Physical Activity, Spirometry and Quality-of-Life in Chronic Obstructive Pulmonary Disease

Abstract

Chronic obstructive pulmonary disease (COPD) has a high level of morbidity. There is limited information about levels of physical activity among community dwelling subjects with the disease and its association with lung function and quality of life. In this study, 176 subjects with COPD were recruited from general practices. Physical activity was measured over 7 days using pedometers, quality-of-life using the St. George's Respiratory Questionnaire, and lung function using spirometry, 124 subjects, 60% male, aged 70 ± 8 years were included in the analysis. Physical activity levels were low, median steps/day for males = 3,621,IQR = 4,247 and for females = 4,287,IQR = 3,063. Overall physical activity (median steps/day = 3,716,IQR = 3682) was significantly associated with the forced expiratory volume in one second (r = 0.39, p < 0.001) and with St. George's Respiratory Questionnaire score (r = −0.28, p < 0.01). In conclusion, we have demonstrated that in a community-based sample of people with COPD, daily physical activity levels were low compared with usual levels reported for the general population. Physical activity was significantly associated with disease severity, measured by lung function, and quality-of-life.

Key words: :

• Pulmonary Disease
• Chronic Obstructive
• Motor Activity
• Pedometers
• Quality of Life
• Respiratory Function Tests

INTRODUCTION

Chronic obstructive pulmonary disease (COPD) is a major cause of morbidity and mortality in Australia. It is the fourth leading cause of death and the fourth highest cause of disease burden for Australians over the age of 65 ([1]). Similar epidemiological data have been documented in other developed countries ([2]). Treatment of COPD has traditionally focused on minimizing symptoms through the use of pharmacological interventions, usually bronchodilators. More recently, the importance of incorporating strategies for maximizing functional status and quality-of-life (QOL) has been recognized ([3], [4]). This is reflected in management guidelines that include recommendations for increasing physical activity as part of the overall management of COPD ([5], [6]). Recommendations for pulmonary rehabilitation programs also emphasize the importance of improving physical activity ([4], [7]), with assessment of the intervention including measures of both QOL and functional status ([7]).

Functional status can be defined in terms of physical function, social function, role function and psychological function and is an objective measure of the ability to perform tasks. It is distinct from health related QOL measurements, which quantify an individual's subjective experience of disease ([8]). Research in COPD has shown that functional status, including physical function, exhibits a strong association with QOL ([9], [10], [11], [12]). The instruments used to measure physical function, such as timed walking tests and cycle ergometry, are measures of physical capacity and distinct from measures of physical activity, which also reflect behaviors. In contrast to physical capacity, little is known about everyday physical activity and its relationship to QOL and lung function. A study of 346 COPD patients attending tertiary level hospitals in Spain ([13]) showed that questionnaire measures of physical activity were associated with quality-of-life but not with lung function as measured by FEV₁. Motion sensors provide a more objective measure of physical activity and several small studies have shown that usual levels of physical activity are significantly lower in patients with COPD than in the general population ([14], [15], [16]). While two of these studies showed associations of objectively measured physical activity with lung function ([15], [16]), they did not examine associations with quality-of-life. In addition, the study samples were selected from COPD patients attending hospitals, which may not be representative of general community samples of people with the disease. In this study, we used pedometers to measure physical activity and investigated whether routine daily physical activity, recorded as steps per day, was associated with quality-of-life and measures of disease severity in people with COPD in a community setting.

MATERIALS AND METHODS

Patient selection

Participants were recruited from the patient population attending general practitioners (GPs) working in the greater Hobart area. Hobart is situated in southern Tasmania. It is Australia's smallest state capital with a population of 195,500. All eligible GPs registered with the Southern Tasmanian Division of General Practitioners (n = 255) were contacted by mail and invited to participate. Fifty-three of the registered GPs (21%) agreed. Participants were identified when potential subjects presented for a consultation and by reviewing the practices' medical files. Subjects with a diagnosis of COPD, either made or accepted by their GP were invited to participate in the study. Inclusion criteria included COPD as their primary functionally limiting illness, aged more than 50 years, with a tobacco smoking history of greater than 10 pack years and both an FEV₁ of less than 65% predicted and a FEV₁/FVC of less than 70% ([6]). Nursing home residents were excluded. The study was approved by the Southern Tasmanian Health and Medical Human Research Ethics Committee and written informed consent was obtained from all participants prior to enrollment.

