This European Respiratory Society (ERS) statement provides a comprehensive overview on physical activity in patients with chronic obstructive pulmonary disease (COPD). A multidisciplinary Task Force of experts representing the ERS Scientific Group 01.02 “Rehabilitation and Chronic Care” determined the overall scope of this statement through consensus. Focused literature reviews were conducted in key topic areas and the final content of this Statement was agreed upon by all members.
The current knowledge regarding physical activity in COPD is presented, including the definition of physical activity, the consequences of physical inactivity on lung function decline and COPD incidence, physical activity assessment, prevalence of physical inactivity in COPD, clinical correlates of physical activity, effects of physical inactivity on hospitalisations and mortality, and treatment strategies to improve physical activity in patients with COPD.
This Task Force identified multiple major areas of research that need to be addressed further in the coming years. These include, but are not limited to, the disease-modifying potential of increased physical activity, and to further understand how improvements in exercise capacity, dyspnoea and self-efficacy following interventions may translate into increased physical activity.
The Task Force recommends that this ERS statement should be reviewed periodically (e.g. every 5–8 years).
An official ERS statement providing a comprehensive overview on physical activity in patients with COPD http://ow.ly/C6v78
Chronic obstructive pulmonary disease (COPD) is a highly prevalent chronic respiratory disease affecting about 10% of the adult population above 40 years of age . In addition to progressive chronic airflow limitation, patients with COPD commonly have multiple extrapulmonary effects and comorbidities, which are associated with physical inactivity [2, 3]. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) has recommended regular physical activity for all COPD patients . However, the clinical relevance of regular physical activity has not been addressed in depth. The purpose of this official European Respiratory Society (ERS) statement is to highlight the existing science regarding physical (in)activity in patients with COPD, including, but not limited to, its prevalence, determinants, consequences, measurement, and potential treatment.
An international group of pulmonologists, physiotherapists, movement scientists, exercise physiologists, health psychologists, social psychologists, and epidemiologists knowledgeable in the area of physical activity and/or COPD research was assembled into a Task Force by the leadership of the ERS Scientific Group 01.02 “Rehabilitation and Chronic Care”. Contributors searched the scientific literature (PubMed and the Cochrane Library) for original studies and systematic reviews relevant to the topic. Selection of relevant studies and reviews was based on the expertise of the contributors. Draft contributions were shared among committee members, reviewed and revised iteratively. Members were vetted for potential conflicts of interest according to the policies and procedures of ERS. This document represents the consensus of these Task Force members.
Physical activity can be defined as any bodily movement produced by skeletal muscles that results in energy expenditure . Physical activity is a complex behaviour that can be characterised by type, intensity, duration, patterns and symptom experience. Exercise is a subset of physical activity. Exercise is physical activity that is planned, structured, repetitive and purposeful . Physical activity also includes, but is not limited to, leisure-time, domestic and occupational activities [4, 5]. Activities of daily living are another subset of physical activity and this term refers to a set of basic, everyday tasks required for personal self-care and independent living [6, 7].
While physical inactivity can be defined simply as “an absence of physical activity” , it is commonly used to represent a level of physical activity that is below an optimal or specified threshold. This concept rests on the strong evidence that lower levels of physical activity are related to poor health and predict negative health outcomes . In general, a healthy individual can be considered physically inactive if one of the following criteria is not met: 1) 30 min of at least moderate-intensity physical activity on ≥5 days every week; 2) 20 min of vigorous-intensity physical activity on at least 3 days every week; or 3) an equivalent combination, which can also be accumulated in shorter bouts usually lasting 10 min of moderate (three times 10 min) or vigorous (two times 10 min) exercise [10, 11]. This represents a framework on which to base recommendations for physical activity promotion. However, the recommended intensity and duration of physical activity for the elderly may be different . Furthermore, the extent to which these recommendations apply to persons with COPD is currently unknown.
In recent years, attention has been raised regarding the adverse health effects of a sedentary lifestyle . A sedentary lifestyle is characterised by behaviours that do not increase skeletal muscle energy expenditure substantially above the resting level . Sedentary individuals expend less than 10% of their total daily energy expenditure in performing moderate- or high-intensity activities .
