Influencing factors of sedentary behaviour in people with chronic obstructive pulmonary disease: a systematic review

Abstract

Background

People with chronic obstructive pulmonary disease (COPD) are more likely to adopt a sedentary lifestyle. Increased sedentary behaviour is associated with adverse health consequences and reduced life expectancy.

Aim

This mixed-methods systematic review aimed to report the factors contributing to sedentary behaviour in people with COPD.

Methods

A systematic search of electronic databases (Medline, CINAHL, PsycINFO and Cochrane Library) was conducted and supported by a clinician librarian in March 2023. Papers were identified and screened by two independent researchers against the inclusion and exclusion criteria, followed by data extraction and analysis of quality. Quantitative and qualitative data synthesis was performed.

Results

1037 records were identified, 29 studies were included (26 quantitative and 3 qualitative studies) and most studies were conducted in high-income countries. The most common influencers of sedentary behaviour were associated with disease severity, dyspnoea, comorbidities, exercise capacity, use of supplemental oxygen and walking aids, and environmental factors. In-depth findings from qualitative studies included a lack of knowledge, self-perception and motivation. However, sedentarism in some was also a conscious approach, enabling enjoyment when participating in hobbies or activities.

Conclusions

Influencers of sedentary behaviour in people living with COPD are multifactorial. Identifying and understanding these factors should inform the design of future interventions and guidelines. A tailored, multimodal approach could have the potential to address sedentary behaviour.

PROSPERO registration number

CRD42023387335.

WHAT IS ALREADY KNOWN ON THE TOPIC.

• The awareness and negative consequences of sedentary behaviour are increasing. But what are the influencing factors of sedentary behaviour in people with chronic obstructive pulmonary disease (COPD)?

WHAT THIS STUDY ADDS

• Influencers of sedentary behaviour are multifactorial. Some are deliberate and not necessarily perceived as unfavourable, for example, a way to enjoy a hobby.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This review provides researchers, health professionals and policymakers insight into the multifactorial contributors to sedentary behaviour in people with COPD. This knowledge could inform interventions to reduce sedentarism and guidelines in this field.

Introduction

Any waking behaviour in a sitting or reclined posture, characterised by an energy expenditure of <1.5 metabolic equivalents, such as watching television, is defined as sedentary behaviour.¹ Extended time spent in sedentary postures is associated with poorer health outcomes^{2 3} and reduced life expectancy⁴ . Sedentary behaviour is acknowledged to be a growing public health issue. Health risks (heart disease, cancer and type 2 diabetes) associated with greater sedentary behaviour have driven this increased awareness. WHO guidelines recommend reducing time spent sedentary, without quantifying an exact time limit.⁵

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide.⁶ It is estimated that globally, 300–400 million people (9%–12%) live with COPD.⁶ Acute exacerbations account for one in eight of all acute hospital admissions, with a 24% 30-day readmission rate.⁷ COPD exacerbations are accountable for 50–75% of direct healthcare costs.⁸

People with COPD spend more of their time sedentary compared with healthy adults of the same age (63% vs 46%–59%).⁹ Pulmonary rehabilitation, smoking cessation, behavioural intervention aimed at self-management and occupational therapy are the current non-pharmacological, evidence-based interventions in people with COPD.^{6 10} However, the importance of reducing sedentary behaviour is not part of the British Thoracic Society guidelines.¹¹ Previous research, including people living with COPD, has primarily focused on increasing physical activity due to convincing health benefits.¹² Of note, it has been argued that physical activity and sedentary behaviour are two different domains.^{4 9 13} Despite the ongoing debate on sedentary behaviour, it is possible for individuals to lead a sedentary lifestyle while still meeting the recommended physical activity guidelines.¹⁴ Furthermore, some people may only spend a little time being sedentary but may not engage in moderate to vigorous-intensity physical activities.¹⁴ Achieving the recommended physical activity guidelines of 150 min per week of moderate-intensity physical activity or 75 min per week of vigorous-intensity physical activity⁵ can be challenging to achieve for those living with COPD .¹² Dyspnoea, exercise capacity and airflow obstruction are associated with reduced physical activity in people with COPD.¹⁵ Since the recognition of the disease-related burden which people with COPD face, Kingsnorth et al suggested that people with COPD are likely to achieve maximal physical activity levels by performing regular daily living activities, instead of focusing on absolute intensity thresholds.¹⁶ Reducing sedentary behaviour, irrespective of the physical activity intensity undertaken, may be a more achievable goal in people with COPD .¹⁷ The most recent Global Initiative for Chronic Obstructive Lung Disease (GOLD) report (2023) mentioned that sedentary behaviour is associated with lifestyle adaptations but without further recommendations.⁶ Reducing sedentary time may deliver long-term health benefits for this population.¹⁸ Therefore, an understanding of the influencers of sedentary behaviour, especially those that are adjustable and adaptable, is essential to inform effective interventions to reduce sedentary time and improve the health of people with COPD.

