Trial Outcomes & Findings for Comparative Study of ELLIPTA Dry Powder Inhaler (DPI) Versus DISKUS DPI Used With HandiHaler DPI in Subjects With Chronic Obstructive Pulmonary Disease (COPD) (NCT NCT03227445)
NCT ID: NCT03227445
Last Updated: 2020-07-08
Results Overview
A checklist for correct use of each inhaler was developed in Patient Instruction Leaflets (PIL's). Participants were guided by trained health care provider (HCP) to demonstrate correct use of inhaler. Baseline assessment was conducted when participant initially dispensed the inhaler. Second assessment was conducted after each 28day dosing period. Correct Use Check list was completed by HCP at each visit. Primary hypothetical estimand is the estimate of treatment effect of all participants who stayed on their randomized study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. Participants could attend visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. Participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included
COMPLETED
PHASE4
240 participants
Up to Day 56
2020-07-08
Participant Flow
This is a randomized, multi-center, open-label study comparing placebo ELLIPTA with placebo DISKUS + HandiHaler to assess correct inhaler use in participants with Chronic Obstructive Pulmonary Disease (COPD). A total of 17 centers across United States participated in the study.
Out of 240 randomized participants (par), 239 entered the study as 1 par was randomized in error. In order to minimize potential bias the order of placebo devices \& order of responses, in preference questionnaire (PQ), was randomized in a crossover design. All par on study received placebo only \& participant flow reflects this as the only treatment
Participant milestones
| Measure |
ELLIPTA/DISKUS+HANDIHALER/Q1
Participants received placebo via ELLIPTA dry powder inhaler (DPI) in period 1 and DISKUS DPI + HandiHaler in period 2, followed by preference questionnaire (PQ) version 1. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
ELLIPTA/DISKUS+HANDIHALER/Q2
Participants received placebo via ELLIPTA DPI in period 1 and DISKUS DPI + HandiHaler in period 2, followed by PQ2. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
DISKUS+HANDIHALER/ELLIPTA/Q1
Participants received placebo via DISKUS DPI + HandiHaler in period 1 and ELLIPTA DPI in period 2, followed by PQ1. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
DISKUS+HANDIHALER/ELLIPTA/Q2
Participants received placebo via DISKUS DPI + HandiHaler in period 1 and ELLIPTA DPI in period 2, followed by PQ2. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
|---|---|---|---|---|
|
Period 1 (28 Days)
STARTED
|
60
|
59
|
60
|
60
|
|
Period 1 (28 Days)
COMPLETED
|
58
|
57
|
59
|
58
|
|
Period 1 (28 Days)
NOT COMPLETED
|
2
|
2
|
1
|
2
|
|
Period 2 (28 Days)
STARTED
|
58
|
57
|
59
|
58
|
|
Period 2 (28 Days)
COMPLETED
|
57
|
56
|
57
|
55
|
|
Period 2 (28 Days)
NOT COMPLETED
|
1
|
1
|
2
|
3
|
Reasons for withdrawal
| Measure |
ELLIPTA/DISKUS+HANDIHALER/Q1
Participants received placebo via ELLIPTA dry powder inhaler (DPI) in period 1 and DISKUS DPI + HandiHaler in period 2, followed by preference questionnaire (PQ) version 1. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
ELLIPTA/DISKUS+HANDIHALER/Q2
Participants received placebo via ELLIPTA DPI in period 1 and DISKUS DPI + HandiHaler in period 2, followed by PQ2. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
DISKUS+HANDIHALER/ELLIPTA/Q1
Participants received placebo via DISKUS DPI + HandiHaler in period 1 and ELLIPTA DPI in period 2, followed by PQ1. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
DISKUS+HANDIHALER/ELLIPTA/Q2
Participants received placebo via DISKUS DPI + HandiHaler in period 1 and ELLIPTA DPI in period 2, followed by PQ2. There was no active treatment and participants continued to take their own prescribed COPD medication and rescue medications for the duration of the study
|
|---|---|---|---|---|
|
Period 1 (28 Days)
Adverse Event
|
1
|
2
|
0
|
1
|
|
Period 1 (28 Days)
Protocol Violation
|
1
|
0
|
0
|
0
|
|
Period 1 (28 Days)
Withdrawal by Subject
|
0
|
0
|
1
|
1
|
|
Period 2 (28 Days)
Adverse Event
|
1
|
0
|
2
|
1
|
|
Period 2 (28 Days)
Protocol Violation
|
0
|
0
|