Consequences of physical inactivity in the general population and in chronic diseases other than COPD
Physical inactivity is a fundamental characteristic of many chronic diseases, both as a cause and as a consequence. Evidence suggests that reduced physical activity predisposes to greater incidence of cardiovascular disease [16, 17], obesity , diabetes [19, 20], cancer , dementia  and physical disability . In addition to its effect on development of chronic disease, physical inactivity may develop or worsen as a result of many diseases, owing to the sequelae of the disease and/or the associated reduction in physiological reserve. These effects of physical inactivity to both potentiate and develop from chronic disease probably explain the fact that reduced physical activity, measured either subjectively or objectively, is associated with higher overall mortality rates in the elderly [24, 25]. The exact mechanisms whereby physical inactivity interacts with anatomical and physiological changes related to ageing and other pathological cofactors to contribute to the evolution of disease or subsequent morbidity or mortality, however, is poorly understood within the human population.
Consequences of physical inactivity on lung function decline and COPD incidence
Several population-based epidemiological studies have assessed the longitudinal effect of regular physical activity on lung function decline or COPD incidence [26–30]; their main characteristics and results are summarised in table 1. Briefly, all studies show an inverse relationship between physical activity levels and the magnitude of lung function decline in at least one of the population subgroups or physical activity variables studied. However, the association between lower physical activity levels and faster lung function decline is not consistent across all population subgroups or physical activity variables. Potential explanations for such inconsistencies include selection bias, lack of adjustment for potential confounders and lack of consideration of changes in physical activity level during follow-up. Only one study overcomes such limitations and, interestingly, shows beneficial effects of regular physical activity on lung function decline and COPD risk in active smokers but not in former or never-smokers [29, 30].
Physical activity assessment
The physical activity of patients with COPD can be assessed using questionnaires. These instruments are commonly used in epidemiological studies and large clinical trials because they are inexpensive and easy to use . A variety of questionnaires exist that capture different aspects of physical activity such as amount, type, intensity, symptom experience and limitations in the performance of “activities of daily living” . The selection of a questionnaire to measure physical activity requires that the specific questionnaire fits the study aim, is properly developed and has strong psychometric properties, for example, validation, test-retest reliability and responsiveness to change . In the specific case of COPD studies, additional criteria could be considered to improve validity of the questionnaire, such as the inclusion of information on low-intensity activities [31, 33], or the availability of a version for interviewer-based administration . Other practical issues include the availability of a culturally validated version, the time required for questionnaire administration and the ability to compare outcome levels across studies.
Recent systematic reviews have assessed all questionnaires available to measure physical activity in the elderly or chronically ill patients [35, 36]. From the 104 questionnaires identified in these systematic reviews, 15 were developed for use in patients with COPD. Validity was assessed in 85% of these instruments, test-retest-reliability in 69% and responsiveness in only 19%, and none of the instruments was based on a conceptual framework . A current Innovative Medicines Initiative project is filling this gap by developing a valid patient reported outcome tool (PROactive) capturing physical activity experience in COPD (www.proactivecopd.com).
A common methodological issue is recall bias, which may become a limitation if not addressed in the development and validation process. Garfield et al.  assessed four questionnaires against physical activity measured directly by accelerometry and found that while the Stanford 7-Day Physical Activity Recall questionnaire could identify patients at both extremes of physical activity the other three had a poor relationship with directly measured activity. Also, in two other studies the relationship between questionnaire-derived physical activity measures and accelerometer-derived physical activity measures was either not given  or did not allow the authors to reliably identify extremely inactive patients .
Despite limitations of questionnaires when used on an individual level, some questionnaires might be used to measure physical activity in groups of patients with COPD. However, the choice depends on matching the question to be addressed with the psychometric properties of the instrument.
Pedometers are small, lightweight, portable and nonintrusive devices which measure the number of steps performed in a given period of time. From that metric estimates of distance and energy expenditure can be made . Many pedometers are available, and variability exists not only in cost but also in mechanism of step detection, data storage and sensitivity. Pedometers are most accurate at step counting, less accurate in distance estimates, and even less accurate at estimating energy expenditure . Pedometers may underestimate the number of steps and energy expenditure during walking at slow speed, which is typical in patients with COPD [42–44]. This may limit their accuracy in patients with moderate-to-very severe disease. However, pedometers may have positive role as a motivational tool aiming to increase daily activity, especially in addition to other combined interventions [45–47].
