Validity and Responsiveness of the Glittre-ADL Test without a Backpack in People with Chronic Obstructive Pulmonary Disease

Abstract

The Glittre-ADL test assesses the functional capacity for activities of daily living of people with chronic obstructive pulmonary disease (COPD). In the test, a weighted backpack is worn (2.5 kg for women and 5.0 kg for men). The differential in weight between men and women is not common in other tests of exercise capacity and may limit the comparison of the test between sexes. The primary aim of this study was to validate the Glittre-ADL test performed without the backpack in people with COPD. Forty participants with mild to severe COPD (mean ± SD age: 70 ± 6 years; FEV₁: 48 ± 20%predicted) were recruited and performed two six-minute walk tests (visit 1); two Glittre-ADL tests with backpack (visit 2), and the Glittre-ADL test with and without the backpack, in random order (visit 3). The Glittre-ADL test time was shorter without the backpack than with the backpack [mean difference −0.37 min (95%CI −0.59 to −0.15)] and heart rate (HR) and oxygen saturation (SpO₂) were equivalents between tests [−1.31 beats/minute (−3.92 to 1.30) and −0.95% (−2.27 to 0.37), respectively]. The Glittre-ADL test without the backpack elicited similar HR and SpO2 responses as the test with the backpack, indicating equivalence of physiological demand. Thus, the Glittre-ADL test without the backpack was a valid, responsive, and appropriate test to assess functional capacity for activities of daily living.

Keywords:

• COPD
• functional capacity
• field tests
• and activities of daily living

Introduction

The Glittre-ADL test [1] has been proposed as a test that is more reflective of functional activities than other functional exercise tests for people with chronic obstructive pulmonary disease (COPD) as it includes standing from a chair, walking, climbing stairs and a grocery shelving task. These activities were chosen to represent common tasks of everyday life [1–3].

In the Glittre-ADL test, the participant wears a backpack filled with 2.5 kg for women or 5.0 kg for men. The weight of 2.5 kg simulates the weight of supplementary oxygen equipment, which can be exchanged for the weight when appropriate [1]. However, men do not usually carry heavier oxygen equipment than women, so the differences in carried weight to standardise the test time for sex, does not seem appropriate. The differential in weight between men and women is not common in other tests of exercise capacity [4] and may limit the comparison of the test between sexes. In addition, many individuals do not use oxygen, so the weighted backpack to simulate oxygen equipment is unnecessary.

The primary aim of this study was to validate the Glittre-ADL test performed without the backpack in people with COPD. The secondary aims were: i) to evaluate the differences between sexes in test time, physiological responses, and symptoms in performing the Glittre-ADL test with and without the backpack and to determine whether there was a difference in responses between sexes; ii) to investigate the responsiveness to change of the Glittre-ADL test performed without the backpack when assessed before and after a pulmonary rehabilitation programme.

Methods

Study design and participants

This was a randomised, cross-over equivalence study. Participants were recruited from referrals to an outpatient pulmonary rehabilitation programme at Royal Prince Alfred Hospital, Sydney, Australia and an outpatient pulmonary rehabilitation programme at Federal University of Minas Gerais, Belo Horizonte, Brazil. Participants were included in the study if they had a diagnosis of COPD [stage I to IV disease severity [5] and were stable over the past month. Exclusion criteria were: self-reported concomitant cardiovascular, orthopaedic or neurological conditions that were likely to impair exercise performance; another significant pulmonary disease; body mass index over 35 kg/m²; diagnosis of major psychiatric illness; an exacerbation during the test days. Written informed consent was obtained from all participants. The trial was registered ANZCTR 12617000920392 and was approved by Ethics Committees from Australia (X14-0199 & HREC/14/RPAH/261) and Brazil (CAAE: 02288818.0.0000.5149).