0
|
1
|
|
Period 2 (28 Days)
Withdrawal by Subject
|
0
|
1
|
0
|
1
|
Baseline Characteristics
Comparative Study of ELLIPTA Dry Powder Inhaler (DPI) Versus DISKUS DPI Used With HandiHaler DPI in Subjects With Chronic Obstructive Pulmonary Disease (COPD)
Baseline characteristics by cohort
| Measure |
All Study Participants
n=239 Participants
Participants received placebo via ELLIPTA dry powder inhaler (DPI) or DISKUS DPI in combination with HandiHaler DPI based on the following treatment sequences: Treatment Sequence A (ELLIPTA in period 1 and DISKUS + HandiHaler in period 2, followed by PQ 1.); Treatment Sequence B (ELLIPTA in period 1 and DISKUS + HandiHaler in period 2, followed by PQ2); Treatment Sequence C (DISKUS + HandiHaler in period 1 and ELLIPTA in period 2, followed by PQ1); or Treatment Sequence D (DISKUS + HandiHaler in period 1 and ELLIPTA in period 2, followed by PQ2). Participants continued to take their own prescribed COPD medication and rescue medications for the entire 56-day study period.
|
|---|---|
|
Age, Continuous
|
65.9 Years
STANDARD_DEVIATION 8.25 • n=5 Participants
|
|
Sex: Female, Male
Female
|
108 Participants
n=5 Participants
|
|
Sex: Female, Male
Male
|
131 Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
African American/African Heritage
|
23 Count of Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
White - Arabic/North African Heritage
|
3 Count of Participants
n=5 Participants
|
|
Race/Ethnicity, Customized
White - White/Caucasian/European Heritage
|
213 Count of Participants
n=5 Participants
|
PRIMARY outcome
Timeframe: Up to Day 56Population: Intent-to-treat (ITT) Population comprised of all randomized participants, excluding those who were randomized in error, and did not have at least one error assessment at Visit 1. Only those participants with data available at specified time point were analyzed.
A checklist for correct use of each inhaler was developed in Patient Instruction Leaflets (PIL's). Participants were guided by trained health care provider (HCP) to demonstrate correct use of inhaler. Baseline assessment was conducted when participant initially dispensed the inhaler. Second assessment was conducted after each 28day dosing period. Correct Use Check list was completed by HCP at each visit. Primary hypothetical estimand is the estimate of treatment effect of all participants who stayed on their randomized study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. Participants could attend visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. Participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Percentage of Participants With Zero Errors After 28 Days of Inhaler Use in Each Treatment Phase (Primary Estimand: Hypothetical)
|
96 Percentage of participants
|
87 Percentage of participants
|
—
|
PRIMARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
Primary estimand is the treatment effect estimated in participants who stayed on their randomised study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. The participant could attend the visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. For primary estimand, a sensitivity analysis using Cochran-Mantel-Haenszel test was performed on participants with discordant results. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Percentage of Participants With Atleast One Error After 28 Days of Inhaler Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Atleast 1 error-ELLIPTA&0 errors-DISKUS+HandiHaler
|
3 Percentage of participants
|
—
|
—
|
|
Percentage of Participants With Atleast One Error After 28 Days of Inhaler Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Atleast 1error-DISKUS+HandiHaler&0errors-ELLIPTA
|
23 Percentage of participants
|
—
|
—
|
PRIMARY outcome
Timeframe: Up to Day 56Supplementary estimand estimated the composite effect of initial randomized treatment. A sensitivity analysis using the Cochran-Mantel-Haenszel test was performed on participants with discordant results. Participants who withdrew during period 2 were included in the analysis using early withdrawal data where available or imputation otherwise. Participants who experienced an intercurrent event in Period 1 were excluded from the analysis.
