The importance of physical activity (PA) for health has been recognized and its beneficial effects have been demonstrated in primary as well as in secondary prevention of several chronic diseases.1,2

In COPD, more regular PA is associated with better quality of life3,4 and lower morbidity and mortality,5,6 and it is a predictive indicator for mortality in this type of patients.7 In addition, several studies dealing with respiratory rehabilitation (RR) programs have observed that patients who are more physically active show slower loss of lung function and fewer exacerbations of the disease.8,9

Despite this evidence, COPD patients show a clear tendency towards sedentary habits, which can be attributed to the bronchial obstruction as well as systemic repercussions10 and comorbidities that frequently accompany this pathology.11 The high prevalence of COPD12 together with the limited accessibility to RR programs in our setting means that very few patients can benefit from them. On the other hand, regular physical activity, theoretically, is within reach of the immense majority of patients.

In recent years, the importance of PA in the evolution of COPD has been emphasized13,14 and information on the transcendence of PA in this disease is growing,15 but our knowledge about the PA habits of COPD patients in their daily lives is still very limited.

The objective of this study is to learn about the daily physical activity habits of stable COPD patients outside of a respiratory rehabilitation program, also analyzing the variables associated with performing more or less physical activity.

We recruited a total of 132 patients, 91.7% men (121/132) and 8.3% women (11/132), with a mean age of 66.3±10 and a BMI of 28.13±5.5. With regards to tobacco habit, 82.4% were ex-smokers and 12.2% active smokers with a mean consumption of 52.17±26 pack-years. Mean FEV1 (% reference value) was 45±13.6; mean distance walked during the 6-min walk test was 364.5±99m. Figs. 1 and 2 show the distribution of the patients according to the severity of their COPD, depending on FEV1 and the BODE index, respectively.

Fig. 1.

COPD severity according to the SEPAR staging system.16

(0.08MB).

Fig. 2.

COPD severity according to the BODE index.17

(0.06MB).

As for physical activity, 32.6% of the patients (43/132) performed PA of less than 1000kcal/week, 38.6% (51/132) between 1000 and 3000, and the remaining 28.8% (38/132) more than 3000kcal/week (Fig. 3). Fig. 4 shows the PA levels according to COPD severity levels classified in accordance with the SEPAR staging system, while Fig. 5 demonstrates the level of PA according to the BODE index.

Fig. 3.

Proportion of patients according to physical activity output.

(0.07MB).

Fig. 4.

Proportion of patients according to level of physical activity and COPD severity.

(0.11MB).

Fig. 5.

Proportion of patients according to the level of physical activity and the BODE index.

(0.1MB).

The socioeconomic and tobacco use characteristics are reflected in Table 1. The low level of education stands out, as do a mid-to-low income level, a high proportion of retired persons and 12% of active smokers, although we did not find significant differences for any of these variables for level of PA.

Some type of comorbidity was present in 65.2% of patients (85/132). There was one comorbidity in 29.5%, two in 31.1% and more than two in 4.5%. The most frequent were: HTA in 42 (31.8%), diabetes in 32 (24.2%) and cardiopathy in 22 patients (16.7%).

Regarding pharmacological treatment, 87% of the patients received a combination of inhaled corticosteroids and long-acting beta-adrenoceptor agonists (LABA), 81% tiotropium and 72.5% the triple-association of these medications. 34.1% of the patients (45/132) received home oxygen therapy and 11.4% (15/132) non-invasive ventilation. We found significant differences neither between the 3 levels of PA and the presence or absence of comorbidities nor between the three levels of PA and the type of treatment that the patients were receiving.

Table 2 shows the data of the health questionnaires and the COPD severity classification, including the BODE index, specifically according to the level of PA. As can be seen, there is a statistically significant relationship between the better scores for general health (SF12, P=.01) and respiratory questionnaires (St. George's, P=.03), fundamentally for the subscales for activity and impact as well as the increased level of PA. Fig. 5 illustrates the decline in physical activity as the BODE index increases. We did not detect a relationship with the state of mental health (HD questionnaire) and the level of PA. By the same measure, we found a statistically significant inverse relationship between disease severity, determined by FEV1 as well as by the BODE index and the level of PA. Independently, each of the variables that make up this index, with the exception of BMI, also showed a clearly significant inverse relationship with the level of PA. In short, the most inactive COPD patients had greater bronchial obstruction (P=.02), more severe disease (P=.03), reported more dyspnea (P=.01) and walked less meters in the 6MWT (P=.02) (Table 2).