Procedures

Participants attended three data collection sessions over 14 days. On the first visit, demographic data such as height, weight and age were recorded. The participants then performed spirometry, completed the St George’s Respiratory Questionnaire (SGRQ), and completed two six-minute walk tests (6MWTs) with the best distance recorded for analysis. On the second visit, participants performed two Glittre-ADL tests (test 1 and test 2) with the backpack to eliminate the learning effect. On the third visit, participants completed one Glittre-ADL test with the backpack and one without the backpack, in random order. The order of the tests was randomised in blocks by a computer programme (https://www.random.org/). The randomisation sequence was kept in sealed, opaque envelopes and only opened prior to tests carried out on visit 3. After completing both tests, participants completed a questionnaire, which asked about the ease of performing each test, particularly control of balance and ability to perform the tasks. Participants rested for 30 minutes between each test or until all parameters had returned to baseline levels. The time to complete five laps of each Glittre-ADL test on the second and third days were recorded as the test outcomes. During all exercise tests, oxygen saturation (SpO₂) and heart rate (HR) were continuously monitored (Masimo Rad 5, Masimo Corporation or Novametric, Respironics for Australian participants and Nonin 8000 R, Nonin Medical Inc., USA, for Brazilian participants). Dyspnoea and rate of perceived exertion (RPE) were assessed before and immediately at the end of each test using the modified 0–10 Scale [6].

In a small sub-group, the responsiveness to pulmonary rehabilitation of the Glittre-ADL test performed without the backpack was tested. The participants attended an eight-week pulmonary rehabilitation programme, twice a week, in which they performed 20 minutes of stationary cycling, 20 minutes of walking, 10 minutes of upper limb strength and endurance exercises and 10 minutes of lower limb strength and endurance exercises. The initial walking exercise intensity was set at 80% of the speed-walked on 6MWT. The initial intensity of stationary cycling was set at 60% of the peak work rate estimated using the distance walked in 6MWT [7]. The intensity of both walking and stationary cycling was progressed each week with the aim to maintain dyspnoea or rate of perceived exertion (RPE) scores of 3–4. Resistance training used functional exercises such as sit-to-stand from a chair and upper limb weights. The initial weights were prescribed as tolerated aiming to achieve 8–12 repetitions for 3 sets of each exercise. The weight was progressed when participants achieved 3 sets of 12 repetitions. The order of the assessments after pulmonary rehabilitation was randomised by the same computer programme. The randomisation sequence was kept in sealed, opaque envelopes and only opened prior to tests at the end of the programme.

Measures

Glittre-ADL test

The Glittre-ADL test has been described previously [1]. In brief, participants stand up from a chair and then walk along a 10-metre track in which there are three steps to walk up and down. At the end of the track, participants perform a shelving task of moving three weights of 1.0 kg each (one by one) from the top shelf to the bottom shelf and then down to the floor and then back to the bottom shelf and the top shelf again. Participants then turn and walk back over the steps and sit down. This is one lap. Participants then immediately start the next lap. Technical details including dimensions of the stairs and adjustments of the shelf heights have been previously published [1]. The primary outcome of the test is the time taken to complete five laps. Participants are allowed to stop and rest during the test, but they are instructed to resume the test as soon as possible. During the test, participants wear a backpack filled with 2.5 kg for women or 5.0 kg for men [1]. In our study, the test performed without the backpack was identical to the Glittre-ADL test, except that participants did not wear a backpack.

Other measures

Spirometry was performed using calibrated portable spirometers (Easy One spirometer; nnd Medical Technologies Inc., Andover, MA, USA for Australian participants and Koko, Ferraris Respiratory, USA, for Brazilian participants) according to standard procedures [8]. Obtained measures were compared to normative data [9,10] and disease severity was classified according to GOLD criteria [5].

Functional capacity was measured by the distance walked in the 6MWT [11], which was performed according to previously published protocols [12].

Health-related quality of life (HRQoL) was measured by the SGRQ, which has 53 items which assess three domains: symptoms, activity limitations and impact of disease [13].

The questionnaire used to assess the ease of performing each test consisted of five questions, which investigated: whether participants had the same level of difficulty performing the Glittre-ADL with and without the backpack; if the difficulty was different, the questionnaire investigated which test was easier, and why was it easier. In addition, the questionnaire investigated how participants felt during the Glittre-ADL without the backpack compared to the Glittre-ADL with the backpack (e.g. more or less able to perform the activities during the test) and if participants had any other comment about the tests.

