were: FEV1 < 70% predicted for age and height, β2-agonist reversibility < 15% or 200 ml (12, 13), and no exacerbations for the previous 4 wk. Exclusion criteria were asthma, bronchiectasis, carcinoma of the bronchus, or inability to complete diary cards. During the study period there were three deaths, one from allergic bronchopulmonary aspergillosis and two from acute exacerbations of COPD. Eleven patients withdrew, 10 because of poor compliance and one because of reactivation of tuberculosis. Thus, the number of patients who remained in the study was 73 (54 male, 19 female) with 70 patients (52 male, 18 female) completing the study.

At recruitment, measurements were made of FEV1 and FVC (rolling seal spirometer; Sensor Medic Corp., Yorba Linda, CA), reversibility to salbutamol (400 μg by metered-dose inhaler) after withdrawing bronchodilators for 4 h, arterialized ear lobe blood gases (model 278 Blood Gas Analyzer; Ciba-Corning, Medfield, MA) (14), the number of exacerbations during the previous year (past exacerbations), and the cardiovascular history. Steroid use, vaccine history, and daily symptoms (dyspnea, cough, wheeze, sputum production) were recorded.

Patients completed daily diary cards after morning medication to record peak expiratory flow rate (PEFR) using a Mini-Wright peak flow meter (Clement Clarke International Ltd, Harlow, UK), increase in upper respiratory tract symptoms (nasal discharge, sore throat), and increase in lower respiratory tract symptoms (dyspnea, sputum, cough, wheezing) and fever. Patients were reviewed monthly until April 1996 and every three months during the summer months for a total follow-up of 12 mo. Approval for this study was obtained from the ethics committee of the East London and City Health Authority.


An exacerbation was diagnosed (15) if the following symptom patterns were experienced for at least two consecutive days: either two or more of three major symptoms (increase in dyspnea, sputum purulence, and increased sputum volume); or any one major symptom together with any one of the following minor symptoms—increase in nasal discharge, wheeze, sore throat, cough, or fever. When patients noticed deterioration in symptoms, they telephoned a member of the clinical team and were seen within 48 h. If their symptoms were compatible with a reported exacerbation, spirometry was measured and treatment prescribed of antibiotics and/or oral steroids and/or increased inhaled steroids. Exacerbations identified from the diary cards at clinic visits when patients were not reviewed acutely were termed “unreported exacerbations.”

Health-related Quality Of Life (HRQOL)

Indices of health-related quality of life (HRQOL) were obtained using the St. George's Respiratory Questionnaire (SGRQ) (2, 16). Three component indices were calculated using empirically derived weightings: Symptom, Activity, and Impact scores from which a Total score was computed. Scores vary from 0 (no disability) to 100 (maximum disability). Patients were also asked to complete the Medical Research Council (MRC) dyspnea scale questionnaire (17) which consists of five grades: 1—shortness of breath with strenuous exercise; 2—shortness of breath when hurrying; 3—walking slower than people of same age; 4—needing to stop after walking 100 yards on level; 5— too breathless to leave the house. The questionnaires were completed by the patients at the last clinic visit without directions from the researchers.

Statistical Analysis

Reported and unreported exacerbations were compared by acute symptoms (chi-square test) and changes in PEFR at the time of exacerbation (Mann-Whitney U test). The Wilcoxon signed rank test was used to test significance of PEFR change over the 2 d prior to exacerbation from zero. Baseline PEFR was taken as the mean of Days 14 to 7 prior to onset of exacerbation. Recovery of PEFR was defined as the time at which the 3-d moving average of PEFR exceeded or was nearest baseline.

The 14 patients who could not complete the study were compared with the 70 patients who completed the study on the baseline physiological measures, history of symptoms, and past exacerbations. Continuous variables were compared using t tests and discrete variables using the chi-squared test. FEV1 and PaCO2 were not normally distributed and log transformation of these variables was used in all analyses. For each patient the mean fall in PEFR over all exacerbations was calculated. MRC grade was divided into two groups, grades 2 and 3, and grades 4 and 5. The median number of exacerbations was 3 per patient and this was taken as a cutoff point to divide the patients into two groups: those who had 0, 1, or 2 exacerbations during the year and those with 3 or more. The relationship between exacerbation group and individual variables was described using univariate logistic regression. The effect of all variables together on exacerbation group was then explored using backward logistic regression. The SGRQ was normally distributed and t tests were used to compare group means for dichotomous variables which included exacerbation group, MRC grade, past exacerbations, sex, and symptoms. Physiological variables were entered into individual regression equations with SGRQ as the outcome variable and then combined in a backward multiple regression.