The correlation analysis showed a significant association between PAI over the previous year and the variables that reflected less COPD severity, as well as with living alone, being retired and being of a lower income level (Table 3).

Our study demonstrates that stable COPD patients who are seen in the outpatient pulmonology consultations of our hospitals, and who had not been included in any type of respiratory rehabilitation program, present a low level of physical activity. More specifically, one-third are inactive, almost 40% perform moderate exercise and only 30% have an energy output in line with the recommendations for obtaining a beneficial effect in their health and a reduction in the risk for morbidity and mortality. The most inactive COPD patients have greater bronchial obstruction, report more dyspnea, walk less meters in the 6MWT and have a poorer quality of life.

Our interest in this study was to analyze PA in a sample that was as representative as possible of stable COPD patients with different levels of severity. Thus, they were recruited from the outpatient consultations of various hospitals in the Community of Valencia, who span several degrees of health-care complexity, as reported in previous studies by our workgroup.22

The profile of these patients reveals a predominance of men, with high BMI, older mean age and either severe (50%) or very severe (19%) COPD, frequently accompanied by comorbidity (65%), usually cardiovascular. From a socioeconomic standpoint, there is a high proportion of retirees with limited income and limited education, most of whom live with either their spouse or family members. We cannot avoid the fact that 12% of these patients still continue to smoke. This description agrees with data previously published in Spain by Esteban et al.23 as well as by the extensive EIME study,24 which is considered to be a true reflection of the type of COPD patients who are seen in the pulmonology consultations. The inclusion of few women in the study, which of course was unintentional, does not enable us to draw conclusions about PA in this sex.

Although there is evidence of the beneficial effect of PA in all the stages of COPD, as reflected in the recommendations of the main clinical practice guidelines,16,25 there is little information about the PA performed by this type of patients. Therefore, and given that the severity of our patients oscillated from moderate to very severe COPD (FEV1 range from 58.6% to 31.4%), we distinguished 3 levels of PA: a group of very active patients (energy output greater than 3000kcal/week), another with an energy output in accordance with the recommendations1,2 on the minimum volume of PA that is beneficial for health at around 150–400kcal/day, and finally a group of clearly inactive patients.

Our study has confirmed that the limited PA of these patients is related with the perception of a low level of general and respiratory health as well as with several COPD severity indicators, both for the analysis of individualized variables (dyspnea, FEV1, 6MWT) and for the BODE index and its multidimensional evaluation of the disease.

In recent years, there has been a growing interest in the literature for learning about the characteristics of the PA in COPD patients. Different authors26–28 have underscored the marked daily inactivity in this type of patients, compared with that of control subjects with similar biological conditions. Likewise, they have observed that PA diminishes after GOLD severity stage II, although the clinical characteristics and the level of COPD severity did not always fully represent the degree of limitation of the daily activity of the patients.27,28

In Spain, there are two previous studies14,29 also analyzing the relationship between the level of PA and the state of health and severity of COPD patients. Our data agree with those of these authors regarding the quality-of-life variables support the fact that, in COPD patients who do not participate in a respiratory rehabilitation program, a higher level of PA is related with a better score in the general and respiratory health questionnaires. Likewise, both García-Aymerich et al.14 and Esteban et al.29 find a statistically significant relationship between the level of PA and various quality-of-life variables. Particularly, in the study by Esteban et al.,29 the patients were followed for 5years and the authors demonstrated that the health-related quality of life (HRQL) improved in those patients who increased their level of PA; meanwhile, the HRQL worsened in those who either maintained a low level of PA or reduced their level of PA over time.

Quite probably, the reasons for which PA improves the HRQL of COPD patients are based on the same multifactorial mechanisms which are attributed to the benefit of PA in the general population (improved cardiac function, improved peripheral muscle function and increased maximal oxygen consumption/intake, among others). They would also explain why, according to our results, the degree of dyspnea and the SF12 questionnaire (P=.01) are variables that show a greater degree of statistical significance related with level of PA.