To our knowledge, no systematic review of the determinants contributing towards sedentary behaviour in people with COPD exists. The objectives of this systematic review are to identify influencing factors of sedentary behaviour in people with COPD to help inform future interventions aiming to reduce sedentarism.

Methods

The systematic review protocol followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)¹⁹ and is registered with the International Prospective Register of Systematic Reviews (PROSPERO) CRD42023387335.

Inclusion criteria

Population: Included were primary studies, including adults with COPD diagnosed by a clinician, or using the GOLD criteria⁶ with forced expiratory volume in one second (FEV₁) of <80% predicted.

Language: Studies in English, German and Portuguese language were included (due to access to free translations).

Setting: People cared for in primary (community and care/nursing homes) and secondary care (hospitals).

Study design: Quantitative and mixed-method studies that measured sedentary behaviour objectively or self-reported. Qualitative data, which included themes or other qualitative data identified in the primary studies and relevant to the review question.

Exclusion criteria

Studies that were only available as an abstract/poster/conference proceeding and case studies/systematic or literature reviews and protocols.

Quantitative studies that measured sedentary behaviour but did not qualitatively explore behavioural influences or objectively test for associations or relationships with other variables.

Studies examining the effectiveness of an intervention.

Information sources

The following electronic databases were searched: Medline, via EBSCO; CINAHL, via EBSCO; PsycINFO, via ProQuest; and Cochrane Library screened for eligibility studies. The search strategy was supported by a clinician librarian. Additional records were identified via hand search. The searches were completed on March–April 2023. A keyword search included: COPD, chronic obstructive pulmonary disease, emphysema, chronic bronchitis, sedentary behaviour, physically inactive, sitting, lying and reclining (see online supplemental materials).

Selection process

One researcher (SH) assessed eligible studies and extracted data from study reports. The retrieved records were imported into a Reference Manager.²⁰ Two researchers (SH and LH) screened all titles and abstracts and then reviewed full texts independently. All identified studies were discussed, disagreements were resolved and a third review author was not required.

Data extraction

Data were extracted to predesigned data extraction forms into the following categories: publication characteristics, participant characteristics, study characteristics and findings. Results from the eligible studies are represented in.

Risk of bias

As qualitative and quantitative studies were included in the search, the mixed-methods appraisal tool (MMAT) was used to help appraise the methodological quality .²¹ Studies were rated ‘yes’, ‘no’ or ‘can’t tell’ in seven categories. Before the qualitative assessment, two researchers (SH and LH) agreed on which categories were important to consider and were applied across all included studies (see online supplemental materials). As per MMAT recommendation, low methodological quality studies were not excluded; instead, a sensitivity analysis approach during the synthesis stage involved interpreting the impact of their findings during the synthesis.²¹

Data analysis

Due to the included studies’ methodological heterogeneity, the researchers applied a narrative synthesis approach, following the methodological guidance for conducting mixed-methods systematic reviews.²² Data are presented in. The study adopted a convergent integrated approach, which involved transforming qualitative and quantitative data into the same, mutually compatible format to address the same research question. Therefore, quantitative data were extracted and converted into a textual description to combine with qualitative data from the qualitative studies. This process is known as ‘qualitising’ and involves a narrative interpretation of the quantitative results. The combined data were then repeatedly and thoroughly inspected to identify any similar factors.²²

Results

Study selection

A total of 1037 records were identified, 281 duplicates were removed and 727 records met exclusion criteria (PRISMA flowchart, figure 1, summaries of studies.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart of study selection.

Study characteristics

The 29 included studies were published between 2008 and 2022 and conducted in 13 countries across 5 continents. Most included studies (n=21) were published in the last 6 years. 15 studies did not record the COPD severity proportion within their study, and 5 did not report the severity of the whole sample, as per recommended guidelines .⁶

Out of the 29, 2 studies included participants following a COPD-related hospital admission; the remaining 27 studies included people with stable COPD. The majority (23/29) were conducted in high-income countries (Australia, Austria, Canada, Denmark, Germany, Netherlands, Norway, Singapore, Spain, Sweden, UK and USA) and 6 studies in upper-middle-income countries (Brazil). The majority (n=26) were quantitative studies and three qualitative studies.