Sample size

The sample size was calculated based on determining whether the Glittre-ADL test performed without the backpack was equivalent to the best test performed with the backpack. SpO₂ and HR were chosen as the physiological variables of equivalence and the standard deviations (SD) of these measurements were based on a pilot study (SD for SpO₂ was 3% and for HR was 18 beats per minute) and were used in the calculation of sample size. We determined an equivalence limit of 3% for SpO₂ [14], and 14 beats per minute for HR [15] based on previously published studies. For the power of 0.90 and an alpha of 5% the required sample size was 22 participants for SpO₂ and 36 for HR.

Data analysis

Data are presented as mean and standard deviation unless otherwise stated and the normality was verified by the Shapiro-Wilk test. The comparisons between sexes for demographic, spirometric and clinical variables were performed by Student t-test for independent samples, and Mann-Whitney or chi-square test, according to the characteristic and/or variable distribution. To validate Glittre-ADL without the backpack, the Glittre-ADL test with the backpack and the 6MWT were considered as gold standards. To verify the relationship between the Glittre-ADL test time performed with and without the backpack and between the Glittre-ADL test time and the distance walked on 6MWT, the Pearson or Spearman correlation coefficient was used, according to data normality. The strength of the correlations was defined as ≥0.30 to 0.49 as weak, ≥0.50 to 0.69 as moderate, ≥0.70 as strong and 1.0 as perfect [16]. Comparisons between Glittre-ADL with backpack test times, between Glittre-ADL with and without the backpack, and responsiveness for all functional tests assessed were performed by paired t-test or Wilcoxon test. Comparisons between sexes were performed by Student t-test for independent samples or Mann-Whitney. The effect sizes of change in Glittre tests and 6MWT after pulmonary rehabilitation were calculated using Cohen’s d statistic. The classification considered was: small (d = 0.20), medium (d = 0.50) or large (d > 0.80) [17]. The level of significance was set at 5%. The Statistical Package for the Social Sciences (SPSS) v 15.0 (Chicago, IL, USA) was used for analyses.

Results

Participant characteristics

The flow chart of study participants is presented in Figure 1. Forty participants were recruited and completed the comparison of Glittre-ADL with and without the backpack, and twenty of these participants completed the comparison of the Glittre-ADL without the backpack before and after pulmonary rehabilitation. No patient presented with signs of exacerbation during the initial three testing days, however, three participants had a change in symptoms during the pulmonary rehabilitation programme and were excluded from the responsiveness analysis.

Figure 1. Flow of participants through the study.

Table 1 presents the demographic, anthropometric, spirometric, and clinical data of the participants. Three participants were classified as GOLD 1, 15 as GOLD 2, 14 as GOLD 3, and 8 as GOLD 4. Participants were divided into two subgroups according to sex. The sub-groups were similar for most characteristics. Men had more severe disease and reported a greater impact of the disease in HRQoL compared with women.

Table 1. Characteristics of study participants (n = 40).

Characteristic	All Mean (SD) n = 40	Male Mean (SD) n = 21	Female Mean (SD) n = 19	Male vs Female p-value
Age, years	70 (6)	70 (7)	70 (6)	0.731
BMI, Kg/m^2	23 (5)	24 (4)	22 (5)	0.132
FEV1, % pred	48 (20)	42 (18)	54 (20)	0.046*
FVC, % pred	83 (27)	74 (23)	92 (28)	0.034*
FEV1/FVC	0.46 (0.13)	0.46 (0.14)	0.47 (0.11)	0.559
SGRQ
Total	45 (16)	49 (16)	41 (15)	0.087
Symptoms	53 (21)	54 (20)	53 (23)	0.830
Activity	60 (19)	63 (19)	57 (20)	0.393
Impact	34 (19)	40 (20)	27 (16)	0.031*

Data presented as mean (SD). All: all participants; BMI: body mass index; FEV₁: forced expiratory volume in one second; FEV₁/FVC: the ratio between forced expiratory volume in one second and forced vital capacity; FVC: forced vital capacity; kg/m²: kilograms per metre squared; pred: predicted; SGRQ: St George’s Respiratory Questionnaire;

* significant difference between sexes.