Accelerometers are portable electronic devices that are worn on the body to detect acceleration and thereby reflect bodily movement. They quantify activity counts, and may provide an estimate of time spent above or below a pre-specified activity level, number of steps and energy expenditure . The use of accelerometers has received increasing interest since they add objective data which cannot be obtained from questionnaires or pedometers.
Accelerometers can detect movement along one axis (uni-axial accelerometers), two axes (bi-axial accelerometers) or three axes (tri-axial accelerometers). Uni-axial devices provide information similar to pedometers, but with the advantage of assessing acceleration in addition to simply detecting steps. Bi-axial and tri-axial devices allow for detection of movement in a wider range of physical activities and are, therefore, more sensitive than uni-axial devices . Activity monitors sometimes combine accelerometers with other physiological sensors (e.g. heart rate or skin temperature) or are used in conjunction with positioning systems with the objective of increasing their accuracy to estimate daily physical activity and energy expenditure [49–51].
The validity of activity monitors for assessment of physical activity in patients with COPD has been the subject of many investigations in recent years [31, 39, 42, 43, 49, 52–61]. Two studies in COPD patients recently investigated the validity of six widely used accelerometers in comparison to the “gold standards” of indirect calorimetry and doubly-labelled-water [62, 63]. Among the six devices, the DynaPort MiniMod (McRoberts BV, the Hague, the Netherlands), the Actigraph GT3X (Actigraph, Pensacola, FL, USA) and the SenseWear Armband (BodyMedia, Inc., Pittsburgh, PA, USA) (all employing triaxial accelerometers) were valid and responsive for use in COPD [62, 63]. These devices were demonstrated to be valid in other studies in COPD as well [42, 43, 49, 52, 54–56, 58].
Several factors may influence the outcome of physical activity monitoring. Vibration from vehicle travel can falsely elevate activity counts measured by some devices, although this may be reduced by filtering the accelerometer signal . The number of assessment days and hours of use per day are also important factors that may influence the reliability of the physical activity assessment [65, 66]. Interestingly, compared with the other days of the week Sundays seem to be days of less physical activity in GOLD stage I to III patients . A recent study demonstrates that for cross-sectional analyses 2 to 3 days are sufficient for reliable measurement of physical activity in GOLD stage IV patients, whereas up to 5 days of measurement are required in patients with GOLD stage I . For measurements that aim to assess longitudinal changes 4 days were shown to be sufficient to demonstrate treatment effects following pulmonary rehabilitation in moderate-to-severe COPD, when weekends were excluded from the analysis .
The use of accelerometers certainly presents limitations. It should be noted that there is little uniformity in output from the various types of accelerometers, which makes it difficult to compare studies using different devices . Another limitation is that estimates of energy expenditure for individual patients may be inaccurate, especially among those with functional limitations due to chronic diseases that affect walking speed and efficiency of movement . Furthermore, purchase costs vary considerably between devices.
Doubly labelled water
The doubly labelled water (DLW) method provides an indirect assessment of total energy expenditure by the body over a substantial period of time (e.g. 2 weeks). The technique has been described elsewhere . Briefly, known doses of deuterium and 18O are ingested (2H2O and H218O). The deuterium washes out of the body through the urine, whereas the 18O is eliminated as urine water and CO2. The difference between the wash out of the two (typically measured in urine samples) provides an estimate of CO2 production by the body, which can be converted to energy expenditure .
The biggest drawback of the technique in the context of physical activity assessment in COPD is that it does not allow separation of energy expenditure linked to physical activity and energy expenditure linked to basal metabolic rate or diet-induced metabolism. Thus, although the DLW technique is used to estimate total energy expenditure in patients with COPD [63, 72, 73] its ability to estimate physical activity is compromised by a number of assumptions that may be correct for healthy subjects but not for patients with COPD. For example, one study used DLW in severe COPD and concluded that active energy expenditure was even higher in COPD compared with controls, which might be related to the increased oxygen cost of breathing and decreased mechanical efficiency in patients with COPD . By contrast, actual physical activity levels are lower in patients with COPD . The DLW technique should probably be restricted to questions about caloric balance in patients with COPD rather than questions concerning the amount and intensity of physical activity. It can be questioned whether the technique should remain the “gold standard” to validate monitors that measure the amount and intensity of movement in COPD, even though DLW will very likely remain a “gold standard” for measuring the caloric cost of physical activity.