Measurement of sedentary behaviour

Most studies (n=23) used an accelerometer or inclinometer (SenseWear Armband n=4, ActiGraph n=8, activPAL n=2, DynaPort n=8, Mox=1). The other six studies were qualitative studies.

Risk of bias

3 studies were classed as low, 9 as moderate and 17 as high quality, as illustrated in table 1. Common criteria found missing in the quantitative studies (n=10) included not accounting for the confounders in the design and analysis and incomplete outcome data (n=14), due to the majority of studies not clarifying severities within their sample size. The most missing criteria in the qualitative studies were insufficient substantiation of the data due to the small homogenous sample size, not including COPD severity. Furthermore, missing COPD severity and limited qualitative data collection caused unable to address the research question adequately (tables 1–3).

Table 1.

Demographics

Author	Quality rating	Sample size (n=)	Age (years)	Males (%)	Season/time of year	Ethnicity	Living arrangements	Occupation
Cani et al (Brazil)45	High	59	Oxygen group (O2), 68±8; control (C), 67±8 mean	O2, 79%; C, not reported	Not reported	Not specified	Not specified	Not specified
Chang et al (Singapore)49	Moderate	6	72 (5) mean	100%	Singapore June–Dec	Not specified	Not specified	Not specified
Cheng et al (Australia)42	High	69	74 (9) mean	48%	All year, Australia (mild–warm temp)	Not specified	Not specified	81% retired
Cordova-Rivera et al (Australia)23	Moderate	64	69 (63–74) median	56%	Not reported	Not specified	Not specified	Not specified
Dogra et al (Canada)33	Low	877	65±10 mean	46%	Not reported	Not specified	Not specified	54% retired
Driver, Novotny and Benzo (USA)29	Low	292	68 (9) mean	51%	Not reported	93% white	Not specified	Not specified
Frykholm et al (Sweden, Canada and Netherlands)44	High	81	67±8 mean	57%	Covering all seasons	Not specified	Not specified	Not specified
Geidl et al (Germany)24	High	326	58±6 mean	68%	Not reported	Not specified	Not specified	25% retired
Hartman et al (Netherlands)28	High	113	65±9 mean	67%	April–Sept and Oct–March	Not specified	27% lived alone	89% retired
Hartman et al (Netherlands)34	High	115	65 (9) mean	68%	Not reported	Not specified	26% lived alone	89% retired
Hirata et al (Brazil)40	High	55	67±8 mean	51%	Not reported	Not specified	Not specified	Not specified
Hoaas et al (Norway (N) + Denmark (D) + Australia (A))48	High	N, 38; D, 36; A, 94	N, 66 (63–72); D, 63 (61–68); A, 66 (60–73) median	N, 63%; D, 55%; A, 55%	N, −7°–15°; D, −2°–21°; A, 6°–26°	Not specified	Not specified	Not specified
Hunt et al (Australia)25	High	141	71 (8) mean	60%	Not reported	Not specified	Not specified	Not specified
Koreny et al (Spain)50	High	404	69 (9) mean	85%	Not reported	Not specified	24% lived alone	88% retired
Lewis et al (Australia)35	Moderate	24	75 (8) mean	75%	Not reported	Not specified	0% lived alone	Not specified
Loprinzi and Walker (USA)41	Low	851	25 (0.2), 40 (0.3), 59 (0.4) mean	20–29 y, 62%; 30–49 y, 55%; 50+, 57%	Not reported	Not specified	Not specified	Not specified
McKeough et al (UK) 39	High	103	73±9 mean	52%	Not reported	Not specified	Not specified	Not specified
McNamara et al (Australia)36	Moderate	50	COPD+physical comorbidities (PC), 73 (11); COPD, 70 (8) mean	COPD+PC, 44%; COPD, 48%	Not reported	Not specified	Not specified	Not specified
Mesquita et al (Netherlands) 46	Moderate	125	67 (62–74) median	55%	Not reported	Not specified	0% lived alone	51% retired
Morita et al (Brazil)43	High	145	65 (60–73) median	54%	Not reported	Not specified	Not specified	Not specified
Munari et al (Brazil)30	High	115	66±8 mean	68%	Not reported	Not specified	Not specified	Not specified
Orme et al (UK)37	High	109	66 (7) mean	61%	Not reported	White-British	Not specified	75% retired
Park et al (USA)31	Moderate	224	70±9 mean	51%	Not reported	Non-Hispanic white 72%	48% lived alone	86% retired
Park et al (USA)38	Moderate	223	70±9 mean	51%	Not reported	Non-Hispanic white 72%	48% lived alone	86% retired
Pitta et al (Austria (A) + Brazil (B))47	High	80	A, 63±7; B, 66±8 median	A, 26%; B, 23%	Warm season	A, all Caucasian; B, 65% Caucasian	A, 48% lived alone; B, 15% lived alone	100% retired
Stevens et al (Canada)51	Moderate	418	58±8 mean	35%	Not reported	Not specified	27% lived alone	95% retired
Weedon et al (UK)26	Moderate	21	67 (52-81) mean	29%	Spring (UK)	Not specified	Not specified	Not specified
Wshah et al (Canada)32	High	14	76 (87, 60) mean	29%	Spring (Canada)	Not specified	71% lived alone	100% retired
Xavier et al (Brazil)27	High	152	68 (62–74) median	68%	Not reported	Not specified	Not specified	91% retired