Description of the Cohort

Table 1 shows the baseline data on the 73 patients who remained in the study, though all further analysis was carried out only on the 70 who completed the study. Twenty-seven of the 70 patients had cardiac failure, 14 had ischemic heart disease, 38 had three or more past exacerbations, and nine were on long-term oxygen therapy. The median MRC dyspnea score was 4 (range, 2 to 5). Twenty-two were current smokers, with a mean of 46 ± 35 pack-years for the cohort. Sixty-six patients were receiving inhaled steroids (beclomethasone, budesonide, or fluticasone) using a mean dose of 1.11 ± 0.66 mg daily for 4.78 ± 3.75 yr. Ten patients were on regular prednisolone (range, 2 to 10 mg/d) and 11 on daily theophylline, 400 mg. The 14 patients who did not complete the study were not significantly different from the 70 who did, in any baseline parameters shown in Table 1. Diary card PEFR data were recorded for a mean of 307 ± 49 or for 84 ± 13% of the year study period.


Age, yr67.88.3
FEV1, L 1.050.45
FEV1, %4019
FEV1 reversibility, % 6.79.1
FVC, L 2.480.82
FEV1/FVC, %4415
PEFR, L/min23589
PaO2 , mm Hg668.3
PaCO2 , mm Hg45.67.2
Daily inhaled steroids, mg 1.110.66
Past exacerbations*  3.02.6
Daily symptoms%n
Cough and sputum4533

*Exacerbations in year prior to recruitment.


There were 190 exacerbations (mean 2.7; median 3; range, 1 to 8) in the 70 patients over the year of follow-up. Six unreported exacerbations were treated at other centers with no recording on their diary cards; these were therefore excluded from further analyses. Of the 184 exacerbations, 93 were reported, providing a reporting rate of 1.5 per patient per year. There were no differences between reported and unreported exacerbations for the major symptoms (p > 0.15). For the minor symptoms, reported exacerbations were associated with a higher incidence of increased cough (p = 0.02), but a lower incidence of increased wheeze (p = 0.03). There were no differences in PEFR fall over the 2 d prior to exacerbation (p = 0.77) nor duration of PEFR change at exacerbation (p = 0.41). There was no difference in reporting rate between the first and second 6 mo of the study (p = 0.85).

At exacerbation the PEFR fell by a median [interquartile range (IQR)] of 6.6 (−16.6, 0) L/min (p = 0.0003). Recovery of PEFR to baseline occurred after 11.1 ± 6.1 d. Seventeen patients were admitted to the hospital for a total of 23 exacerbations. There was no difference in PEFR change at exacerbation between those who were admitted for exacerbation and those who were not (p = 0.95).

Quality-of-life Scores

The mean (± SD) SGRQ Total score for the cohort was 56.7 ± 16.7 (component scores: Impacts, 43.5 ± 19.1, Symptoms 66.4 ± 19.8, and Activities 74.5 ± 17.8). The mean time between completion of the SGRQ and the preceding exacerbation was 101 ± 74 d. The Total quality-of-life score was worse in MRC grades 4 and 5 patients (p = 0.002), among younger patients (p = 0.0029), and those with frequent past exacerbations (p = 0.0014). Wheeze was the only daily symptom associated with a reduced quality of life. Blood gases and lung function showed no association with the SGRQ Total score. The Symptom score was worse for patients with frequent past exacerbations (r = 0.45, p = 0.0005), daily wheeze (r = 0.39, p = 0.001), daily sputum (r = 0.30, p = 0.011), bronchitic symptoms (r = 0.26, p = 0.03), and daily dyspnea (r = 0.27, p = 0.023). The Activities score was higher with frequent past exacerbations (r = 0.24, p = 0.047) and with MRC Groups 4 and 5 (r = 0.37, p = 0.002) and was related to FEV1 (r = −0.26, p = 0.03). The Impact score was worse in younger patients (r = −0.39, p = 0.001) frequent past exacerbations (r = 0.26, p = 0.033), MRC Groups 4 and 5 (r = 0.35, p = 0.003), and in patients with wheeze (r = 0.39, p = 0.001). Table 2 shows that SGRQ Total and component scores were all significantly higher in patients who had frequent exacerbations (p < 0.005). There was a 15-point mean difference in SGRQ Total score between the frequent and infrequent exacerbators.