Levels of physical activity in patients with COPD
Patients with COPD have significantly lower levels of physical activity as compared with healthy controls [65, 74–81]. Existing data show that time spent walking is significantly lower in COPD patients compared with healthy, age-matched persons [74, 76, 77, 80, 82]. These findings appear to be consistent across settings, cultural background, geographic area and methods used to measure physical activity. In addition, movement intensity of patients with COPD is lower compared with age-matched healthy subjects, which indicates that patients with COPD walk at a slower pace . Current data suggest that patients with COPD reduce their physical activity early in the course of the disease [83–85]. Accordingly, most patients do not meet currently recommended physical activity levels. For example, in one study only 26% of 177 patients with a mean forced expiratory volume in 1 s (FEV1) 52% predicted achieved at least 30 consecutive minutes of moderate intensity activity on at least 5 days; this increased to 50% if the 30 min was accomplished in bouts of at least 10 min each . Corroborating these results, another study reported that only 29% of 73 patients with COPD achieved a mean of at least 30 min of moderate physical activity summed up throughout the day . Compared with controls, COPD patients had a reduction by 50% of their minutes of moderate physical activity per day, which was even more reduced when bouts of at least 10 min were compared between controls and COPD patients .
Factors associated with physical activity in COPD
This section addresses associations between physical activity and clinical characteristics of patients with COPD, such as disease severity, comorbidities, exacerbations and behavioural factors. Since most of the studies are cross-sectional it is not possible to draw conclusions regarding the directionality of the established associations . It should be acknowledged that, in general, physical activity is dependent of many factors, which include biological, behavioural, genetic, social, environmental, cultural and policy factors . In this section we will discuss those aspects that are specifically studied in COPD.
FEV1 shows a weak-to-moderate positive association with objectively measured physical activity in patients with COPD (table 2) [39, 65, 74, 77, 79, 80, 91, 92]. In general, FEV1 explains only a small proportion of the variation in physical activity in subjects with COPD. Directly measured maximal voluntary ventilation may provide an increased correlation with physical activity in this population . Fewer studies have investigated the relationship between physical activity and other lung function measures. While three studies found weak-to-moderate positive associations between physical activity and diffusion capacity [74, 80, 93], one study showed a strong and independent linear association . A robust inverse association was found between dynamic hyperinflation (measured in the laboratory during cardiopulmonary exercise test on a stationary cycle ergometer) and physical activity . Overall, while increasing severity of lung function impairment is associated with reduced physical activity in subjects with COPD, the relationship is relatively weak. Therefore, levels of physical activity cannot be accurately predicted from resting lung function parameters.
Physical fitness, which can be measured by various exercise tests, comprises a set of attributes that relates to the ability to perform physical activity . Accordingly, most studies in COPD have found moderately positive associations between either 6-min walking distance or peak work rate in an incremental exercise test and objectively measured physical activity (table 3) [40, 65, 74, 79, 96].
Two studies evaluated the predictive power of the 6-min walking distance to identify physically inactive COPD patients with an objectively measured physical activity level lower than 1.4 (i.e. <40% of total daily energy expenditure is related to physical activity). In both studies 6-min walking distance, even though moderately associated with physical activity, was found to be of limited value to reliably identify physically inactive COPD patients [40, 65].
Self-efficacy, an individual’s belief in his or her capability of performing a specific task in a specific situation, is influenced by expectations of ability to perform the task and its outcome . Theoretically, higher levels of self-efficacy may be associated with increased physical activity; and higher levels of physical activity may result in enhanced self-efficacy belief. However, self-efficacy for walking, assessed in COPD using the Self-Efficacy Questionnaire – Walking (SEQ-W) instrument, was only weakly associated with objectively measured physical activity . Furthermore, in another study of 165 patients with COPD, general self-efficacy as measured with the General Self-Efficacy Scale (SES6) was not associated with physical activity .
Sociodemographic factors and environment
Several sociodemographic and environmental factors, including ethnicity, socioeconomic status, job availability, education level, patient choices about where to live, and seasonal variations in temperature and humidity, have the potential to influence performance of daily physical activity among persons with COPD.