COPD, chronic obstructive pulmonary disease.

Table 2.

Severity, design

Author	GOLD I, >80%; II, 50–70; III, 30–49; IV, <30%	Severity of COPD, FEV1	Type of study	Type of tool used	Hospital/community based
Cani et al 45	Oxygen group (O2): III, 21%; IV, 79%. Control (C): III, 20%; IV, 80%	O2, 25±7; C, 25±9 mean	Cross-sectional	DynaPort	Community
Chang et al 49	III, 50% people; II, 50% people	51% (14) mean	Interview	Not measured	7–14 days posthospital discharge
Cheng et al 42	I and II, 51%; III and IV, 49%	55% (19) mean	Cross-sectional	ActivPAL	Community
Cordova-Rivera et al 23	Not reported	52% (33–67) median	Cross-sectional	ActivPAL	Community
Dogra et al 33	Not reported	85±20 mean	Cross-sectional	Self-reported	Community
Driver, Novotny and Benzo29	Not reported	41% (17) mean	Cross-sectional	SenseWear Armband	Community
Frykholm et al 44	I, 15%; II, 46%; III, 28%; IV, 11%	57±19 mean	Cross-sectional	DynaPort	Community
Geidl et al 24	I, 9%; II, 44%; III, 38%; IV, 8%	54±18 mean	Cross-sectional	ActiGraph	Community
Hartman et al 28	I, 27%; II, 27%; III, 28%; IV, 19%	52% (14–119) median	Cross-sectional	DynaPort	Community
Hartman et al 34	I, 26%; II, 27%; III, 29%; IV, 18%	58% (28) mean	Mixed methods	DynaPort	Community
Hirata et al 40	I, 2%; II, 62%; III, 20%; IV, 16%	55% (38–62) median	Cross-sectional	SenseWear Armband	Community
Hoaas et al 48	Norway (N) I, 0%; II, 53%; III, 29%; IV, 18%. Denmark (D) I, 0%; II, 8%; III, 42%; IV, 50%. Australia (A) I, 0%; II, 43%; III, 40%; IV, 17%	N, 51% (39–64); D, 33% (25–41); A, 44% (32–62) median	Cross-sectional	SenseWear Armband	Community
Hunt et al 25	Not reported	50% (17) mean	Cross-sectional	ActiGraph	Community
Koreny et al 50	I, 9%; II, 53%; III, 30%; IV, 8%	57% (18) mean	Cross-sectional	DynaPort	Community
Lewis et al 35	Not reported	54% (23) mean	Cross-sectional	ActiGraph and SenseWear Armband	Community
Loprinzi and Walker41	Not reported	Not available	Cross-sectional	ActiGraph	Community
McKeough et al 39	Not reported	56±23 mean	Cohort study	Self-reported	Community
McNamara et al 36	COPD+physical comorbidities (PC) I, 0%; II, 36%; III, 4%; IV, 10%. COPD I, 0%; II, 30%; III, 18%; IV, 2%	COPD+PC: 51% (17). COPD: 54% (11) mean	Cohort study	SenseWear Armband	Community
Mesquita et al 46	Not reported	51% (33–65) median	Cross-sectional	Mox	Community
Morita et al 43	I and II, 39%; III and IV, 61%	45±15 mean	Cross-sectional	DynaPort	Community
Munari et al 30	Not reported	35±16 mean	Cross-sectional	DynaPort	Community
Orme et al 37	Not reported	76% (18) mean	Cross-sectional	ActiGraph	Community
Park et al 31	Not reported	Not available	Cross-sectional	ActiGraph	Community
Park et al 38	Not reported	Not available	Cross-sectional	ActiGraph	Community
Pitta et al 47	Austria (A) I, 0%; II, 20%; III, 18%; IV, 11%. Brazil (B) I, 0%; II, 20%; III, 23%; IV, 8%	A: 48±17. B: 46±17median	Cross-sectional	DynaPort	Community
Stevens et al 51	Not reported	Not available	Cross-sectional	Self-reported	Community
Weedon et al 26	Not reported	Not available	Interview	Not measured	In hospital
Wshah et al 32	Not reported	52% (91, 24) mean	Interview	Not measured	Community
Xavier et al 27	I, 3%; II, 32%; III, 51%; IV, 13%	44% (35–53) median	Cross-sectional	ActiGraph	Community

COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in one second; GOLD, Global Initiative for Chronic Obstructive Lung Disease; PC, physical comorbidities.