Exacerbation FrequencynTotalSymptomsActivitiesImpacts
0–23248.9 ± 15.653.2 ± 17.267.7 ± 17.236.3 ± 18.2
3–83864.1 ± 14.677.0 ± 15.880.9 ± 16.050.4 ± 17.6
Mean difference−15.1−21.9−12.2−14.1
CI−22.3 to −7.8−29.7 to −14.0−21.2 to −5.3−22.9 to −5.6
p Value< 0.0005< 0.00050.0010.002

Definition of abbreviation: CI = 95% confidence interval.

*Mean ± SD.

Factors Predisposing to Frequent Exacerbations

Exacerbations were more frequent in patients with frequent past exacerbations (odds ratio [OR] = 5.5, p = 0.001), those with daily cough (OR = 3.3, p = 0.019), or daily wheeze (OR = 4.2, p = 0.011). Exacerbation frequency was also related to bronchitic symptoms (OR = 3.8, p = 0.009), but not to daily sputum production alone nor dyspnea. No other baseline factors or PEFR change at exacerbation were related to exacerbation frequency. When all factors were taken together, three factors that showed a univariate association with exacerbation frequency contributed independently to exacerbation frequency: frequent past exacerbations (OR = 1.43, p = 0.013), daily wheeze (OR = 1.34, p = 0.042), and bronchitic symptoms (OR = 1.56, p = 0.009).

SGRQ Scores and Predisposing Factors

Table 3 shows the multiple regression analyses between the SGRQ scores and the factors predictive of either a high exacerbation frequency or showing a univariate association with SGRQ. The exacerbation frequency was strongly correlated to the SGRQ Total and each of the component scores. Inclusion of time to preceding exacerbation did not affect the regression. Past exacerbations were only related to the SGRQ Symptoms score (p = 0.0428). MRC grade was significantly related to the SGRQ Total, Impacts, and Activities scores. SGRQ score was not related to PEFR changes at exacerbation nor to frequency of hospital admissions (p > 0.59 in all cases). This model accounted for 46% of the variance in SGRQ Total score and 40% of that in the Symptoms component.


R2 b* SE b* p Value
SGRQ Total score0.46
 Exacerbation frequency 11.543.260.0008
 MRC grades 4 and 5 12.683.040.0001
 Daily wheeze  7.833.480.0280
SGRQ Symptoms0.40
 Exacerbation frequency16.14.180.0003
 Past exacerbations  8.694.210.0428
SGRQ Activities0.29
 Exacerbation frequency 13.443.710.0006
 MRC grades 4 and 5 13.283.720.0007
SGRQ Impacts0.41
 Exacerbation frequency  9.183.840.0200
 Daily wheeze
 MRC grades 4 and 5 13.843.630.0003

*b = regression coefficient.

This study was designed to evaluate the effect of exacerbations on health status in COPD patients, and to determine the predisposing factors to exacerbations. Patients were monitored daily at home and asked to report exacerbations as soon as possible after onset. Health status, measured with the SGRQ, was strongly related to the number of exacerbations over the year study period and to the number of past exacerbations.

Patients with COPD develop significant disability with progressive disease, and measures of lung function such as FEV1 may not accurately predict disability (18, 19). Respiratory disease–specific questionnaires, such as the SGRQ, provide sensitive measurements of disturbance to daily life and well-being. The SGRQ has been evaluated in patients with all grades of severity of COPD, including chronic respiratory failure (2, 7, 20) and correlates quite well with exercise capacity (3, 6, 21). Although the level of the FEV1 was not predictive of exacerbation in this study, the SGRQ Total and three component scores all related strongly to exacerbation frequency. Thus a high SGRQ score may be used to predict patients at risk from frequent exacerbations. As the SGRQ was administered at the last clinic visit we cannot be certain that reduced health status is a consequence of high exacerbation frequency. Because the mean time between administering the SGRQ and the preceding exacerbation was 101 d, it is unlikely that a current exacerbation affected the response. In bronchiectatic patients, SGRQ scores were also found to be related to the exacerbation frequency (22). In our study the MRC dyspnea score was strongly related to SGRQ scores, but not to exacerbation frequency; thus indicating that while dyspnea is a weak discriminating factor for frequent exacerbations, it is associated with reduced quality of life. PEFR changes at exacerbation, though significant, were not related to SGRQ scores.

