Sedentary Behaviour and Physical Inactivity in Patients with Chronic Obstructive Pulmonary Disease: Two Sides of the Same Coin?

Abstract

Despite the growing interest in sedentarism, there is no available information on the profile of patients with COPD according to sedentary behaviour (SB) and with a detailed analysis of minute-by-minute bouts. Hence, the aims of this study were to quantify the time spent in SB, light activities and moderate-to-vigorous physical activities (MVPA) and to verify the relationship of MVPA and SB in individuals with COPD, as well as to identify the profile of those physically (in)active and (non)-sedentary. A cross-sectional study in which physical activity in daily life was objectively assessed through the use of SenseWear Pro 2 Armband (BodyMedia) during 2 consecutive weekdays, 12 h/day. Analysis was performed minute-by-minute for each day of each patient. MVPA comprised time spent >3 metabolic equivalents (MET), whereas light activities corresponded to time spent between 1.5 and 3 MET and SB to time spent <1.5 MET. A total of 137 subjects with COPD (66 ± 8years; FEV₁ 46 [31–57] %pred; BMI 26 [22–30] kg/m²) were analysed. Time spent in MVPA and SB presented strong negative correlation (r = −0.72, P < 0.001). Minute-by-minute analysis showed that patients with COPD spend most of their time in SB. SB accounted for 40% of all bouts >1 minute, whereas only 14% these bouts concern MVPA. Patients combining two positive characteristics (physically active and non-sedentary) have better clinical profile than others. In conclusion, SB is negatively correlated with MVPA in patients with COPD. Furthermore, patients classified as physically active (i.e., those who reach MVPA recommendations) in combination with a non-sedentary lifestyle present markedly better clinical conditions.

Keywords:

• Pulmonary disease
• chronic obstructive
• sedentary lifestyle
• motor activity
• exercise
• physical activity

Introduction

The level of physical activity in daily life (PADL) is the main predictor of mortality for all causes in subjects with chronic obstructive pulmonary disease (COPD) (1). Patients with COPD are considered less active in daily life than healthy older adults, and a sedentary lifestyle is related to dyspnoea and fatigue (2, 3). Furthermore, it is well established in healthy adults that physical inactivity is linked to higher risk of developing cancer, cardiovascular disease, type II diabetes and many other health effects (4).

Physical inactivity and sedentarism are not synonyms. Being physically (in)active refers to achieving (or not) a recommended amount of physical activity (PA), generally that proposed by the American College of Sports Medicine (ACSM), i.e., at least 30 min/day of moderate-to-vigorous physical activities (MVPA) in bouts of at least 10 min, during at least 5 days of the week, or 20 min in vigorous activity at least 3 times per week (4). Moreover, ACSM recommendations also suggest that individuals with worse physical condition can perform PA bouts shorter than 10 min when presenting difficulty to perform it continuously. On the other hand, sedentarism has two most common definitions: one based solely on low intensity PA (activities performed at an intensity <1.5 metabolic equivalent of task [MET]) and another which combines low intensity PA (≤1.5 MET) with large proportion of the day spent in the seated or reclined posture (5). Therefore, to be sedentary refers to spending most of the day in activities requiring minimal energy expenditure or sitting/lying, regardless whether the subject has achieved or not the minimum recommended MVPA (6–10).

Sedentary time is not considered during the subject’s period of sleep but only during awake time, which therefore includes work, leisure and housework activities. Most individuals with COPD spend their time awake in sedentary behaviour (SB) (2, 6, 8, 9, 11, 12). However, it is important to note that a subject can be considered physically active according to ACSM recommendations, but at the same time accumulate large amounts of SB by spending the whole rest of the day in marked sedentarism (11). Therefore, these two negative concepts (physical inactivity and sedentarism) may (or not) be present in the same individual.

It is worthwhile to mention that Furlanetto et al. (13) recently proposed a cut-off of objectively assessed SB with prognostic value for patients with COPD, which classifies them as severely sedentary or not. According to that study, individuals with COPD who spend >8.5 h/day in activities requiring <1.5 MET present a mortality risk more than 4 times higher in comparison to those who do not reach this cut-off (13).