Table 3.

Findings

Author	Findings
Cani et al 45	LTOT use is associated with increased SB and physical inactivity (Cramer’s V=0.29, p=0.040)
Chang et al 49	Lack of awareness to replace SB with light PA and poor adherence to movement-related advice
Cheng et al 42	SB was inversely correlated with light (r = −0.97, p<0.01) and moderate-to-vigorous intensity PA (r = −0.55, p<0.01) and exercise capacity (r = −0.33, p<0.01), but not with age, BMI or lung function. Seasonal variation not associated with total SB
Cordova-Rivera et al 23	Less severe COPD less SB. COPD compared with controls: mean-difference (CI) of 59.0 more minutes of total sedentary time, 6.4 min longer ×50, −6.2 less sit-to-stand transitions
Dogra et al 33	Those who reported highest weekly SB had higher odds of reporting poor perceived health (OR=2.70, CI: 1.72 to 4.24), poor perceived mental health (OR=1.99, CI: 1.29 to 3.06) and unhealthy ageing (OR=3.04, CI: 1.96 to 4.72)
Driver, Novotny and Benzo29	Individuals with 1-point better dyspnoea scores averaged 24.5 (8.4–40.5) min less SB per day
Frykholm et al 44	Isotonic quadriceps endurance was the only muscle contributor that improved daily steps (ΔR2=0.04 (relative improvement 13%) p=0.026), daily sedentary time (ΔR2=0.07 (23%), p=0.005) and MVPA minutes (ΔR2=0.08 (20%), p=0.001)
Geidl et al 24	Clusters differed significantly in disease-related parameters of GOLD severity FEV1, CAT and 6MWT
Hartman et al 28	SB same throughout COPD severity stages (mean SB, % per day: GOLD I, 36; GOLD II, 36I GOLD 3, 38; GOLD 4, 39; p = 0.440), more SB associated with more positive perception of treatment control, less motivation to exercise, not using sleep medication, and O2 use. Use of sleeping meds was associated with shorter SB
Hartman et al (B)34	Reported reason to be SB was the weather (48% of participants), followed by health problems (43% of participants) and lack of intrinsic motivation (20% of participants)
Hirata et al 40	Sleep quantity of ≥9 hours is associated with increased average SB (599 min/day) compared with sleep quality of <7 hours (533 min/day). ≥9 hours also associated with less time per day in PA and had more fragmented sleep
Hoaas et al 48	Danes more SB (median 784 min/day (660–952) vs 775 min/day (626–877) for Norwegians vs 703 min/day (613–802) for Australians, p=0.013). People with COPD, higher PA during the summer
Hunt et al 25	Higher BMI, airflow obstruction, dyspnoea and exercise capacity were associated with greater SB (p<0.0001)
Koreny et al 50	(1) Higher population density associated with less steps and more SB and worse exercise capacity (−507 (95% CI: 1135 to 121) steps, +0.2 (0.0, 0.4) hours/day and −13 (−25, 0) m per IQR). (2) Pedestrian street length related to more steps and less SB (156 (9, 304) steps and −0.1 (−0.1, 0.0) hour/day per IQR). (3) NO2 exposure relates to increased SB
Lewis et al 35	Level of physiological impairment was an independent predictor of waking SB, people with chronic conditions spent more time sedentary (COPD 62% vs healthy group 45% of daily waking hours)
Loprinzi and Walker41	Smokers 50+ years of age with greater nicotine dependence engaged in more SB (β=11.4, p=0.02) and less light-intensity physical activity (β = −9.6, p=0.03) and moderate-to-vigorous physical activity (MVPA; β = −0.14, p=0.003)
McKeough et al 39	Found no associations between SB and any functional performance outcome or with health-related quality of life
McNamara et al 36	People with COPD and comorbidities are more sedentary than people with COPD without comorbidities (COPD+PC mean 771 (98) min/day vs COPD (603 (148) min/day) vs control group (567 (76) min/day) (p<0.001)
Mesquita et al 46	People with COPD more SB and less active than their partner (p<0.0001). If the partner was active, they were also found more active but not less sedentary
Morita et al 43	Patients with delayed 1 min heart rate recovery after 6MWT exhibited worse exercise capacity as well as a more SB (472±110 min vs 394±129 min, p =0.002) and worse functional status
Munari et al 30	SB is associated with COPD GOLD D (with mMRC of >2)
Orme et al 37	Low-intensity movement, comprising SB/light activity/average, movement intensity negatively associated with limitations with mobility, daily activities and BMI but positive association with health status independent of high-intensity movement and sleep
Park et al 31	Race, level of education, working status, shortness of breath significantly associated with SB
Park et al 38	Waist circumference and glucose level were significantly associated with SB
Pitta et al 47	Austrians significantly lower walking time (p=0.04), higher sitting time (p=0.02) and lower movement intensity (p=0.0001) compared to Brazilians
Stevens et al 51	Amount of average daily SB was positively associated with education and urbanicity (β = 0.113; t value = 1.71; P = 0.011), negatively associated with PA and self-rated health (β = −0.159; t value = −2.42; P = 0.016)
Weedon et al 26	(1) SB to enable them to perform activities, such as housework; (2) sitting, experienced as enjoyable; (3) the most ill participants experienced sitting in terms of sadness, as the only thing they could do
Wshah et al 32	Determinants for reducing ST: knowledge, beliefs about consequences, beliefs about capabilities, environmental context and resources, social influences, social/professional role and identity, and behavioural regulation
Xavier et al 27	Participants with greater SB were associated with higher age, dyspnoea and airflow obstruction. Participants in phenotype 1 (n = 61) had median (IQR) age of 65 (60–70) years and moderate to severe airflow obstruction. Compared to participants in phenotypes 2 and 3, phenotype 1 was more physically active (p<0.0001), less sedentary (p<0.0001)