A learning effect was evident for the Glittre-ADL test with the backpack on the second visit (test 1 and test 2), demonstrated by a significant decline in test time from test 1 to test 2 [4.84 ± 1.98 minutes vs. 4.47 ± 1.95 minutes; mean difference 0.36 minutes (95%CI 0.03 to 0.70; p = 0.036)] i.e. the second test was completed more quickly which indicates a learning effect of 9%. For the Glittre-ADL test performed with the backpack on visit 3, there was no significant difference in test time compared to test 2 on visit 2 [mean difference 0.21 minutes (95%CI −0.003 to 0.42; p = 0.053)], indicating no further learning effect.

Validity of Glittre-ADL test without the backpack and differences between sexes

Correlations between Glittre-ADL tests and between Glittre-ADL tests and 6MWT are presented in Table 2. Glittre-ADL test time performed without the backpack was strongly correlated with Glittre-ADL test time performed with the backpack for the total group, males, and females.

Table 2. Relationship between Glittre-ADL tests with and without the backpack and between the Glittre tests and the six-minute walk test.

	Glittre-ADL test time without BP, minutes
	All	Male	Female
Glittre-ADL test time with BP, minutes	rho = 0.869	rho = 0.782	rho = 0.943
	p ≤ 0.001	p ≤ 0.001	p ≤ 0.001
	Six-minute walk test distance, metres
	All	Male	Female
Glittre-ADL test time with BP, minutes	rho = −0.606	rho = −0.466	rho = −0.687
	p ≤ 0.001	p = 0.033	p = 0.001
Glittre-ADL test time without BP, minutes	rho = −0.683	rho = −0.699	rho = −0.676
	p ≤ 0.001	p ≤ 0.001	p = 0.001

Definition of abbreviations: ADL: activity daily life; All: all participants; BP: backpack; Glittre-ADL: Glittre-activities of the daily living test; rho: Spearman correlation coefficient.

Correlations between Glittre-ADL tests and 6MWT are also presented in Table 2. Glittre-ADL Test time performed with the backpack was moderately correlated with the six-minute walk distance (6MWD) for the total group, and females while the correlation for males was weak. On the other hand, the correlation for Glittre-ADL test performed without the backpack was moderate for the total group, males, and females.

Table 3 shows the results for Glittre-ADL tests performed with and without the backpack for the total group and males and females. No differences in baseline variables between tests were observed for mean ± SD HR (85 ± 16 vs. 84 ± 13, p = 0.310), SpO₂ (97 ± 3 vs. 97 ± 3, p = 0.164), median (Q1-Q3) dyspnoea [0 (0–1) vs. 0 (0–1), p = 0.707], RPE legs [0 (0–0.5) vs. 0 (0–1), p = 0.343), and RPE arms [0 (0–0) vs. 0 (0–0), p = 0.201] for Glittre-ADL test performed with and without the backpack, respectively. The time for the test was significantly shorter without the backpack for the total group and males. The Glittre-ADL test performed without the backpack provoked the same physiological responses and symptoms to the Glittre-ADL test with the backpack, except for upper limb fatigue that was reported by males as significantly higher for the test performed with the backpack. No differences were observed between sexes for the variables assessed in both tests.

Table 3 (a). Glittre-ADL test performed with and without a backpack.