Despite the growing interest in this field, there is no available study investigating the profile of patients with COPD according to the above mentioned definition of SB and with a detailed analysis of minute-by-minute bouts of MVPA or SB. Hence, the aims of this study were (1) to investigate the association between time spent in SB and MVPA in subjects with COPD; (2) to identify the pattern of bouts (i.e., frequency and duration) of continuous MVPA, light activities or SB in a minute-by-minute analysis; and (3) to identify the profile of those physically (in)active and (non)-sedentary.

Methods

Study population

This was a cross-sectional study which involved subjects with the following inclusion criteria: diagnosis confirmed by spirometry of COPD according to GOLD criteria (14), clinical stability (i.e., no infections and exacerbations within the last 3 months before the study); not performing either pulmonary rehabilitation or any kind of high intensity regular exercise training in the last year; and absence of severe and/or unstable cardiac disease and musculoskeletal co-morbidities that could interfere with the assessments. This study concerns a retrospective analysis of baseline assessment data from patients recruited during the initial assessment for admission in a outpatient-based physical training programme from a previously published study (15) and an ongoing study in the same centre (Clinical Trials: NCT01537627). Data collection occurred from 2006 until 2016, and comprised a convenience sample of patients with COPD who sought treatment due to symptoms or were referred to pulmonary rehabilitation. The study was approved by the ethics committee of the University and all patients provided informed consent prior to inclusion. Exclusion criteria were: unavailability of PADL assessment data; patients who did not achieve the pre-established minimum wearing time (at least 10 h of use per day); and occurrence of osteoneuromuscular complication or acute exacerbation during the assessment period.

Assessments

Spirometry was performed using the SpiroBankG according to the ATS/ERS guidelines (16) and reference values by Pereira et al (17). The 6-min walking test (6MWT) was performed according to the standards by Holland et al. (18) and reference values by Britto et al. (19) were used. Dyspnoea in daily life was assessed by the modified Medical Research Council scale (20).

PADL was objectively assessed using a validated (21, 22) multisensor PA monitor SenseWear Pro 2 Armband. Subjects were instructed to wear the monitor during awake time of two consecutive routine weekdays for 12 h/day (2, 23), and the average of the 2 days was used for analysis. A valid assessment day was considered a day with at least 10 h of wearing time, in agreement with the existing literature (24, 25). Time spent in SB, light activities and MVPA were analysed minute-by-minute in each day of each patient as follows: MVPA were performed at intensity >3 MET, light activities at intensity between 1.5 and 3 MET and SB at intensity <1.5MET. Meticulous minute-by-minute analysis looked for each bout of continuous MVPA, light activity or SB. A bout was considered a period of at least 1 min (or more) of continuous activity at the same intensity.

A posteriori, four groups were composed: active + non-sedentary (A + NS), active + sedentary (A + S), inactive + non-sedentary (IN + NS) and inactive + sedentary (IN + S). Physically active subjects were those who achieved >30 min/day of MVPA (4), whereas those classified as physically inactive did not reach this recommendation. In addition, sedentary subjects were those who spent ≥8.5 h/day in sedentary activities (i.e., activities of intensity <1.5METs) (13), whereas non-sedentary subjects were those who did not reach this cut-off.

Statistical analysis

Data distribution was analysed by the Shapiro–Wilk test. According to normality in data distribution, data were described as mean ± standard deviation or median (interquartile range), correlated with Pearson’s or Spearman’s coefficient and compared with one-way ANOVA or Kruskal–Wallis test with Bonferroni correction. The chi-square test was used to compare categorical variables between groups. Statistical analyses were carried out using SPSS 22 and GraphPad Prism 6.0, and the significance level was set at P < 0.05.

Results

Baseline characteristics

Out of 186 patients who were included, 49 were excluded due to unavailability of PADL assessment. Baseline characteristics of the remaining 137 patients are shown in Table 1. In general, the sample had normal-to-overweight BMI, moderate-to-severe airflow obstruction and relatively preserved exercise capacity.

Table 1. Baseline characteristics.