Study design

The study was a cross-sectional study of community dwelling people with a diagnosis of COPD. A research nurse interviewed participants, either at their GP's surgery (68%) or at home (32%). Information on demographics, smoking, symptom history, medication use and treatment interventions for COPD over the preceding 12 months was obtained from the participants using a standardized questionnaire. Height and weight measurements were recorded and spirometry performed (MicroLab 3300, Micromedical Ltd, UK).

Quality of life

The St. George's Respiratory Questionnaire (SGRQ) ([9]) was used to assess health-related QOL. The SGRQ has been specifically designed to measure QOL in airways disease. It is a self-administered questionnaire composed of 50 weighted items. The questionnaire measures three domains, namely symptoms, activity and impacts. Both the total score and sub-component scores are recorded as percentages, with a score of 100% indicating maximum impairment (i.e., the higher the score the greater the impairment, and vice versa).

Physical activity

Physical activity was recorded over a 7-day period using a digital pedometer (HJ 003, Omron Healthcare, Singapore). Participants were given instructions on how to wear the pedometer and to record the number of steps indicated on the pedometer at the end of each day on a diary card before resetting the pedometer to zero. The time spent wearing the pedometer, and on periods of non-walking exercise (cycling, swimming, gym work and upper body exercises) was also recorded. All pedometers were individually calibrated using a walking test of 20 paces. The completed diary cards and pedometers were returned by post at the conclusion of the 1-week recording period. The median of the 7-day recordings, or fewer than 7 days if the diary card was incomplete, was used as a measure of everyday physical activity.

Statistical analysis

The Kolmogorov–Smirnov test was used to determine the distribution of the variables. Variables with a normal distribution were summarized as means and standard deviations. Variables with non-parametric distributions were summarized as medians and interquartile ranges (IQR). Comparisons between groups were assessed using independent t-test, χ² test or Mann–Whitney U-test. Correlations were conducted using Pearson correlation coefficients. Logistic regression analysis was used to examine the independent associations with participation in non-walking exercise. Log transformations were performed for non-parametric variables used in all correlations and the linear regression analyses. Data were analyzed using the Statistical Package for the Social Sciences 11.0 for Windows (SPSS; Chicago, USA).

RESULTS

Demographics

First, 218 people were identified by GPs as potentially suitable for inclusion, of which 176 agreed to participate in the study. Then, 37 subjects were deemed to be ineligible following the baseline assessment and were excluded from the analysis: 25 did not fulfill the spirometric inclusion criteria for COPD; 5 had a total smoking history of less than 10 pack years; and 7 subjects were excluded on the basis of having another primary functionally limiting illness (dementia, multiple sclerosis, renal failure, paraplegia and rheumatoid arthritis). A further 15 subjects were excluded from the analysis because of dysfunctional use of pedometers: 4 were unable to read the pedometer display, 8 failed to return their diary cards and 3 wore their pedometers for less than 70% of waking hours on 6 or more days.

Then, 124 subjects (57% of those identified by their general practitioner as being eligible) included in the final analysis. The cohort was predominantly elderly, male and retired (Table 1). There were no significant differences between male and female subjects in terms of age, or occupational status, but women were more likely to be widowed than men. The mean body mass index was 26 ± 6 Kg/m². The mean total cigarette smoking history for the study population was 57 ± 30 pack years. Of these, 31% of participants were current smokers, with more females currently smoking than males, although this difference was not statistically significant, (χ² = 2.1; df = 1; p = 0.14). Only 27% of the participants had participated in a formal cardio-respiratory rehabilitation program, females more commonly.