	With BP Mean(SD)	Without BP Mean(SD)	With BP Mean (SD)		Without BP Mean (SD)		Without BP – With BP Mean difference (95% CI)			Male – Female Mean difference (95% CI)
	All	All	Male	Female	Male	Female	All	Male	Female	With BP	Without BP
Glittre-ADL, minutes	4.27 (1.84)	3.90 (1.44)	4.22 (1.44)	4.32 (2.24)	3.79 (1.19)	4.01 (1.70)	−0.37 (−0.59 to −0.15)*	−0.43 (−0.71 to −0.15)*	−0.31 (−0.68 to 0.06)	−0.11 (−1.30 to 1.09)	−0.23 (−1.16 to 0.71)
SpO2, %	92 (6)	91 (7)	92 (6)	92 (5)	91 (8)	92 (6)	−0.95 (−2.27 to 0.37)	−1.19 (−3.54 to 1.16)	−0.67 (−1.85 to 0.52)	−0.24 (−3.85 to 3.38)	−1.15 (−5.78 to 3.48)
HR, bpm	118 (16)	116 (16)	120 (17)	116 (15)	118 (17)	114 (14)	−1.31 (−3.92 to 1.30)	−1.29 (−5.52 to 2.96)	−1.33 (−4.61 to 1.94)	4.04 (−6.26 to 14.33)	4.16 (−5.98 to 14.29)

Definition of abbreviations: ADL: activity daily life; All: all participants; BP: backpack; bpm: beats per minute; HR: heart rate; SpO₂: oxygen saturation;

* significant difference between tests. Values for SpO₂ and HR are from the end test.

Table 3 (b). Glittre-ADL test performed with and without a backpack.

	With BP Median (Q1-Q3)	Without BP Median (IQR)	With BP Median (Q1-Q3)		Without BP Median (Q1-Q3)		Differences between tests p-value			Differences between sexes p-value
	All	All	Male	Female	Male	Female	All	Male	Female	With BP	Without BP
Dyspnoea	4 (3–5)	3 (3–5)	4 (3–5)	3 (3–4)	4 (3–5)	3 (3–4)	0.172	0.067	0.855	0.076	0.319
RPE Legs	4 (3–5)	3 (3–4)	4 (3–5)	3 (0.5–4)	4 (3–5)	3 (0.5–4)	0.200	0.461	0.244	0.554	0.305
RPE Arms	3 (0–4)	3 (0.5–3)	3 (0.5–4)	2 (0–3)	3 (1–3)	3 (0.5–3)	0.406	0.028*	0.389	0.298	0.758

Definition of abbreviations: ADL: activity daily life; All: all participants; BP: backpack; IQR: interquartile range; Q1: quartile 1; Q3: quartile 3; RPE: rate of perceived exertion;

* significant difference between tests. Values for Dyspnoea, RPE Legs, and RPE Arms are from the end test.

Seventy-eight percent of the participants reported greater ease in performing the test without the backpack. Of these, 61% mentioned greater ease in standing up from the chair and bending down and bending over during the shelving task, 57% reported greater ease in walking up and down the steps, 51% reported greater ease in walking and 39% in controlling balance. One participant reported that it was easier to perform the test with the backpack and three participants considered the same level of difficulty performing the tests.

Responsiveness of Glittre tests to pulmonary rehabilitation

Twenty of the 40 participants completed the Glittre-ADL test with and without the backpack before and after pulmonary rehabilitation. Reasons for non-completion of pulmonary rehabilitation are provided in Figure 1. Participant characteristics are presented in Table 4. Eleven participants were classified as GOLD 2, 5 as GOLD 3, and 4 as GOLD 4. After pulmonary rehabilitation, there were significant improvements in the Glittre-ADL test time performed with and without the backpack and in the 6MWD (Table 5). The effect size of the Glittre-ADL test performed without the backpack was high and similar to that of the Glittre-ADL test performed with the backpack.

Table 4. Characteristics of study participants who performed Glittre-ADL test with and without the backpack before and after pulmonary rehabilitation (n = 20).

Characteristic	Mean (SD)
Sex, M/F	13/7
Age, years	69 (7)
BMI, Kg/m^2	24 (5)
FEV1, % predicted	50 (17)
FVC, % predicted	86 (24)
FEV1/FVC	0.47 (0.14)
SGRQ
Total	46 (14)
Symptoms	55 (26)
Activity	59 (17)
Impact	35 (17)

Data presented as mean (SD). BMI: body mass index; FEV₁: forced expiratory volume in one second; FEV₁/FVC: the ratio between forced expiratory volume in one second and forced vital capacity; FVC: forced vital capacity; kg/m²: kilograms per metre squared; pred: predicted; SGRQ: St George’s Respiratory Questionnaire.