Characteristics
Sex (M/F)	75/62
Age (years)	66 ± 8
BMI (kg/m^2)	26 [22–30]
FEV1 (L)	1.1 [0.8–1.5]
FEV1 (%pred)	46 [31–57]
FEV1/FVC	54 [42–65]
GOLD (I/II/III/IV)	1/55/51/30
mMRC	3 [2–3]
6MWD (m)	452 [403–510]
6MWD (%pred)	83 [73–95]
BODE index	4 [3–5]

M: male; F: female; BMI: body mass index; kg/m²: kilogram per square metre; FEV₁: forced expiratory volume in first second; L: litre; %predicted: % of the predicted value; mMRC: modified Medical Research Council scale; 6MWD: 6-minute walking distance; BODE index: Body Composition, Obstruction, Dyspnoea and Exercise index.

Data were expressed as mean ± SD or median [QR].

SB, light activities and MVPA in the minute-by-minute analysis

Results presented in Table 2 show that patients with COPD spent the vast majority of time in SB and light activities, therefore spending little time in MVPA. Median wearing time of the activity monitor was 721 (718–725) min/day, corresponding to 12 h of assessment per day. In total, 41,238 bouts >1 min were identified. Noteworthy, SB bouts presented longer duration than bouts in light activities and MVPA. The frequency of bouts/day of SB, light activities and MVPA corresponded to 40%, 46% and 14% of all bouts, respectively. Moreover, the proportion of number of breaks (i.e., frequency of bouts/total time) according to each intensity was different among MVPA (90 [65–126]%), light activities (70 [57–92]%) and SB (29 [19–38]%), thus the higher the intensity of activity, the higher the number of interruptions that will occur.

Table 2. Characteristics of physical activity in daily life (PADL) in the study sample.

PADL characteristics
Wearing time (min)	721 [718–725]
Time spent in SB (min)	451 ± 125
Time spent in light activities (min)	207 ± 95
Time spent in MVPA (min)	62 ± 73
Sedentary patients (n/%)	44/32
Physically active patients (n/%)	76/56
Steps/day	4,859 [2,498–7,887]
Total energy expenditure/day (kcal)	1,174 [1,003–1,529]
Average MET/day	1.8 [1.7–2.1]
Frequency of SB bouts (bouts/day)	120 ± 40
Frequency of light activities bouts (bouts/day)	138 ± 48
Frequency of MVPA bouts (bouts/day)	35 [15–61]
Duration of SB bouts (min)	7 [5–10]
Duration of light activities bouts (min)	3 [2–4]
Duration of MVPA bouts (min)	2 [2–3]

SB: sedentary behaviour; MVPA: moderate-to-vigorous physical activity; min: minutes; kcal: kilocalories; MET: metabolic equivalents of task; bouts/day: periods of continuous activity (>1 min) at a given intensity (sedentary, light or moderate-to-vigorous).

Correlations

MVPA and SB had a strong negative correlation (r = −0.72, P < 0.001) which is depicted in Figure 1. Similarly, light activities and SB were also strongly and negatively correlated (r = −0.81, P < 0.001) (Figure 2). In addition, there was positive correlation between frequency and duration of bouts/day of MVPA (r = 0.52, P < 0.001), and similar but negative correlation for bouts/day of SB (r = −0.55, P < 0.001).

Figure 1 Correlation of time spent in sedentary behaviour and time spent in moderate-to-vigorous physical activities in patients with COPD.

Figure 2 Correlation of time spent in sedentary behaviour and time spent in light activities in patients with COPD.

Comparison among groups

Out of the 137 patients, 67 were classified as A + NS, 9 as A + S, 26 as IN + NS and 35 as IN + S. Comparisons regarding the minute-by-minute PADL analysis among the groups are shown in Table 3. Group A + NS presented higher time in MVPA, steps/day, energy expenditure and less time in SB in comparison to all the other groups. A + S and IN + NS presented similar steps/day and duration of bouts of light activities. Finally, both non-sedentary groups (IN + NS and A + NS) presented similar duration and frequency of SB bouts and light activities bouts, whereas both active groups (A + NS and A + S) presented higher duration of MVPA bouts in comparison to IN + S.

Table 3. Comparison among groups concerning variables of physical activity in daily life.