BMI, body mass index; C, control group; CAT, COPD Assessment Tool; COPD, chronic obstructive pulmonary disease; FEV1, Forced Expiratory Volume in one second; GOLD, Global Initiative for Chronic Obstructive Lung Disease; LTOT, long-term oxygen therapy; mMRC, Modified Medical Research Council; MVPA, moderate- to vigorous-intensity physical activity; 6MWT, Six Minute Walk Test; PA, physical activity; PC, physical comorbidities; SB, sedentary behaviour.

Disease and health-related factors

Severity of COPD

Mixed findings were reported regarding the severity of COPD as an influencing factor of sedentary behaviour. Five studies^23–27 found a positive association with disease severity. These findings were contradicted by a study²⁸ in which sedentary behaviour was independent of disease severity.

Dyspnoea

The Modified Medical Research Council Dyspnoea Scale was predominantly used to assess breathlessness (n=6). Dyspnoea was found to be an influencing factor in six studies.25 27 29–31 People spent more time sedentary to reduce dyspnoea .²⁶ This implies this was not only a passive consequence of the symptoms but a rationalised self-management strategy. Dyspnoea was also identified as a barrier to progressing from sedentary behaviour to light physical activity interventions.³²

Comorbidities

Four studies associated poor health and comorbidities with greater sedentary time.^33–36 Comorbidities concern people with COPD.³² Higher waist circumference^{37 38} and body mass index^{25 37} were associated with time spent sedentary.

Health-related quality of life

One study used the St George’s Respiratory Questionnaire to report findings on health-related quality of life and its impact on sedentary positions. No association between health-related quality of life and sedentariness were found.³⁹ This study did not report occupation or GOLD severity within their population.

Age

Older age was associated with increased sedentary behaviour in one study.²⁷

Lifestyle and behavioural factors

Sleep

The study by Hirata et al assessed sleep with objective measuring tools and found a positive correlation between time (more than 9 hours) spent in bed and sedentary behaviour, but increased time spent in bed had a negative correlation with sleep quality.⁴⁰ In contrast, Hartman et al’s study assessed sleep subjectively, using the Pittsburgh Sleep Quality Index, and found a negative correlation between the use of sleeping medication and daily sedentary behaviour .²⁸

Nicotine dependence

Loprinzi and Walker found a positive correlation between nicotine dependency and sedentary behaviour in people over 50 years of age.⁴¹

Physical factors

Exercise capacity

An inverse relationship between exercise capacity and sedentary behaviour was reported in five studies.24 25 28 42 43 All studies assessed exercise capacity with the Six Minute Walk Test.