Table 5. Responsiveness for Glittre-ADL test with and without the backpack after 8 weeks of pulmonary rehabilitation (n = 20).

	Baseline Mean (SD)	Week 8 Mean (SD)	Baseline- week 8 Mean difference (95%CI)	Cohen’s d
Glittre-ADL test with BP, minutes	3.71 (0.88)	3.19 (0.60)	−0.52 (−0.81 to −0.22)*	0.82
Glittre-ADL test without BP, minutes	3.49 (0.67)	3.08 (0.54)	−0.41 (−0.63 to −0.18)*	0.86
6MWT distance, metres	479 (81)	549 (68)	49 (28 to 70)*	1.10

Definition of abbreviations: ADL: activity daily life; BP: backpack; 6MWT: six-minute walk test.

* significant difference between tests.

Discussion

The main findings of this study demonstrated that: 1) the Glittre-ADL test performed without the backpack was highly correlated with the Glittre-ADL test performed with the backpack, and moderately correlated with the 6MWD, which demonstrated that it was a valid test of function in people with COPD; 2) the time to perform the Glittre-ADL test without the backpack was significantly shorter than the Glittre-ADL test with the backpack for the total group and males; 3) the test performed without the backpack provoked similar physiological responses as the Glittre-ADL test with the backpack; 4) No differences were observed between sexes for the variables assessed in both tests; 5) the Glittre-ADL test performed without the backpack was responsive to change following a pulmonary rehabilitation programme.

The Glittre-ADL test has been increasingly used to assess the functional activity of patients with COPD [1,2,18–23]. The test has the advantage of assessing the function of both the arms and legs [3], has a strong correlation with the 6MWT [1,19,24] and shows good responsiveness to pulmonary rehabilitation [1].

In the present study, the total group and males performed the test without the backpack 9% faster than with the backpack. In the test with the backpack, participants have to carry a backpack filled with 2.5 kg for women and 5.0 kg for men, thus it was expected that the performance of the test without the backpack would make more difference for men than women.

According to the authors who proposed the test [1], the backpack simulates the weight of supplementary oxygen equipment which can be exchanged for oxygen cylinder when appropriate. However, the weight of supplementary oxygen equipment is the same for both sexes, thus, based on this, the weight for men and women in the backpack should be the same. Moreover, the fact that individuals who do not use oxygen therapy carry the backpack with the same weight that those individuals who use oxygen therapy should be questioned.

The authors attributed the necessity to double the weight in the backpack for men was to standardise the test time for sexes [1]. However, this method of correcting for sex differences makes it difficult to understand the differences in functional capacity between sexes, and also compromises the comparison with other field tests [25]. It is well known that sex influences the performance on functional tests such as the 6MWT [26], the incremental shuttle walk test and the endurance shuttle walk test [12]. However, the differences are expected and reference equations include sex as a variable [12]. Reis et al. [27] developed reference equations for the Glittre-ADL test. Only age, body mass index, and height were included as predictor variables, differently from other reference equations developed for other field tests, which include sex as a predictor variable. This may be due to the backpack, which imposes a greater weight for males, which is different from other field tests where males and females perform the test under the same conditions. Our study showed that in males, the correlation with the 6MWT was stronger with the Glittre-ADL test without the backpack compared to the Glittre-ADL test with the backpack, demonstrating that the weighted backpack was likely the reason for the difference. Therefore, the necessity to standardise test time for sexes by doubling the weight for men to decrease the physiological advantage does not seem logical.

No differences were observed between sexes for the variables assessed in both tests, even for the test performed without the backpack where men and women performed the Glittre test under the same conditions. The higher performance of men compared to women in the field walking tests is a consequence of the larger muscle mass and greater stride length observed in men [28]. These characteristics can contribute to higher performance in walking tests, but not necessarily for a test which measures functional activities for daily living like the Glittre test.