Variables	A + NS (67/49%)	A + S (9/6%)	IN + NS (26/19%)	IN + S (35/26%)	P value
Time spent in SB (min)	357 ± 85	559 ± 39*	459 ± 56*^,^†	598 ± 51*^,^†^,^#	<0.0001
Time spent in light activities (min)	254 ± 85	118 ± 38*	244 ± 58^†	112 ± 50*^,^#	<0.0001
Time spent in MVPA (min)	84 [54–139]	43 [32–50]*	16 [11–24]*^,^†	7 [3–17]*^,^†^,^#	<0.0001
Steps/day	7,895 [5,839–10,488]	3,939 [1,437–4,608]*	3,774 [2,935–4,965]*	1,946 [1,360–2,770]*^,^#	<0.0001
Total daily energy expenditure (kcal)	1,433 [1,171–1,605]	1,038 [825–1,221]*	1,150 [1,006–1,217]*	962 [792–1,108]*^,^#	<0.0001
Average MET	2 [1.9–2.3]	2 [1.8–2.1]	1.7 [1.6–1.8]*^,^†	1.6 [1.4–1.7]^†	<0.0001
Frequency of SB bouts (bouts/day)	131 ± 32	101 ± 45	142 ± 35^†	88 ± 32*^,^#	<0.0001
Frequency of light activities bouts (bouts/day)	162 ± 37	100 ± 45*	153 ± 34^†	91 ± 35*^,^#	<0.0001
Frequency of MVPA bouts (bouts/day)	61 [43–88]	35 [14–54]*	17 [13–25]*^,^†	9 [4–16]*^,^†^,^#	<0.0001
Duration of SB bouts (min)	6 [4–7]	10 [8–19]*	6 [5 –9]^†	13 [10–20]*^,^#	<0.0001
Duration of light activities bouts (min)	3 [2–4]	2 [1–3]	3 [3–4]	2 [2–3]*	0.01
Duration of MVPA bouts (min)	3 [2–4]	3 [1–7]	2 [2–2]*	2 [1–2]*^,^†	<0.0001

A + NS: active and non-sedentary; A + S: active and sedentary; IN + NS: inactive and non-sedentary; IN + S: inactive and sedentary; SB: sedentary behaviour; MVPA: moderate-to-vigorous physical activity; kcal: kilocalories; MET: metabolic equivalents of task; bouts/day: periods of continuous activity (>1 min) at a given intensity (sedentary, light or moderate-to-vigorous).

* P < 0.05 vs A + NS.

^† P < 0.05 vs A + S.

^# P < 0.05 vs IN + NS.

Table 4 shows that group A + S (although comprising a very small proportion of the sample) was non-overweight, had more severe airflow obstruction and worse dyspnoea than others, as well as had worse exercise capacity in comparison to A + NS and IN + NS. There were almost no differences between A + S and IN + NS groups.

Table 4. Comparisons among groups concerning variables of demographic data, lung function, dyspnoea in daily life, exercise capacity and BODE index.

Variables	A + NS (67/49%)	A + S (9/6%)	IN + NS (26/19%)	IN + S (35/26%)	P value
Sex (M/F)	39/28	5/4	11/15	21/14	0.51
Age (years)	64 ± 8	70 ± 7	66 ± 7	70 ± 9*	0.003
BMI (kg/m^2)	25 [21–28]	20 [19–31]	27 [23–31]	27 [23–33]*	0.04
FVC (L)	2.4 [1.8–3.1]	2.1 [1.2–2.6]	2 [1.2–2.4]	1.9 [1.5–2.3]*	0.004
FVC (%pred)	72 ± 15	62 ± 19	62 ± 15	62 ± 17*	0.01
FEV1 (L)	1.24 [0.94–1.63]	0.84 [0.67–1.34]	1.06 [0.84–1.46]	0.89 [0.68–1.08]*	0.002
FEV1 (%pred)	48 [37–58]	35 [23–49]	46 [26–55]	38 [26–54]	0.08
FEV1/FVC	54 [43–67]	51 [34–62]	55 [42–63]	50 [40–61]	0.68
mMRC	2 [1–3]	3 [2–4]	3 [3–3]*	3 [2–3]*	0.003
6MWD (m)	500 [445–535]	420 [408–435]*	437 [393–479]*	381 [340–465]*	<0.0001
6MWD (%pred)	90 [82–100]	78 [70–83]*	82 [73–92]*	72 [66–83]*	<0.0001
BODE index	4 [2–5]	3 [3–5]	5 [3–6]	5 [4–6]*	0.009

A + NS: active and non-sedentary; A + S: active and sedentary; IN + NS: inactive and non-sedentary; IN + S: inactive and sedentary; M: male; F: female; BMI: body mass index; FVC: forced vital capacity; L: litre; %pred: % of the predicted value; FEV₁: forced expiratory volume in the first second; mMRC: modified Medical Research Council scale; 6MWD: 6-min walking distance; BODE Index: Body Composition, Obstruction, Dyspnoea and Exercise index.