Among healthy adults, lower physical activity levels have been associated with lower socioeconomic status, lower education level and non-Caucasian race [99–101]. However, this may not be the case for patients with COPD, since two studies have demonstrated a relationship between lower physical activity levels and higher socioeconomic status [102, 103]. In these studies, it is difficult to distinguish the impact of socioeconomic status from that of other influences, such as geographical location and ethnic and cultural differences, but it is possible that increased dependence on walking and public transportation among persons of lower socioeconomic status accounts for the findings. Also, both studies focused on persons with severe airflow limitation; hence the findings may not be applicable to patients with less severe lung function impairment.
Weather, climate and altitude conditions can also influence physical activity levels among persons with COPD. Extremes of heat or cold and/or high environmental levels of particulates or other air pollutants can trigger increased symptoms, bronchoconstriction and acute exacerbations [104–106] and may pose a barrier to exercise and activity adherence [107, 108]. Accordingly, seasonal variations in daily physical activity have been reported [93, 109–111], with a tendency for lower activity during periods with lower temperature. High altitude environments may also influence physical activity levels, since elevation in altitude leads to worsening resting and/or exercise hypoxaemia and reduced exercise tolerance among patients with COPD . Geographical location per se may not strongly influence activity levels (e.g. independently of other factors such as climate, altitude or socioeconomic status) as no significant differences in physical activity were observed among patients with varying severity of COPD across diverse geographical locations in Europe and the USA [78, 79]. Finally, physical activity levels may be influenced by the day of the week; total activity and activity intensity were lower on Sundays or during the weekend as compared with other days [63, 65].
Exacerbations of COPD
Physical activity is dramatically reduced during and after hospitalisation due to an exacerbation of COPD [113, 114]. Furthermore, recovery time is prolonged over several weeks and physical activity may not return to pre-exacerbation activity levels [113, 114].
Even patients with milder exacerbations, which do not require hospitalisation, tend to stay indoors during the exacerbation period . Moreover, patients with a history of frequent exacerbations reduce their time spent outdoors at a faster rate compared with those with infrequent exacerbations, and thus are more likely to become housebound . This was confirmed by another study demonstrating that a history of more than one exacerbation was correlated with lower physical activity levels .
Comorbidities are common in COPD [116–118] and may independently impact physical activity levels. A cross-sectional study of 170 patients with COPD demonstrated that left ventricular cardiac dysfunction (assessed by elevated N-terminal pro-brain natriuretic peptide and echocardiographic assessment of diastolic function) was associated with reduced physical activity levels, independent of COPD severity assessed by GOLD stage or the multidimensional BODE (body mass index, airflow obstruction, dyspnoea, exercise capacity) score . In this study, depression, anaemia, systemic arterial blood pressure and nutritional depletion were not associated with reduced physical activity levels . In the same cohort, physical activity levels were significantly lower among patients with metabolic syndrome and COPD across all GOLD stages as compared with the patients without metabolic syndrome . It is not yet clear which of the components of metabolic syndrome contributed to this finding. For instance, obese COPD patients have lower activity levels compared with underweight and normal weight patients [121, 122]. Furthermore, diabetes is strongly associated with inactivity in COPD, independent of other confounders . In a cohort of newly diagnosed COPD patients, physical inactivity was more strongly associated with the presence of comorbidities compared with airflow obstruction .
Quadriceps muscle strength and mass, which are commonly reduced in patients with COPD, were positively correlated with physical activity in several studies [74, 79, 85]. In one study quadriceps muscle strength was predictive of physical activity, independent of FEV1 . By contrast, handgrip strength did not correlate with physical activity in patients with COPD [94, 119].
Mood disturbances such as symptoms of anxiety and depression are highly prevalent in patients with COPD . Of the six studies examining the relationship between physical activity and depression in patients with COPD [79, 80, 119, 125–127], only two found significant associations [80, 127]. In one study higher levels of anxiety were associated with higher levels of physical activity, whereas more depressive symptoms were associated with lower physical activity only when anxiety was part of the statistical model .
In the general population a growing body of evidence demonstrates that physical activity, inflammation and immunity are tightly linked, with regular moderate physical activity reducing systemic inflammation [128–131]. Potential mechanisms underlying these observations include the release of anti-inflammatory myokines by contracting skeletal muscles . In COPD, four studies have demonstrated that higher levels of low-grade systemic inflammation are associated with lower levels of physical activity even after adjusting for relevant confounders [79, 94, 119, 133]. Whether anti-inflammatory mechanisms of regular physical activity or COPD-specific mechanisms of systemic inflammation underlie the associations between systemic inflammation and physical activity in COPD needs to be elucidated in further longitudinal studies .