In 2008, Dechman & Scherer [25] suggested that weight adjustments for sexes could affect Glittre-ADL test validity. A number of studies since then have evaluated adjustments to the use of a backpack in modified versions of the Glittre-ADL test. Santos et al. [22] evaluated a modified shelf protocol (higher shelf) using 2.5 kg in the backpack for both sexes. The authors justified that the change was made to reduce the overload caused by the backpack weight carried by men. Monteiro et al. [29] later, in a modified version of Glittre-ADL test, used a backpack with 10% of post-bariatric surgery patient’s body weight. The same adjustment was made for obese individuals and healthy controls using 10% of their body weight in the backpack and all participants carried out each activity component of Glittre-ADL test for 2 minutes [29]. More recently, Martins et al. [30] also suggested using a backpack weight based on children’s weight according to the child’s age and sex.

Our findings showed that the test without the backpack provoked the same physiological responses and symptoms as the test with the backpack, except for the higher fatigue of upper limbs reported by males for the test performed with the backpack. Previously, Cavalheri et al. [31] have observed similar physiological responses and symptoms for patients with COPD walking on the level for one minute with a weighted backpack (5 kg for men and 2.5 kg for women) and without a backpack. In addition, the energy expenditure measured by indirect calorimetry and pedometers were also the same for these activities [31]. Similar results were observed by Monteiro et al. [29] evaluating the energy expenditure for post-bariatric surgery patients walking with and without a backpack for 2 minutes. Although only walking was evaluated in these studies rather than including all the activities in the Glittre-ADL test, these findings are an indication that the backpack may not make much difference to test performance and hence may not be necessary to evaluate functional exercise capacity. Although in our study males reported higher fatigue of the upper limbs for the test performed with the backpack, the physiological response of SpO₂ and HR were the same, indicating that the backpack might cause more discomfort but that does not make a difference to these physiological measures. The upper limb fatigue was only observed for males who had to carry double the weight of females during the test. While there were slight sex differences due to the differences of the backpack weight in the Glittre-ADL test performed with the backpack, when the total group was considered, the test without the backpack provoked the same physiological responses and symptoms as the test performed with the backpack.

The validity of the Glittre-ADL test has been previously demonstrated by comparisons with the 6MWT which showed strong correlations between Glittre-ADL time and the 6MWT distance (r = −0.82, p < 0.001) [1] and strong correlation between the oxygen uptake (VO₂) measured at the end of the Glittre-ADL test and the 6MWT (r = 0.87, p < 0.05) [19]. In addition, Glittre-ADL test time showed moderate correlations with the London Chest Activity of Daily Living (r = 0.67, p < 0.05) [2]. In our study, the Glittre-ADL test and 6MWT were chosen as the comparison tests for the Glittre-ADL test without the backpack because the Glittre-ADL test is currently considered the most accepted test of functional daily activities and the 6MWT is the most widely used test to assess functional capacity in people with COPD [12]. Our study showed that the Glittre-ADL test performed without the backpack was strongly correlated with the test performed with the backpack, and moderately correlated with the 6MWD. These findings demonstrate the validity of the Glittre-ADL test performed without the backpack to assess functional capacity for daily activities in people with COPD.