Patients were asked to report exacerbations according to worsening of symptoms. However, despite considerable instruction and clinic visits, only 50% of the exacerbations were reported. The reporting rate in this study is comparable to a study in children with asthma, where 33% of exacerbations were reported (23). Patients with COPD are accustomed to frequent symptom changes and thus may tend to underreport exacerbations to physicians. These patients have high levels of anxiety and depression and may accept their situation (7), though there is no evidence that unreported exacerbations led to increased morbidity. There were no differences in any of the major symptoms, exacerbation frequency, duration, or peak flow changes between reported or unreported exacerbations, though reported exacerbations were associated with changes in the minor symptoms; a greater incidence of cough and less wheeze. We found no differences in PEFR changes in those presenting with cough and/or wheeze, and thus for analysis, combined reported and unreported exacerbations. The tendency of patients to underreport exacerbations may explain the higher total rate of exacerbation of 2.7 per patient per year detected in this study as compared with that reported by Anthonisen and coworkers (15) where the exacerbation rate was 1.1 per patient per year and unreported exacerbations were diagnosed from patients' recall of symptoms.

A previous study of acute infective exacerbations of chronic bronchitis found that one of the factors predicting exacerbation was the number in the previous year (24). However, this study was limited to exacerbations presenting with purulent sputum only, without bacteriological evidence of infection and no baseline or exacerbation-related physiological data was available. We also found that patients with frequent past exacerbations were more likely to have an exacerbation during the study year. Patients who had chronic cough or wheeze also tended to have a higher exacerbation frequency, though frequency was not related to chronic sputum production. Patients with chronic productive cough may be more susceptible to infection, or to environmental factors such as air pollution, temperature changes, or humidity.

Coexisting cardiac and pulmonary disease has been previously identified as the only other risk factor for exacerbations (24). We found no relationship between congestive cardiac failure, ischemic heart disease, and exacerbation frequency, but this may be due to the relatively small number of patients with cardiac disease in this study. However our study revealed a strong relationship between bronchitic symptoms and exacerbation frequency. Jousilahti and colleagues (25) found that the presence of bronchitic symptoms predicted the risk of coronary artery disease independently of other known risk factors. Thus there may be an association between exacerbation frequency and coronary artery events in COPD patients.

In our study, only 16% of the exacerbations required hospital admission; this is representative of community-acquired COPD exacerbations. There was no relation between hospital admission and exacerbation frequency. One recent retrospective study found a relationship between poor quality-of-life measures and hospital readmission (8). However, reasons for hospital admission are complex, with patients with COPD exacerbations often admitted to the hospital late in the evolution of the disease, when usual therapy has failed. The purpose of this study was to consider COPD exacerbations at onset in the community as commonly found in clinical practice before secondary complications have occurred.

This study shows that COPD exacerbation may have important effects on health status and is a useful outcome measure in clinical studies of the disease. Reduction of exacerbation frequency would be expected to improve well-being, though this has not been formally tested in an interventional study. Health status measures show a strong relationship with exacerbation frequency and thus may be useful in determining which patients are at risk of exacerbation and associated disability. Earlier identification of patients at particular risk may considerably reduce the morbidity and mortality from complications associated with COPD exacerbations.

The writers thank Claire Evans for help in data collection, and the staff of the Respiratory Function unit of the London Chest Hospital for assistance with physiological measurements.

Supported by the British Lung Foundation and the Hedley Foundation.