Muscle function

A study by Frykholm et al found that increased isotonic quadriceps endurance is associated with reduced sedentary time.⁴⁴

The use of supplemental oxygen therapy, ambulatory oxygen or walking aids

Two studies reported a positive association between the use of supplemental oxygen and the time spent sitting, lying or reclining.^{28 45} Furthermore, it was found that ambulatory oxygen and walking aids could hinder reducing sedentariness despite being a necessity for some of the COPD population.^{26 32}

Interpersonal and family-related factors

Living arrangements

The work by Mesquita et al found people with COPD to be more sedentary than their loved ones, and the amount of time spent sedentary did not change if living with an active or inactive partner.⁴⁶ This contradicts the study by Wshah et al, describing social influences, such as support from family and loved ones, as well as home and neighbourhood support, as determinants of sedentary behaviour.³² All of the participants in the study by Mesquita et al were living with a partner or loved ones.⁴⁶

Sociodemographic and environmental factors

Employment status

Studies in which participants were retired reported overall higher levels of sedentary behaviour.³¹

School education

Few studies explored educational impact, and only one study reported an association between lower high school education and increased sedentary behaviour.³¹

Demographic

A study by Pitta et al compared behaviour in a developing country (Brazil) with a developed country (Austria).⁴⁷ Most participants were of Caucasian origin. The Brazilian COPD population observed lower sedentary behaviour, greater walking time and higher movement activity, indicating that people with COPD from lower-income countries are less sedentary .⁴⁷ An ethnic association was seen in a US study, which reported that non-Hispanic white populations were less sedentary compared with other ethnic groups, which included Hispanic, Mexican-American and non-Hispanic Black populations.³¹ A study by Hoaas et al found that Danes were more sedentary than Norwegians and Australians, though the Danish population had a greater COPD severity.⁴⁸

Environment

An association between weather and sedentary behaviour has been made in studies conducted in Canada, Netherlands, Norway, Denmark and Australia.^{32 34 48} Colder weather has a negative association with sedentary behaviour,³² and people were found to be more active during the summer months.⁴⁸ These findings were contradicted by the study by Cheng et al, which did not find weather-related changes associated with sedentary behaviour.⁴² Of note, this study was performed in Australia, with a mild to warm climate, likely not generalisable to the COPD population living in other countries and facing colder seasons.⁴² Three other studies that found no association between the weather and sedentary time were conducted during mild to moderately mild weather.^{26 47 49}

Urbanicity and sedentary behaviour have been found to have a direct relationship.^{50 51} Higher population density and higher nitrogen dioxide exposure were also associated with greater sedentary behaviour.⁵⁰ In contrast, long pedestrian street length was associated with less time spent sedentary.⁵⁰

Psychological factors

Some people with COPD are unaware of the benefits and the importance of reducing sedentary behaviour.^{32 49} Demotivation,³⁴ self-monitoring and adherence to movement-related advice,^{32 49} and beliefs about capabilities to reduce sedentary behaviour were also identified.³² In the study by Weedon et al, sedentary behaviour was described as a conscious approach and being part of enjoyment when participating in activities such as TV watching and other sedentary hobbies in people with COPD.²⁶ This implies that people include sedentary time into their daily routine to enable independence, empowering people to choose activities they would like to participate in.²⁶ Of note, severely ill people with COPD associate sedentary time with negative feelings, as the only thing they can participate in.²⁶

Sensitivity analysis, comparing the quality of the studies with the findings of the review

To better evaluate the quality of the included studies, Hong et al advised performing a sensitivity analysis instead of calculating an overall score.²¹ The majority of the included papers on the most common influencing factors of sedentary behaviour were considered as moderate (n=12) to high quality (n=25). Only two of the studies were of low quality and were found in the following categories: dyspnoea²⁹ and the presence of comorbidities.³³ The eight other studies in the dyspnoea category and seven in the presence of comorbidities were of moderate to high quality. Therefore, the effect of the included studies being of low quality is insignificant.

Discussion

This is the first systematic review to identify influencing factors of sedentary behaviour in people with COPD. This review highlights that sedentary behaviour has become of growing interest, with the majority of papers being published in the last 6 years. Themes across the 29 included studies were related to disease severity, lifestyle, physical and psychological, interpersonal, sociodemographic and environmental factors.

Multiple influencing factors of sedentary behaviour in people with COPD were identified. Whether sedentary behaviour impacts these factors, or vice versa, is not clear. The most common associations were disease severity, dyspnoea, presence of comorbidities, exercise capacity, the use of supplemental oxygen and walking aids and environmental factors. The results also imply socioeconomic and ethnic factors could influence time spent sedentary. People living in higher-income countries have also been seen as more physically inactive compared with middle-income populations.⁵² However, the majority of our included papers were conducted in high-income countries.