The Glittre-ADL test has been shown to be responsive to change after pulmonary rehabilitation [1,31]. Skumilen et al. [1] demonstrated a mean difference in the Glittre-ADL test time of –0.89 minutes, 95% CI (−0.48 to 1.30) after 4 weeks of pulmonary rehabilitation. In contrast, Calik-Kutukcu et al. [24] did not show improvement in the Glittre-ADL test time in people with COPD after 23 supervised pulmonary rehabilitation sessions, consisting only of arm strength training rather than a standard programme. Our study demonstrated a mean difference of –0.52 minutes, 95% CI (−0.81 to −0.22) for the test performed with the backpack and a mean difference of –0.41 minutes, 95% CI (−0.63 to −0.18) for the test performed without the backpack after eight weeks of pulmonary rehabilitation. The effect sizes for both tests were large [17], and similar to those previously reported for the Glittre-ADL test performed with the backpack [32], but lower than the effect size observed for the 6MWT in our study. The greater effect size of the 6MWT may be explained by the fact that this assessment after pulmonary rehabilitation involved only one functional activity (walking) while Glittre-ADL test with the backpack involved at least ten activities (carrying on shoulders and back, sitting, standing, walking, stair climbing, reaching, lifting, lowering objects, grasping and releasing) making the 6MWT less complex compared to the Glittre-ADL tests. Karloh et al. [19] showed that Glittre-ADL test induced a higher final VO₂ than 6MWT reinforcing that Glittre-ADL test required greater energy expenditure than the 6MWT. Recently, Gulart et al. [32], in a study of 60 participants with COPD who completed 24 sessions of pulmonary rehabilitation, determined the minimally important difference for the Glittre-ADL test performed with the backpack, as a reduction of 0.38 minutes. Participants in our study achieved this minimal important difference for the Glittre-ADL test with the backpack after 16 sessions of pulmonary rehabilitation. The minimal important difference for Glittre-ADL test performed without the backpack has not yet been determined, however, the time reduction after pulmonary rehabilitation observed in our study was greater than the proposed minimal important difference for the test performed with the backpack. A 3.5-minute cut-off point on Glittre-ADL test was shown to discriminate the functional status of individuals with COPD [33]. The Glittre-ADL test performed without the backpack did not reduce the discriminatory power of the test in our sample. Although participants performed the test without the backpack faster, the percentage of participants below the cut-off point was only 2.5% higher in the Glittre-ADL performed without the backpack (one participant).

A learning effect for Glittre-ADL test has been previously reported with test-retest repeatability varying between 6% and 7% [1,22] to 17% [24] from the first to the second test. Our findings showed a learning effect of 9% for the first two tests performed with the backpack, with no further learning effect when the test was carried out for the third time on a subsequent day.

Despite the complexity, the Glittre-ADL test has been considered a comprehensive functional test for clinical practice as it includes eleven links to the International Classification of Functioning, Disability and Health as it assesses functional upper and lower limb exercise capacity through different activities which are important for mobility and participation in daily life [3].

A limitation of this study was that there was no direct measure of metabolic or ventilatory responses during the tests. Such measures could contribute to understanding the metabolic and ventilatory differences in performing the Glittre-ADL tests with and without the backpack. In addition, the study did not include participants using oxygen therapy, therefore it is not possible to generalise the findings to this group. The sample size was calculated to evaluate the equivalence of the physiological variables of SpO₂ and HR when the Glittre-ADL without the backpack was compared to with the backpack, but not for symptoms of dyspnoea or perceived exertion. Despite this, only small differences, which were not statistically or clinically significant, were shown in dyspnoea or RPE Leg or RPE Arm when the Glittre-ADL tests with and without the backpack were compared for the total group, further supporting the equivalence of the Glittre-ADL without a backpack to that with the backpack. It would be interesting to know test-retest reliability for the Glittre-ADL without the backpack, however, our study design did not enable this assessment. Furthermore, not performing two Glittre-ADL tests with the backpack and two Glittre-ADL tests without the backpack after pulmonary rehabilitation may have affected the responsiveness, but both tests showed similar and high responsiveness.

Conclusion

In conclusion, the Glittre-ADL test performed without the backpack was shown to be a valid and responsive measure of functional activities for daily living and provoked the same physiological responses as the test with the backpack. No differences were observed between sexes for the variables assessed in both tests. The test without the backpack was reported by participants as more tolerable, than the tests with the backpack. It is reasonable to suggest that the Glittre-ADL test without the backpack is an appropriate test to assess functional capacity for activities of daily living.

Declaration of interest

The authors declare no conflict of interest.

Additional information

Funding

This work was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brazil (CAPES) Finance Code 001; Programa de Doutorado-sanduíche no Exterior (PDSE) Process number 88881.135337/2016-01; and Australian Education International, Australian Government – Department of Education and Training (Endeavour Leadership Programme).