Table 1. Characteristics of the study participants

	Males (n = 74)	Females (n = 50)
Age (years)
50–59	8	18
60–69	34	36
70–79	47	36
80–89	11	10
Occupational status
Retired	96	94
Employed	4	6
Marital status
Married	60	36
Widowed	23	52
Separated/divorced	8	12
Never married	9	0
Body Mass Index (Kg/m^2)
< 20	15	22
20–25	31	34
> 25	54	44
Smoking history (pack years)
< 20	7	8
20–40	19	26
41–60	30	38
61–80	22	20
> 80	22	4
Current smokers	26	38
Participation in a cardio-pulmonary rehabilitation program	18	44
Acute interventions for COPD in the preceding 12 months
Antibiotic use	57	68
Short course of oral corticosteroids	30	44
Treatment at a hospital emergency department	15	16
Hospitalization	27	32
Duration of symptoms (yrs)
Cough	5 (20.0)	1 (10.5)
Phlegm	3.5 (12.5)	3 (10.0)
Dyspnea	8 (9.3)	6.5 (9.0)

Data are presented as percentages, except duration of symptoms, which are presented as median (Inter-Quartile Range).

Morbidity

All subjects included in the analysis had symptoms consistent with COPD. Dyspnea was of longest duration (median = 7 years, IQR = 9), with cough (median = 3 years, IQR = 15) and phlegm (median = 3 years, IQR = 10) present for shorter periods. Medications for COPD had been used for a median of 5 years (IQR = 7), 89% of subjects using inhaled bronchodilators, 48% inhaled corticosteroids and 7% oral corticosteroids as maintenance therapy. The majority of subjects were up to date with immunizations, 91% having received an influenza vaccination within the last year and 74% a pneumococcal vaccination within the last 5 years. During the 12 months preceding entry into the study, 61% of the participants had been prescribed antibiotics and 36% oral corticosteroids for an exacerbation of COPD. Fifteen percent had attended a hospital emergency department and 29% hospitalized for COPD.

Physical activity

Physical activity, defined as the median number of steps/day measured by pedometer readings, was 3716 (range 276–14,511) steps per day. Physical activity was not significantly different between males and females, although the median number of steps was slightly higher for females at all age ranges (Figure 1). The median number of steps per day decreased with increasing age for both males and females (r = 0.27, p < 0.01). Then, 65% of the participants recorded additional non-walking exercise; upper limb exercises (n = 68), use of an exercise bike (n = 10) and swimming (n = 1) on one or more days (Table 2). The median number of steps was higher for those subjects who participated in additional non-walking physical activities (3752, IQR = 3354), than those who did not (3377, IQR = 4623), and this difference was statistically significant (p = 0.04). Subjects who participated in additional non-walking exercise were more likely to be female (p < 0.01) and to have participated in a formal cardio-respiratory rehabilitation program (p < 0.05). Neither the mean FEV₁, expressed as percent of predicted, nor mean SGRQ scores were significantly different for subjects who participated in additional non-walking exercise than those who did not. There was no relationship between physical activity and duration of respiratory symptoms, total smoking history, current smoking status, marital status or BMI. Ten of the subjects were on domiciliary oxygen therapy; the median number of steps/day was significantly lower for this group than those not using domiciliary oxygen (2381 vs. 3744, r = 0.23, p = 0.009).

Figure 1. Median number of steps per day.