* P < 0.05 vs A + NS.

Discussion

The present study provides a novel and detailed analysis of bouts of time spent in sedentarism, light and moderate-to-vigorous activities. In general, patients with COPD spend the vast majority of time in SB and light activities, although a considerable proportion of patients in this sample was classified as active and non-sedentary, i.e., 56% are physically active with 36 MVPA bouts/day in average. Noteworthy, there was strong and negative correlation between MVPA and SB. Additionally, bouts of SB presented longer duration than bouts of light activities and MVPA, and there was higher frequency of bouts of SB and light activities than of MVPA. Finally, comparisons among the four groups analysed according to the presence of (in)activity and (non)-sedentarism showed that patients combining two positive characteristics (i.e., physically active and non-sedentary) had better clinical profile than others.

It has been suggested that healthy adults, spend approximately 46–59% of the day in sedentary activities and 37–46% of the day in light activities (26). In line with these results, the present study showed that patients with COPD spend approximately 62.6% of the day in sedentary activities and 28.8% of the day in light activities. In fact, it is well known that patients with COPD walk less time/day than healthy elderly without the disease (2). However, part of the current challenge in the PA field is about breaking sedentary bouts and turning this into increase in MVPA bouts or even in light activities bouts. Since a greater quantity of low-intensity PA already leads to lower risk of COPD hospitalisations (27), detailed results pertaining the frequency and duration of bouts of light intensity activity become even more relevant.

Interestingly, patients spend longer duration of bouts at SB intensity, but slightly higher frequency in bouts of light activities. Therefore, it is important to consider the proportion of breaks according to the total time spent in each intensity. For instance, MVPA breaks occurred in lower quantity but, proportionally, patients spent markedly less time performing high intensity activities. Thus, it is likely that the higher the intensity of activity, the higher the number of interruptions that will occur. Perhaps this happens because patients naturally break periods of activities due to symptoms of the disease (28). In this line of reasoning, the present study demonstrated that the intensity of activity is associated with the number of breaks since frequency and duration of bouts were positively associated for MVPA whereas negatively associated for SB.

The strong negative correlation between time spent in SB and MVPA, despite somewhat intuitive, was never previously shown and features that there is a direct link between longer time in sedentary activities and less MVPA. It is known that accumulating MVPA leads to important health benefits, whereas interrupting longer periods of SB with any mild or moderate intensity may also be beneficial (4, 29–31). These findings have helped to develop recommendations to minimise the amount of continuous time spent in prolonged sitting or reclining by breaking this SB more often (4, 32). Different strategies have been used to modify PA in patients with COPD, however leading these patients to reduce their time in sedentary activities (such as television viewing and others) in order to replace them by light activities may be a feasible first initiative (24). In fact, strategies to increase light intensity PA rather than focussing solely on increasing MVPA are a novel feature in the literature (6, 33, 34). The strong negative correlation between time spent in light activities and SB confirms the fact that the increase in light activities may reduce SB. However, the present results also highlighted that increasing the duration and frequency of MVPA bouts should not be left aside, as discussed ahead.

The present study was the first to present a comparison among four groups according to a minute-by-minute analysis of patterns of PA and sedentarism, combining these positive and negative characteristics. By performing this analysis, it was possible to show that the group of subjects combining two negative characteristics (physically inactive and sedentary) had the worst profile, whereas the group combining two positive characteristics (physically active and non-sedentary) had the best profile. Of note, group A + NS presented better results of dyspnoea and exercise capacity in comparison to group IN + NS, highlighting the importance of being physically active. On the other hand, it seems reasonable to encourage patients to reduce SB despite achieving 30 min of MVPA, since exercise capacity was also worse in the A + S group in comparison to A + NS. Previous studies using different ways to identify inactive or sedentary patients suggested that, in general, worse clinical conditions (e.g., worse airway obstruction, dyspnoea and exercise capacity, as well as higher BMI and scores in the BODE index) are associated with inactivity and/or sedentarism (35–37).