1. Fletcher C., Peto R.The natural history of chronic airflow obstruction. B.M.J.1197716451648
2. Jones P. W., Quirk F. H., Baveystock C. M., Littlejohns P.A self-complete measure for chronic airflow limitation: The St. George's Respiratory Questionnaire. Am. Rev. Respir. Dis.145199213211327
3. Guyatt G. H., Berman L. B., Towensend M., Pugsley S. O., Chambers L. W.A measure of quality of life for clinical trials in chronic lung disease. Thorax421987773778
4. Ketelaars C. A. J., Schlosser M. A. G., Mostert R. T., Abu-Saad H. H., Halfens R. J. G., Wouters E. F. M.Determinants of health-related quality of life in patients with chronic obstructive pulmonary disease. Thorax5119963943
5. Jones P. W., Bosh T. K.Quality of life changes in COPD patients treated with salmeterol. Am. J. Respir. Crit. Care Med.155199712831289
6. McSweeney A. J., Heaton R. K., Grant I., Cugell D., Solliday N. N., Timms R.Chronic obstructive pulmonary disease: socioemotional adjustment and quality of life. Chest771982309311
7. Okubadejo A. A., Jones P. W., Wedzicha J. A.Quality of life in patients with chronic obstructive pulmonary disease and severe hypoxaemia. Thorax5119964447
8. Osman L. M., Godden D. J., Friend J. A. R., Legge J. S., Douglas J. G.Quality of life and hospital re-admission in patients with chronic obstructive pulmonary disease. Thorax5219976771
9. Traver G. A.Measures of symptoms and life quality to predict emergent use of institutional health care resources in chronic obstructive airways disease. Heart Lung171988689697
10. Siu A. L., Reuben D. B., Ouslander J. G., Osterweil D.Using multidimensional health measures in older persons to identify risk of hospitalisation and skilled nursing placement. Quality of Life Res.21993253261
11. Cox N. J., Hendricks L. C., Binkhorest R. A., van Heywaarden C. L.A pulmonary rehabilitation program for patients with asthma and mild chronic obstructive pulmonary diseases (COPD). Lung1711993235244
12. American Thoracic SocietyStandards for the diagnosis and care of patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med.1521995S78S83
13. Nagai A., Thurlbeck W. M., Konno K.Responsiveness and variability of airflow obstruction in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med.1511995635639
14. Pitkin A. D., Roberts C. M., Wedzicha J. A.Arterialised ear lobe blood gas analysis: an underused technique. Thorax491994364366
15. Anthonisen N. R., Manfreda J., Warren C. P. W., Hershfield E. S., Harding G. K. M., Nelson N. A.Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease. Ann. Intern. Med.1061994196204
16. Quirk F. H., Jones P. W.Patients' perceptions of distress due to symptoms and effect of asthma on daily living and investigation of possible influential factors. Clin. Sci.7919901721
17. Medical Research Council Committee on Aetiology of Chronic BronchitisStandardised questionnaires on respiratory symptoms. B.M.J.219601665
18. McGavin, C. R., S. P. Gupta, and G. J. McHardy. 1976. Twelve minute walking test for assessing disability in chronic bronchitis. B.M.J. 1: 6073;822–823.
19. Killian K. J., Summer E., Jones N. L., Campbell E. J.Dyspnea and leg effort during incremental cycle ergometry. Am. Rev. Respir. Dis.145199213391345
20. Meecham Jones, D. J., E. A. Paul, P. W. Jones, and J. A. WedzichaNasal pressure support ventilation plus oxygen compared with oxygen therapy alone in hypercapneic COPD. Am. J. Respir. Crit. Care Med.1521995538544
21. Jones P. W., Baveystock C. M., Littlejohns P.Relationship between general health measured with sickness impact profile and respiratory symptoms physiologic measures and mood in patients with chronic airflow limitation. Am. Rev. Respir. Dis.140198915381543
22. Wilson C. B., Jones P. W., O'Leary C. J., Cole P. J., Wilson R.Validation of the St. George's Respiratory Questionnaire in bronchiectasis. Am. J. Respir. Crit. Care Med.1561997536541
23. Johnston S. L., Pattemore P. K., Sanderson G., Smith S., Lampe F., Josephs L., Symington P., O'Toole S., Myint S. H., Tyrrell D. A., Holgate S. T.Community study of role of viral infections in exacerbations of asthma in 9–11 year old children. B.M.J.310199512251229
24. Ball P., Harris J. M., Lowson D., Tillotson G., Wilson R.Acute infective exacerbations of chronic bronchitis. Q. J. Med.8819956168
25. Jousilahti P., Variainen E., Tuomilehto J., Puska P.Symptoms of chronic bronchitis and the risk of coronary disease. Lancet3481996567572
Correspondence and requests for reprints should be addressed to Dr. J. A. Wedzicha, Academic Department of Respiratory Medicine, The London Chest Hospital, Bonner Road, London E2 9JX, UK.


No related items
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record