Table 2. Comparison of clinical characteristics and outcome variables for subjects who participated in additional, non-walking activities and those who did not participate in additional, non-walking activities

	Participation in non-walking activities (n = 65)	No participation in non-walking activities (n = 59)
Age yrs	69± 8	72± 8
Steps/day	3,752 (3,288)	3,661 (4,952)
Gender %
Males	42	79
Females	58	21
Pack years	55± 25	59± 34
Current smokers %	27	35
Participation in a rehabilitation program %	38	17
FEV1 % pred	46± 16	45± 16
SGRQ
Total	47± 18	48± 20
Symptoms	62± 22	62± 21
Activity	64± 22	65± 24
Impacts	32± 18	34± 21

FEV₁: forced expiratory volume in one second.; % pred:percentage of predicted value; SGRQ: St. George's Respiratory Questionnaire. Data presented as mean ± SD or median (IQR), unless otherwise indicated.

Disease severity

We assessed COPD severity in the population by spirometry and quality of life (Table 3). The overall mean FEV₁ was 1.1 ± 0.4 liters, which represented 46 ± 16% of predicted and the overall mean SGRQ total score was 47 ± 19. FEV₁, expressed as percent of predicted (FEV₁% predicted), correlated with the SGRQ total score (r = −0.42, p < 0.001), as well as all sub-components. Physical activity also showed a statistically significant correlation with FEV₁, when expressed as percent of predicted (r = 0.39, p < 0.001), and SGRQ total score (r = −0.28, p < 0.01) (Table 4). Subgroup analysis revealed the strength of associations to be similar for males and females, although the correlation between physical activity and SGRQ scores for females, apart from the activity sub-component, were not statistically significant. The correlation between physical activity and FEV₁% predicted was significant, irrespective of whether additional non-walking activities were undertaken (r = 0.32, p < 0.01 for additional exercise, r = 0.56, p < 0.001 for no additional exercise). However, no associations were found between physical activity and other indices of disease severity, such as hospitalization, attendances at hospital emergency departments or treatment with antibiotics or oral corticosteroids in the 12 months preceding the study (data not shown).

Table 3. Spirometry and quality-of-life measurements

	Males (n = 74)	Females (n = 50)
FEV1	1.19± 0.46	0.88± 0.33
FVC	2.26± 0.86	1.67± 0.47
FEV1 % predicted	44.5± 14.9	47.5± 17.4
FEV1/FVC	55± 14.8	52.3± 13.6
SGRQ
Symptoms	62± 23	62± 20
Activity	66± 23	63± 23
Impacts	34± 21	33± 20
Total	48± 20	47± 17

Data are presented as mean ± SD. FEV₁: forced expiratory volume in 1 second; FVC: forced vital capacity; SGRQ: St. George's Respiratory Questionnaire.

Table 4. Associations of physical activity (steps/day) with spirometry and the St. George's Respiratory Questionnaire score

	Associations with steps per day (Pearson correlation coefficient)
	Males (n = 74)	Females (n = 50)
FEV1 % predicted	0.43	0.43^†
FEV1/FVC	0.13	0.28
SGRQ:
Total	−0.30^†	−0.22
Symptoms	−0.23^‡	−0.21
Activity	−0.28^‡	−0.29^‡
Impacts	−0.27^‡	−0.13

FEV₁: forced expiratory volume in one second; FVC: forced vital capacity; SGRQ: St. George's Respiratory Questionnaire.

*p < 0.001;

^† p < 0.01;

^‡ p < 0.05.

DISCUSSION

This cross-sectional study provides the first description of physical activity, measured by pedometer, in a large sample of subjects with COPD recruited from the community. Levels of physical activity, measured as steps per day, were significantly associated with lung function and quality-of-life, particularly the activity domain of the SRGQ.