Qualitative methodologies elicited more in-depth, at times contrasting findings. These included a lack of knowledge, self-perception, motivation and beliefs about capabilities. These findings are congruent with a meta-analysis of sedentary behaviour in general adult populations.⁵³ Of the studies that explored psychological factors, sedentary behaviour was suggested to be a choice rather than a lack of awareness of the behaviour. For example, participating in sedentary hobbies, such as TV watching and fishing, provided pleasure. Sedentary behaviour was used as an empowering tool to maintain independence as part of a pacing strategy. This was also reported in a systematic review of general adult populations, where increased sedentary behaviour was associated with positive attitudes towards sedentary behaviour.⁵² An interview study with older women found that sedentary behaviour was an important coping mechanism to manage pain and fatigue.⁵⁴ This highlights the importance of managing sedentary behaviour during individuals’ daily lives in order to comply with the recommendations to decrease sedentary time and continue to have enjoyment.

This review highlights the multifactorial reasons for adopting a sedentary lifestyle, in particular breathlessness, fatigue and management of other health-related comorbidities. This knowledge must be taken into account when considering the design of an intervention to reduce sedentary time. Further research regarding the awareness of sedentary behaviour, versus consciously choosing a sedentary lifestyle to enable empowerment, would be of interest.

The current review importantly identifies potential future treatment considerations for health professionals, summarised in. Given the multifactorial nature of sedentary behaviour, it is likely that a tailored, multimodal approach must be taken in this population. This would include individualised positive affective experiences to sustain engagement to promote associated health benefits. Raising awareness and educating on the importance of reducing sedentary behaviour, perhaps during pulmonary rehabilitation, hospital admission or annual COPD reviews, should be considered. These should also include individualised goal setting to increase capabilities, confidence and self-belief. The multimodal approach might incorporate COPD optimisation, symptom control such as breathlessness,

pharmacological optimisation, sleep hygiene and fatigue management throughout an individual’s daily routines. Given the multifactorial influencers which contribute to a sedentary lifestyle in people with COPD, focusing on reducing sedentary behaviour might be a more suited approach than increasing physical activity. Figure 2

Figure 2.

Influencing factors of sedentary behaviour to inform a multimodal approach by healthcare workers in partnership with individuals living with chronic obstructive pulmonary disease (COPD). The outer boxes represent strategies to potentially address sedentary behaviour, while the inner circles represent influencing factors.

Future research in this field could include qualitative and longitudinal studies to explore determinants contributing to a sedentary lifestyle covering all COPD severities, including people during or following an acute exacerbation. Second is exploring whether these factors could be influenced to help develop interventions to reduce sedentary time. Subsequently, randomised controlled trials, including those set in lower and middle-income countries, are needed to identify sustainable health strategies in this field.

This review followed a comprehensive search strategy and is the first to systematically review influencing factors of sedentary behaviour in people with COPD, including qualitative and quantitative approaches. Sedentary behaviour in people with COPD is multifactorial, and psychological, environmental, social and health-related factors should be considered.

The available evidence summarised by this review provides health professionals with important information to support a tailored approach to patients with COPD to address sedentary behaviour, for example, combining breathlessness management with pacing techniques. The findings of this review help raise the profile of sedentary behaviour and help inform future guidelines.

Limitations include heterogeneity of design and outcome, which limited direct comparison. A narrative synthesis rather than a meta-analysis was therefore performed, which limited objective appraisal of the findings. There is limited literature on people following hospital admission; only two studies explored people with an acute exacerbation of COPD. The limited data following an acute exacerbation of COPD need to be acknowledged, and a generalisability of these findings should be made cautiously. Following the qualitising method may have simplified some outcomes. However, integrating qualitative and quantitative studies provides a more comprehensive understanding of the phenomena, compared with undertaking two syntheses separately without combining them when data pooling is not possible.

As this study explored influencers of sedentary behaviour, including cross-sectional design, this limits causal inference to determine exposure or outcome. Although most of the studies included were of moderate to high quality, there is a potential for selection bias. For example, people who agreed to be participants may be more motivated to change their behaviour than those who declined to participate in a study. Measuring devices and time varied throughout the studies, limiting direct comparisons between studies and future calculations of effect sizes. Some studies did not include all COPD severities, or the severity of the disease was unevenly proportioned, and the outcome might not be generalisable.

Conclusion

The results of this review suggest that the influences of sedentary behaviour in people living with COPD are multifactorial. Identifying and understanding factors, including disease severity, dyspnoea, comorbidities, physical capacity and environmental and personal influencers, could help inform future interventions. The study highlights the lack of longitudinal studies accounting for the chronic nature of COPD. More insight from interventional influences is needed to aid the establishment of guidelines measuring and impacting on sedentary behaviour in people with COPD.

The protocol is available on PROSPERO protocol ID: CRD42023387335.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.