Outcome measures
| Measure |
ELLIPTA
n=224 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Percentage of Participants With Atleast One Error After 28 Days of Inhaler Use in Each Treatment Phase (Supplementary Estimand: Composite)
Atleast 1 error-ELLIPTA&0 errors-DISKUS+HandiHaler
|
4 Percentage of participants
|
—
|
—
|
|
Percentage of Participants With Atleast One Error After 28 Days of Inhaler Use in Each Treatment Phase (Supplementary Estimand: Composite)
Atleast 1error-DISKUS+HandiHaler&0errors-ELLIPTA
|
27 Percentage of participants
|
—
|
—
|
PRIMARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
Supplementary estimand estimated the composite effect of initial randomized treatment. The analysis was performed using stratified exact logistic model. Participants were included in the model as fixed strata, treatment option was included in the exact statement and period included as fixed effects. Participants who withdrew during period 2 were included in the analysis using early withdrawal data where available or imputation otherwise. Participants who experienced an intercurrent event in Period 1 were excluded from the analysis.
Outcome measures
| Measure |
ELLIPTA
n=224 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=224 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Percentage of Participants With Zero Errors After 28 Days of Inhaler Use in Each Treatment Phase(Supplementary Estimand: Composite)
|
95 Percentage of participants
|
85 Percentage of participants
|
—
|
SECONDARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
The occurrence of each type of error for each inhaler (ELLIPTA, DISKUS or HANDIHALER) were evaluated based on the information collected in Correct Use Checklists. Primary estimand is the treatment effect estimated in participants who stayed on their randomised study device sequence \& did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. The participant could attend the visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. For primary estimand, a sensitivity analysis using Cochran-Mantel-Haenszel test was performed on participants with discordant results. Only those participants who completed error assessments for both randomized treatment groups \& experienced no intercurrent events were included. The number of error is reported as NA for the type of error which was not applicable to the particular inhaler type
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par exhales while holding inhaler away from mouth
|
6 Errors
|
9 Errors
|
14 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par takes long steady deep breath in through mouth
|
2 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par breathes out slowly and gently
|
1 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par opened cover
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par slid lever away from mouthpiece
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par didn't tilt/shake device post dose preparation
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
2 Errors
|
1 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par took a quick deep breath through the Diskus
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par removed Diskus from mouth,held breath for long
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
2 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par breathed out slowly
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par closed the cover of the inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
2 Errors
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par did not exhale into the mouthpiece
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
10 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par opened mouthpiece & center chamber is showing
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par opened only 1 capsule from the blister card
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
5 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par placed capsule in center chamber of inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par pressed green piercing button&released button
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
4 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par exhaled fully
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
7 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par closed lips tightly around mouthpiece
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
9 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par didn't block air intake vents with fingers
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
9 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par breathed in deeply
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
10 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par heard or felt the capsule vibrate
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
14 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par held breath for few seconds & removed inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
10 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par resumed normal breathing
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
9 Errors
|
|
Number of Errors by Type for Each Inhaler After 28 Days of Use in Each Treatment Phase (Primary Estimand: Hypothetical)
Par didn't pierce use capsule for 2nd inhalation
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
NA Errors
NA indicates error not applicable for that particular inhaler
|
1 Errors
|
SECONDARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
Participants were provided with PIL explaining correct use of inhaler. Overall error includes both critical and non-critical errors. Primary estimand is the treatment effect estimated in participants who stayed on their randomised study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. The participant could attend the visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. For primary estimand, a sensitivity analysis using Cochran-Mantel-Haenszel test was performed on participants with discordant results. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Number of Errors Per Participant for Each Treatment Group After 28 Days of Use (Primary Estimand: Hypothetical)
|
0.0 Errors per participant
Interval 0.0 to 1.0
|
0.6 Errors per participant
Interval 0.0 to 9.0
|
—
|
SECONDARY outcome
Timeframe: Day 1 and Day 28 of each treatment groupPopulation: ITT Population. Only those participants with data available at specified time point were analyzed.