Regarding the relationship between PA and COPD severity, the results of the different studies are not as homogeneous. Specifically, García-Aymerich et al.,14 unlike our results, found no relationship between the level of PA and disease severity, although their COPD patients were more severe than ours (means FEV1 35%), came from tertiary hospitals and had had several re-hospitalizations. In fact, their patients were recruited after one of these exacerbation episodes. The study by Esteban et al.29 included patients with different stages of severity (50% GOLD stage II, 44% GOLD stage III and 6% GOLD stage IV), somewhat less severe than our subjects. The authors state, in agreement with our results that a low level of PA is accompanied by greater dyspnea and greater COPD severity. Given that the quantification of the level of PA in the different studies was performed by the individualized administration of a specific questionnaire and that both García-Aymerich and we used the same one, it may be argued that perhaps the discrepancy in the results is due to the difference in the level of severity of the patients.

Certainly, there are currently more precise methods for determining the level of PA than simply obtaining data using questionnaires. Specifically, the use of pedometers and accelerometers can reflect the daily reality of patient mobility with greater accuracy. Particularly, accelerometers have been recommended for use in COPD patients as they can detect ranges of physical activity from light to intense,30 but the price of these devices is not insignificant, even more so in the context of a multi-center study in which not all participating hospitals can absorb such an investment. On the contrary, the MLTPAQ questionnaire validated for the Spanish population by Elosua et al.18,19 compiles an extensive list of possible activities carried out by the patient, for a short period (1week) as well as a long period (1year), excluding the seasonal bias and obtaining a global evaluation of the energy output performed by our patients. In fact, despite the differences in the method of determining the level of PA, our data also agree with those provided in previous studies.26–28 These authors used devices based on the incorporation of accelerometers and energy output sensors that measured the PA of the patients during the period of examination, and like us, they also detected a decrease in the PA in COPD patients from the early stages of the disease. This limitation was related with dyspnea, greater severity of the disease and with the decrease in walking capacity as evaluated by the 6MWT. Meanwhile GOLD stages III and IV were the best predictors for very inactive patients.27

Thus, despite the possible limitation of the study that can be involved in using the MLTPAQ questionnaire compared with other more precise methods for determining PA, we think that the results that we present in this study are valid and highly consistent. Furthermore, we feel that there are two aspects that should be mentioned. First of all, as ours is a multi-center study carried out by a recently-created COPD workgroup,22 it covers a wide number of hospitals with different degrees of health-care complexity, meaning that our results reflect the reality of the COPD patients usually seen in pulmonology consultations. Second, the fact that data were obtained for the PA of the patients for a short period (1week) as well as a long period (1year) enables us to obviate inaccuracies due to forgetfulness in the first case, while spanning the different seasons of the year in the second case, eliminating seasonal PA bias due to meteorological reasons.

Home oxygen therapy (HOT) and/or non-invasive ventilation (NIV) can have repercussions on patient PA levels. In our study, we found no relationship between the use of HOT or NIV and the level of PA (Table 1), although we should mention as a limitation of the study the fact that we did not know whether the patients had either a stationary or portable oxygen source.

It is known that air trapping has a limiting role in PA of COPD patients due to dynamic pulmonary hyperinflation31 and that daily PA in COPD patients is associated with it.32 In advanced COPD, dynamic hyperinflation during exercise is associated with greater dyspnea and poorer quality of life.33 Another limitation of our study is the incapability to collect this information in our patients.

One of the main features that characterizes COPD is the gradual deterioration in lung function, accentuated by exacerbations, which have been demonstrated to be a predictive factor for mortality in COPD,34 but the symptoms, functional status, measurements of the state of health and exercise capacity do not flow in a parallel manner. The complex varieties of cell, organic, functional and clinical manifestations that take place during COPD have increased the interest in unraveling the phenotypic heterogeneity of this disease.35 Likewise, we also know that performing PA is the result of a complex interaction of many factors, and that its behavior can only be analyzed from a multidimensional standpoint, in which in addition to disease severity other variables are taken into consideration, such as psychosocial and physical conditions and even cultural aspects. In fact, in our study the most inactive patients were those who lived with family members, who also had a sedentary occupation and whose income level was low. These data should stimulate us to develop more studies that help us to predict the start of the inactivity of our patients and to determine which indicators can alert us before its onset.

In the meantime, and in concluding, our study has allowed us to more thoroughly understand the habits of PA in the daily life of patients with stable COPD outside a respiratory rehabilitation program who are seen in pulmonology outpatient consultations. In these patients, there is evidence of low PA levels, which are associated with a poorer state of health and with greater disease severity.