Our subjects were recruited through their GPs, with 57% of those identified included in the final analysis. This is in contrast to previous studies, which have included subjects recruited though hospital outpatients ([13], [15], [17]). Our sample was elderly and predominantly male, reflecting the historically higher prevalence of tobacco smoking among males in this age group. Physical activity levels were low, reflecting substantial impairment in this sample, but was similar to results obtained in previous studies on physical step activity in COPD ([13], [15], [17]) and other chronic illnesses ([14]). Subjects on LTOT had a significantly lower level of physical activity than those not requiring oxygen, which is in keeping with findings of other studies ([13], [15]). Other chronic diseases have similarly decreased levels of daily activity, with an average 3,540–3,571 steps/day for chronic heart failure and 4,116 steps/day for claudication, compared to twice these levels for healthy people in the same age range ([14]). Schonhofer, et al. ([17]) recruited 25 subjects with COPD and evaluated daily activity using pedometers. The findings were in accord with ours, their group had a similar level of lung function, mean FEV₁% predicted 47± 9%, and daily activity, mean daily movement count 3,781± 2,320, despite being younger. They, like us, found a significant correlation between daily activity and lung function, as has been found in other studies ([15]). We also found that physical activity correlated to quality-of-life, a relationship that has not been demonstrated previously. This relationship was strongest for the activity sub-component of the SGRQ in both males and females. Despite our sample having only moderate COPD as classified by the severity of airflow limitation ([18]), all subjects had a high SGRQ total score compared to the population mean of 8.4 ([19]), indicating a marked reduction in their QOL.

For measurement of physical activity we used pedometers, rather than traditional methods such as self-report instruments (diaries and questionnaires), Self-report methods have been the most widely used instruments for measuring physical activity, but their accuracy is limited by recall bias, variability in the instruments, and different scoring techniques ([14], [20]). Questionnaire measures of physical activity correlate only weakly with “gold-standard” measures of physical activity, including accelerometers, time in observed activity and total energy expenditure, whereas pedometer measures correlate strongly ([21]). The use of sympathomimetics in the treatment of COPD restricts the use of heart rate monitors as a valid measure of physical activity, while the cost and complexity of radioisotope energy expenditure studies limits their usefulness in community-based studies. Most participants were able to successfully record their everyday physical activity over a 7-day recording period using a pedometer. The use of a 7-day recording period provides a more representative recording of everyday physical activity than shorter recording intervals, which are less effective at capturing the variability of activity on different days of the week ([22]). Pedometers are an inexpensive method of quantifying physical activity in COPD, and have the advantage of recording incidental activity. Their patient acceptability is high and they overcome many of the limitations associated with alternative methods of measuring physical activity ([17]).

A potential confounder in our study was engagement in non-walking exercises, of which upper limb exercises were the most frequently recorded. We found the median number of steps undertaken by subjects who participated in additional non-walking exercise was higher than those who did not, at a statistically significant level. This increase in daily steps is in keeping with the finding of an association between daily walking and non-walking exercise in the general population ([23]), suggesting steps/day is a good surrogate of overall activity. Given that activity is influenced by aerobic fitness and muscle strength, it is possible that both associations may be due to greater cardiovascular fitness and muscle strength in subjects undertaking non-walking exercise, reducing the limitation imposed by impaired lung function. Although we found similar magnitude correlation coefficients between physical activity and QOL as for lung function for males and females, these correlations this did not reach statistical significance for most QOL sub-components in the female sub-group. This may reflect either the higher rate of non-walking activity, or a greater importance of psychosocial factors in determining QOL in female subjects. We used the SGRQ, which has demonstrated validity and reliability ([9]), to assess QOL. Although QOL in patients with chronic lung disease attending family practitioners, measured using the Sickness Impact Profile (SIP) ([24]), has been found to be less impaired than in hospital outpatients ([25]), the SIP is not as sensitive as the SGRQ in detecting differences in COPD ([9]). The mean SGRQ total score in this study was similar to that found in hospital outpatients ([9], [25]), and correlated significantly with the FEV₁, expressed as % predicted, as seen in previous reports ([26]).

In conclusion, we have demonstrated that in a community-based sample of people with COPD, daily physical activity levels were low overall with a median of 3,621 steps per day in men and 4,287 steps per day in women. Objectively measured daily physical activity was significantly associated with lung function and quality-of-life.

We are grateful to participants and the research nurses, Sue Davoren and Elizabeth Hammer, for their work on the study.

This study was funded by an unrestricted grant from Boehringer Ingelheim.