Assessment of errors was conducted by HCP trained in the correct inhaler use of the three inhalers based on the checklist of errors. The median number of overall errors made by per participant was assessed for ELLIPTA and DISKUS + HandiHaler both on Day 1 and Day 28 for each treatment group. The difference was calculated by subtracting values of Day 28 from Day1 for each treatment regimen. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Change in Errors Per Participant for Each Treatment Group After 28 Days of Use
|
0.0 Errors per participant
Interval -3.0 to 1.0
|
0.0 Errors per participant
Interval -10.0 to 9.0
|
—
|
SECONDARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
Participants were provided with the PIL, explaining correct use of the inhaler. Overall error includes both critical and non-critical errors. Assessment of errors was conducted by HCP trained in the correct inhaler use of the three inhalers based on the checklist of errors. Primary estimand is the treatment effect estimated in participants who stayed on their randomised study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. The participant could attend the visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. For primary estimand, a sensitivity analysis using Cochran-Mantel-Haenszel test was performed on participants with discordant results. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Number of Errors for Each Treatment Group in Participants With One or More Errors After 28 Days of Use (Primary Estimand: Hypothetical)
|
1.0 Errors per participant
Interval 1.0 to 1.0
|
4.4 Errors per participant
Interval 1.0 to 9.0
|
—
|
SECONDARY outcome
Timeframe: Day 1 and Day 28 of each treatment groupPopulation: ITT Population. Only those participants with data available at specified time point were analyzed.
Assessment of errors was conducted by health care professionals trained in the correct inhaler use of the three inhalers based on the checklist of errors. The median of overall errors made by each participant was assessed for ELLIPTA and DISKUS + HandiHaler both on Day 1 and Day 28. The difference was calculated by subtracting values of Day 28 from Day1 for each treatment regimen. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Change in Errors for Each Treatment Group in Participants With One or More Errors After 28 Days of Use
|
0.7 Errors per participant
Interval -1.0 to 1.0
|
2.6 Errors per participant
Interval -4.0 to 9.0
|
—
|
SECONDARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
A checklist for correct use of each inhaler was developed based on the steps identified in the PIL. A critical error was defined as an error that was most likely to result in no, or a significantly reduced amount, of medication being inhaled by the participant. Primary estimand is the treatment effect estimated in participants who stayed on their randomised study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. The participant could attend the visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. For primary estimand, a sensitivity analysis using Cochran-Mantel-Haenszel test was performed on participants with discordant results. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
n=217 Participants
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Number of Participants With Zero Critical Errors After 28 Days of Each Treatment Group Use (Primary Estimand: Hypothetical)
|
215 Participants
|
193 Participants
|
—
|
SECONDARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
A critical error was defined as an error that was most likely to result in no, or a significantly reduced amount, of medication being inhaled by the participant. Primary estimand is the treatment effect estimated in participants who stayed on their randomised study device sequence and did not change their standard COPD maintenance medication device to one delivered via ELLIPTA, DISKUS or HANDIHALER, throughout both 28 day treatment periods. The participant could attend the visit without the device/s they were randomised to, in which case correct use cannot be assessed as described in the protocol. For primary estimand, a sensitivity analysis using Cochran-Mantel-Haenszel test was performed on participants with discordant results. Only those participants who completed error assessments for both randomized treatment groups and experienced no intercurrent events were included.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Number of Participants With Atleast One Critical Error After 28 Days of Each Treatment Group Use (Primary Estimand: Hypothetical)
Atleast 1 critical error with ELLIPTA
|
1 Participants
|
—
|
—
|
|
Number of Participants With Atleast One Critical Error After 28 Days of Each Treatment Group Use (Primary Estimand: Hypothetical)
Atleast 1 critical error with DISKUS+HandiHaler
|
23 Participants
|
—
|
—
|
SECONDARY outcome
Timeframe: Up to Day 56Population: ITT Population. Only those participants with data available at specified time point were analyzed.