Various assessments of health status have been related to physical activity measurements in patients with COPD. Most studies showed that an impaired health status assessed either by generic or disease-specific instruments is weakly to moderately related to a lower amount and intensity of physical activity [79, 91, 102, 135, 136]. Changes in physical activity over time parallel trends in health status . Since most health status questionnaires have items addressing physical activity this could obviously affect these relationships.
Breathlessness on exertion is the primary symptom limiting exercise, which in turn leads to reduced physical activity in patients with COPD . Accordingly, patients’ perception of breathlessness on exertion is reported to be a barrier for participation in exercise and activities of daily living [110, 139]. Indeed, greater levels of dyspnoea as measured by the modified Medical Research Council dyspnoea scale are related to lower levels of physical activity in patients with COPD [65, 79].
Fatigue is also reported to be a frequent symptom in patients with COPD . There is one study that evaluated the association of fatigue assessed by the Functional Assessment of Chronic Illness Therapy – Fatigue questionnaire and physical activity. After adjusting for several confounders physical activity was demonstrated to be associated with fatigue in patients with COPD . Whether, and to what extent, pain symptoms may affect physical activity in patients with COPD remains currently unknown.
Effect of physical inactivity on hospitalisations in patients with COPD
To date, seven prospective longitudinal studies have assessed the association between levels of regular physical activity and hospital admission or readmission due to COPD exacerbations [30, 113, 141–145]. All studies consistently showed a statistically significant association of low physical activity levels with increased risk of hospitalisations. Although all studies reported associations adjusted for confounders, just one study adjusted for previous COPD hospitalisations, which is one of the most important risk factors for readmission . Importantly, the studies have shown that the amount of regular physical activity needed to obtain a significant effect on admissions due to COPD is relatively small, equivalent to walking or cycling for 2 h per week [30, 113, 141–145].
Effects of physical inactivity on survival in COPD
Three longitudinal studies with a follow-up of 3–12 years have demonstrated that low levels of physical activity predict all-cause mortality in patients with COPD after controlling for relevant confounding factors [144, 145, 147]. The relationship was consistent across differing settings, patient characteristics and methods used to measure physical activity. In one study that separated respiratory-related from cardiovascular and all-cause mortality, the greatest effect was on the former . Recently, an assessment of physical activity has also been included as a prognostic factor in a multidimensional prognostic score for stable COPD patients .
Whether, and to what extent, improvements in physical activity levels may lower the risk of dying remains unknown in patients with COPD. It seems reasonable to hypothesise that patients who have a decline in physical activity over time have a worse prognosis compared with those patients who remain physically active.
Treatment strategies to improve physical activity
While it is well-known that bronchodilators improve dyspnoea and exercise tolerance, only a few studies have investigated the impact of bronchodilator therapy on physical activity [149–151]. Two studies had positive results with regard to improvements in physical activity following bronchodilator treatment. Of these, one study (23 patients) was a nonrandomised open-label study of a long-acting β-agonist in a small number of patients , while the other study was a retrospective subgroup analysis . By contrast, two randomised, placebo-controlled, multicentre studies could not demonstrate a change in physical activity following therapy with a long-acting bronchodilator [149, 152]. Therefore, it remains currently unknown whether bronchodilator therapies that are known to improve exercise capacity will also either improve physical activity or prevent deterioration of physical activity over time in patients with COPD.
Ambulatory oxygen therapy
Oxygen therapy improves exercise tolerance in hypoxaemic COPD patients, but whether activity levels are also enhanced is unclear. A small, randomised trial of replacing heavier oxygen tanks with lightweight ambulatory oxygen therapy failed to detect improvements in activity monitor assessments of physical activity over a 6-month period .
Pulmonary rehabilitation is a “comprehensive intervention to improve the physical and psychological condition of people with chronic respiratory disease and to promote the long-term adherence to health-enhancing behaviours” . Pulmonary rehabilitation has been demonstrated clearly to reduce dyspnoea, increase exercise capacity and improve quality of life in patients with COPD. Increases in exercise capacity in combination with behavioural change may also have the potential to increase physical activity in patients with COPD. Despite this rationale, the 10 studies that evaluated the effect of pulmonary rehabilitation on physical activity [80, 82, 155–162] have yielded inconsistent results; four showed an increase in physical activity [80, 160–162] and six failed to do so [82, 155–159]. In the negative studies, the lack of improvement in physical activity was observed despite a concomitant increase in exercise capacity and quality of life. A systematic review and meta-analysis of single-group and randomised trials of the effect of exercise training (not necessarily pulmonary rehabilitation) on physical activity concluded that this intervention conferred a significant, but small, increase in this outcome .