Results shown in Table 3 reinforce the message to focus on being both physically active and non-sedentary as the optimal goal for patients with COPD. Although patients who have only one positive characteristic (IN + NS and A + S) have similar number of steps/day, it is better to present the two positive characteristics since results of time in MVPA and SB, energy expenditure and also steps/day were better in the A + NS group. On the other hand, based on the present results, avoiding sedentarism (even in an inactive patient) or avoiding inactivity (even in a sedentary patient) present some benefits. For instance, there was similar pattern for duration and frequency of bouts between A + NS and IN + NS, whereas A + S presented similar pattern of MVPA bouts in comparison to A + NS. In fact, SB and physical inactivity have been separately associated with worse prognosis for patients with COPD (1, 13, 38). Therefore, an ideal goal would be to have as many patients as possible being physically active and non-sedentary; however, if this is not possible, patients may benefit from presenting at least one of these positive characteristics, avoiding the worse profile (i.e., physically inactive and sedentary).

According to the literature, the goal of performing ≥150 min of MVPA each week may not be realistic for many patients with COPD (7). Long programmes of exercise training are more effective than shorter ones in increasing PA levels, and the combination of PA coaching with the feedback of an activity monitor and behaviour change can also enhance daily activity levels in COPD (39). It is also known that patients with COPD who have a physically active loved one have a higher likelihood of being physically active themselves (40), as well as those who sleep better (i.e., non-fragmented sleep) (41). Another potential strategy to reduce SB in patients with COPD is to increase time in light activities such as dog walking and grandparenting (42). Furthermore, according to Mesquita et al. (35), the ‘couch potatoes’, a very inactive cluster of patients, may benefit from shorts bouts of high-intensity activity interspersed regularly with periods of rest or lower intensity exercise. On the other hand, the ‘busy bees’, a very active cluster, may benefit from home-based exercise combined with monthly phone calls, while strategies such as leisure walking can produce more sustainable results in these subjects (35).

Potential limitations of this study include its cross-sectional design (not allowing to infer causality) and the fact that group comparison may have been hindered by the small sample of the A + S group, although this imbalance could also be setting-dependent or even a characteristic of the COPD population in case this is replicated in larger samples in the future. Also, patients in the GOLD I classification were underrepresented in this sample. Furthermore, exacerbation history (43), presence of co-morbidities (44), daylight time (25) and climatic variations (45) were not assessed, and these outcomes could provide valuable further insights in future studies. Additionally, although 2 days of PADL assessment in cross-sectional studies might be sufficient for patients with very severe COPD (46), more days of assessment might be recommended for patients with milder disease severity. And finally, it was not possible to identify the time spent sleeping during the day (i.e., during the 12 h of assessment per day), and it was therefore regarded as sedentary activity. Despite a recent study showing that sleep disturbances can also be associated with PADL in COPD (41), it was not possible to assess sleep in this study. On the other hand, the assessment of minute-by-minute bouts of SB, light activities and MVPA is an important novel feature of this study, due to the scarce previous literature in this topic. The classification of patients in four groups according to the characterisation as physically active/inactive and sedentary/non-sedentary is also novel and provides further insight to the current literature, although different ways of classifying these features are available and may provide different findings.

Conclusion

The present study concludes that SB is negatively correlated with MVPA in patients with COPD. Bouts of SB last longer than those of light activities and MVPA, and the frequency of bouts in SB and light activities is considerably higher than bouts in MVPA. Furthermore, patients classified as physically active (i.e., those who reach MVPA recommendations) in combination with a non-sedentary lifestyle present markedly better clinical conditions.

Although sedentary time and physical inactivity are strongly correlated, it is unlikely that they are two sides of the same coin. They are different domains, with their own particularities and factors, and therefore are more likely to be two coins which closely connect with each other (please see Online Supplement).

Acknowledgements

We would like to thank all the colleagues from Laboratory of Research in Respiratory Physiotherapy for their support and assistance and the patients for agreeing to participate.

Disclosure statement

The authors have no conflict of interest to disclose.