Supplementary estimand estimated the composite effect of initial randomized treatment. A sensitivity analysis using the Cochran-Mantel-Haenszel test was performed on participants with discordant results. Participants who withdrew during period 2 were included in the analysis using early withdrawal data where available or imputation otherwise. Participants who experienced an intercurrent event in Period 1 were excluded from the analysis.
Outcome measures
| Measure |
ELLIPTA
n=217 Participants
Participants were randomized to use placebo via ELLIPTA in either period 1 or 2.
|
DISKUS + HANDIHALER
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
HandiHaler
Participants were randomized to use placebo via DISKUS + HANDIHALER in period 1 or 2.
|
|---|---|---|---|
|
Number of Participants With Atleast One Critical Error After 28 Days of Each Treatment Group Use (Supplementary Estimand: Composite)
Atleast 1 error-ELLIPTA&0 errors-DISKUS+HandiHaler
|
3 Participants
|
—
|
—
|
|
Number of Participants With Atleast One Critical Error After 28 Days of Each Treatment Group Use (Supplementary Estimand: Composite)
Atleast 1error-DISKUS+HandiHaler&0errors-ELLIPTA
|
27 Participants
|
—
|
—
|
Adverse Events
All Study Participants
Serious adverse events
| Measure |
All Study Participants
n=239 participants at risk
Participants received placebo via ELLIPTA dry powder inhaler (DPI) or DISKUS DPI in combination with HandiHaler DPI based on the following treatment sequences: Treatment Sequence A (ELLIPTA in period 1 and DISKUS + HandiHaler in period 2, followed by PQ 1.); Treatment Sequence B (ELLIPTA in period 1 and DISKUS + HandiHaler in period 2, followed by PQ2); Treatment Sequence C (DISKUS + HandiHaler in period 1 and ELLIPTA in period 2, followed by PQ1); or Treatment Sequence D (DISKUS + HandiHaler in period 1 and ELLIPTA in period 2, followed by PQ2). Participants continued to take their own prescribed COPD medication and rescue medications for the entire 56-day study period.
|
|---|---|
|
Respiratory, thoracic and mediastinal disorders
Chronic obstructive pulmonary disease
|
2.1%
5/239 • Number of events 5 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Respiratory, thoracic and mediastinal disorders
Acute respiratory failure
|
0.84%
2/239 • Number of events 2 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Respiratory, thoracic and mediastinal disorders
Pneumonia aspiration
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Infections and infestations
Pneumonia
|
1.3%
3/239 • Number of events 3 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Infections and infestations
Bronchitis
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Cardiac disorders
Cardiac failure congestive
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
General disorders
Chest pain
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Metabolism and nutrition disorders
Hyponatraemia
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Nervous system disorders
Dysarthria
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
|
Nervous system disorders
Hemiparesis
|
0.42%
1/239 • Number of events 1 • Up to Day 56
Intent-to-Treat Population was used to collect Adverse Events. Since all participants received placebo as treatment, data has been clubbed for all the arms. The non-serious AEs are reported using a frequency threshold of \>5%
|
Other adverse events
Adverse event data not reported
Additional Information
Results disclosure agreements
- Principal investigator is a sponsor employee GSK agreements may vary with individual investigators, but will not prohibit any investigator from publishing. GSK supports the publication of results from all centers of a multi-center trial but requests that reports based on single-site data not precede the primary publication of the entire clinical trial.
- Publication restrictions are in place
Restriction type: OTHER