Behaviour changes, feedback, counselling
Solely increasing the exercise capacity of patients with COPD may be insufficient to increase participation in self-directed leisure time activity [47, 80, 82, 155, 157–162, 164]. Widespread acknowledgement of the exercise maintenance problem has led to the identification of behavioural factors related to participation in daily physical activity and to the development of interventions targeting these factors.
Interventions including self-monitoring of activity behaviour using activity monitors in combination with behavioural counselling in patients with COPD might have the potential to change physical activity behaviour [45–47, 165–167]. However, the evidence base is weak since only a few studies have been conducted using small samples [46, 47], often without a control group [45, 165, 166]. Larger studies with long-term follow-up are needed in patients with COPD.
Key components that increase the effectiveness of behavioural interventions have already been summarised in several meta-analyses and international guidelines [168–172], and include mobilising social support, using well-described/established behaviour change and self-regulatory techniques (self-monitoring, stimulus control, problem solving, relapse prevention management, goal setting, self-reinforcement, providing feedback on performance and developing action plans), providing higher contact time or contact frequency, and assessing the readiness/motivation to change [168, 170]. Motivational interviewing techniques have further been recommended as a collaborative communication approach .
Summary of treatment strategies to improve physical activity
To date, only a few randomised controlled trials have studied the efficacy of pharmacological or non-pharmacological treatment strategies on daily physical activity in patients with COPD. Therefore, there is an urgent need for additional well-designed trials. Based on the correlates of physical activity, it seems reasonable to focus in future studies on physical, non-physical and environmental factors to improve physical activity in patients with COPD. Moreover, future trials should not only focus on improvements in physical activity but also on the prevention of loss of physical activity.
The scientific foundation regarding the clinical importance of assessing and improving physical activity in patients with COPD has grown considerably in the past decade. Nevertheless, many questions remain unanswered and the methodology of physical activity assessment needs to be further standardised. Therefore, this Task Force identifies the following major areas that need to be further addressed in the coming years.
1) In addition to smoking cessation the disease-modifying potential of increased physical activity in smokers without COPD and in patients with all degrees of airflow limitation should be explored.
2) Further understanding is needed regarding the concepts for optimising the impact of the pharmacological and non-pharmacological interventions that aim to maintain or increase physical activity levels in patients with COPD.
3) Research should be undertaken to understand how improvements in exercise capacity, dyspnoea, and self-efficacy following an intervention (e.g. pulmonary rehabilitation or pharmacological therapy) might translate into increased physical activity.
4) The methodology to measure physical activity needs to be further standardised and formal guidance on this should be provided in the future. Such methodologies could rely on objective and accurate assessment of physical activity, patient reported assessment of physical activity experience, or a combination thereof.
The realisation of the Official ERS Statement on Physical Activity in COPD would not have been possible without the support of Guy Brusselle (ERS Guidelines Director), and Sandy Sutter (ERS CME and Guidelines Coordinator). Moreover, the Task Force co-chairs are grateful to the ERS for funding this ERS Statement.
The authors’ affiliations are as follows. Henrik Watz: Pulmonary Research Institute at LungenClinic Grosshandorf, Airway Research Center North, German Center for Lung Research, Grosshansdorf, Germany; Fabio Pitta: Laboratory of Research in Respiratory Physiotherapy, Dept of Physiotherapy, Universidade Estadual de Londrina, Brazil; Carolyn L. Rochester: Yale University School of Medicine and VA Connecticut Healthcare System, New Haven, CT, USA; Judith Garcia-Aymerich: Centre for Research in Environmental Epidemiology (CREAL); CIBER Epidemiología y Salud Pública (CIBERESP); and Universitat Pompeu Fabra, Departament de Ciències Experimentals i de la Salut, Barcelona, Spain; Richard ZuWallack: St Francis Hospital Medical Center, Hartford, CT, USA; Thierry Troosters: Dept of Rehabilitation Sciences, KU Leuven, Leuven, Belgium, and Respiratory Division and Pulmonary Rehabilitation, UZ Gasthuisberg, Leuven, Belgium; Anouk W. Vaes: Dept of Research and Education, CIRO+, Center of Expertise for Chronic Organ Failure, Horn, The Netherlands; Milo A. Puhan: Institute of Social and Preventive Medicine, University of Zurich, Switzerland; Melissa Jehn: Dept of Pneumological Onkology and Transplantology, Charite Universitätsmedizin, Berlin, Germany; Michael I. Polkey: NIHR Respiratory Biomedical Research Unit at the Royal Brompton and Harefield NHS Foundation Trust and Imperial College, London, UK; Ioannis Vogiatzis: Medical School, 1st Dept of Respiratory Medicine, Pulmonary Rehabilitation Unit, National and Kapodistrian University of Athens, Athens, Greece; Enrico M. Clini: Dept of Medical and Surgical Sciences, University of Modena Reggio Emilia, Modena, Italy, and Ospedale Villa Pineta, Pavullo n/F, Modena, Italy; Michael Toth: Dept of Medicine, University of Vermont, College of Medicine, Burlington, VT, USA; Elena Gimeno-Santos: Centre for Research in Environmental Epidemiology (CREAL); CIBER Epidemiología y Salud Pública (CIBERESP); and Universitat Pompeu Fabra, Departament de Ciències Experimentals i de la Salut, Barcelona, Spain; Benjamin Waschki: Pulmonary Research Institute at LungenClinic Grosshandorf, Airway Research Center North, German Center for Lung Research, Grosshansdorf, Germany; Cristobal Esteban: Pneumology Dept, Hospital Galdakao-Usansolo, Bizkaia, Spain; Maurice Hayot: INSERM U-1046, University Montpellier I, Université Montpellier II, Montpellier, France, and Dept of Clinical Physiology, CHRU Montpellier, Montpellier, France; Richard Casaburi: Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Los Angeles, CA, USA; Janos Porszasz: Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Los Angeles, CA, USA; Edward McAuley: Dept of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Urbana, IL, USA; Sally J. Singh: Centre for Exercise and Rehabilitation Science, Glenfield Hospital, University Hospitals of Leicester NHS Trust, Leicester, UK, and Faculty of Health and Life Sciences, Coventry University, Coventry, UK; Daniel Langer: Dept of Rehabilitation Sciences, KU Leuven, Leuven, Belgium, and Respiratory Division and Pulmonary Rehabilitation, UZ Gasthuisberg, Leuven, Belgium; Emiel F.M. Wouters: Dept of Research and Education, CIRO+, Center of Expertise for Chronic Organ Failure, Horn, The Netherlands, and Dept of Respiratory Medicine, Maastricht University Medical Center (MUMC+), Maastricht, The Netherlands; Helgo Magnussen: Pulmonary Research Institute at LungenClinic Grosshandorf, Airway Research Center North, German Center for Lung Research, Grosshansdorf, Germany; Martijn A. Spruit: Dept of Research and Education, CIRO+, Center of Expertise for Chronic Organ Failure, Horn, The Netherlands, and REVAL - Rehabilitation Research Center, BIOMED - Biomedical Research Institute, Faculty of Medicine and Life Sciences, Hasselt University, Diepenbeek, Belgium.
Support statement: Michael Polkey’s contribution to this work was part funded by the NIHR Respiratory Biomedical Research Unit at the Royal Brompton and Harefield NHS Foundation Trust and Imperial College London, UK, who part fund his salary. Thierry Trooster’s contribution was partly funded by the Flemish Research Foundation (#G.0871.13). Anouk Vaes’ contribution to this work was partially funded by “Stichting de Weijerhorst” and Point-One funding from AgentschapNL, Dutch Ministry of Economic affairs, the Netherlands. Martijn A. Spruit’s contribution to this work was partially funded by Point-One funding from AgentschapNL, Dutch Ministry of Economic affairs, the Netherlands. Benjamin Waschki’s contribution to this work was partially funded by the German Center for Lung Research, Germany. The Task Force co-chairs are grateful to the ERS for funding this ERS statement.
Conflict of interest: Disclosures can be found alongside the online version of this article at erj.ersjournals.com
- Received March 11, 2014.
- Accepted July 29, 2014.
- ©ERS 2014