Clinical Trial Results:
Trial of Optimal Therapy for Pseudomonas Eradication in Cystic Fibrosis
Summary


EudraCT number 
200901257510 
Trial protocol 
GB SE 
Global end of trial date 
10 Apr 2018

Results information


Results version number 
v1(current) 
This version publication date 
10 Apr 2019

First version publication date 
10 Apr 2019

Other versions 

Summary report(s) 
Protocol Amendments V2.0, V4.0 and V6.0 
Trial Information
Subject Disposition
Baseline Characteristics
End Points
Adverse Events
More Information
Subject Disposition
Baseline Characteristics
End Points
Adverse Events
More Information


Trial identification


Sponsor protocol code 
CH/2007/2661


Additional study identifiers


ISRCTN number 
ISRCTN02734162  
US NCT number 
  
WHO universal trial number (UTN) 
  
Other trial identifiers 
Funder reference: HTA 07/51/01  
Sponsors


Sponsor organisation name 
University Hospitals Bristol NHS Foundation Trust


Sponsor organisation address 
Bristol Royal Children's Hospital, Upper Maudlin Street, Bristol, United Kingdom, BS2 8BJ


Public contact 
Ashley Jones, Clinical Trials Research Centre, University of Liverpool, +44 151 795 8751, ctrcqa@liverpool.ac.uk


Scientific contact 
Ashley Jones, Clinical Trials Research Centre, University of Liverpool, +44 151 795 8751, ctrcqa@liverpool.ac.uk


Sponsor organisation name 
University of Liverpool


Sponsor organisation address 
Research Support Office, 2nd Floor Block D, Waterhouse Building, 3 Brownlow Street, Liverpool, United Kingdom, L69 3GL


Public contact 
Ashley Jones, Clinical Trials Research Centre, University of Liverpool, +44 151 795 8751, ctrcqa@liverpool.ac.uk


Scientific contact 
Ashley Jones, Clinical Trials Research Centre, University of Liverpool, +44 151 795 8751, ctrcqa@liverpool.ac.uk


Paediatric regulatory details


Is trial part of an agreed paediatric investigation plan (PIP) 
No


Does article 45 of REGULATION (EC) No 1901/2006 apply to this trial? 
No


Does article 46 of REGULATION (EC) No 1901/2006 apply to this trial? 
No


Results analysis stage


Analysis stage 
Final


Date of interim/final analysis 
27 Sep 2018


Is this the analysis of the primary completion data? 
Yes


Primary completion date 
10 Apr 2018


Global end of trial reached? 
Yes


Global end of trial date 
10 Apr 2018


Was the trial ended prematurely? 
No


General information about the trial


Main objective of the trial 
This trial assessed whether fourteen days intravenous ceftazidime with tobramycin is superior to three months (12 weeks) oral ciprofloxacin. Both treatment regimes were in conjunction with three months (12 weeks) nebulised colistin.
Primary outcome:
Successful eradication of P. aeruginosa infection at three months (12 weeks) after allocated treatment had started, remaining infection free through to 15 months (60 weeks) after the start of allocated treatment.


Protection of trial subjects 
There were no formal accountability measures required for the trial. TORPEDOCF has put the
following measures in place to safeguard the safety of trial participants:
 Participant completed treatment diaries to record treatment compliance
 Collection of unused trial IMP
 Collection of packaging of used IMP
 Verbal interview conducted by the PI at the 3 month followup visit
 Local procedures should be used if a manufacturer issues a recall


Background therapy 
Participants recruited into the study were randomised to one of the following treatment arms: Arm A: 14 days* intravenous (IV) antibiotics as follows:  Ceftazidime 150 milligram (mg)/kilogram (kg)/day, in 3 divided doses (maximum of 3 grams (g) three times daily (tds)). Some centres used a once daily continuous infusion (where the maximum daily dose would usually be 6g/day) or twice daily regimen for ceftazidime. These centres continued to use this regimen for the study and followed their local dosing guidelines.  Tobramycin 10mg/kg/day once daily (od) (maximum 660mg /day). Some centres used a twice daily or three times daily regimen for tobramycin. These centres continued to use their current regimen for the study and should follow their local dosing guidelines. Therapeutic drug monitoring should be used to guide tobramycin dosing as per national guidelines (https://www.cysticfibrosis.org.uk/media/82010/CD_Antibiotic_treatment_for_CF_May_09.pdf) and usual clinic procedures. *Recommended treatment duration should be 14 days, minimum treatment duration should be no less than 10 days Arm B: 3 months oral ciprofloxacin twice daily (bd). (Ciprofloxacin dose will be 20 mg/kg twice daily (maximum 750mg twice daily. This is in line with the BNF for children (http://bnfc.org/bnfc/). Some clinicians prefered to use a lower dose of 15mg/kg twice daily for children under 5 years, as used in national CF guidelines (https://www.cysticfibrosis.org.uk/media/82010/CD_Antibiotic_treatment_for_CF_May_09.pdf). Both treatment arms received 3 months#of nebulised colistin in conjunction to the randomised treatment. Colistin dose was as recommended by the UK CF Trust: 1,000,000 units twice daily for children aged ≤ 2 years and 2,000,000 units twice daily for children aged > 2 years and adults. If the colistin was administered via an Ineb a lower dose of 1,000,000 units twice daily for all ages should be used.  
Evidence for comparator 
The rationale for choosing fourteen days of intravenous treatment and for choosing three months for oral treatment is that both of these are standard practice for many UK CF centres identified in the HTA feasibility study and both are standard recommendations within the published UK guideline and believed to represent current best practice.  
Actual start date of recruitment 
18 Jun 2010


Long term followup planned 
No


Independent data monitoring committee (IDMC) involvement? 
Yes


Population of trial subjects


Number of subjects enrolled per country 

Country: Number of subjects enrolled 
United Kingdom: 284


Country: Number of subjects enrolled 
Italy: 2


Worldwide total number of subjects 
286


EEA total number of subjects 
286


Number of subjects enrolled per age group 

In utero 
3


Preterm newborn  gestational age < 37 wk 
0


Newborns (027 days) 
0


Infants and toddlers (28 days23 months) 
70


Children (211 years) 
161


Adolescents (1217 years) 
37


Adults (1864 years) 
15


From 65 to 84 years 
0


85 years and over 
0



Recruitment


Recruitment details 
60 UK based centres and 1 centre based in Italy took part in the trial. The first patient was randomised on 5th October 2010 and the last patient was randomised on 27th January 2017.  
Preassignment


Screening details 
There were 1522 screenings from 1308 patients for eligibility and of those 554 were ineligible, 193 were not approached, 489 didn't provide consent and 3 patients were excluded but no reason was given. 286 patients were randomised into the trial. (Note that patients could be screened more than once)  
Period 1


Period 1 title 
Baseline


Is this the baseline period? 
Yes  
Allocation method 
Randomised  controlled


Blinding used 
Not blinded  
Blinding implementation details 
Randomisation lists were generated in a 1:1 ratio using simple block randomisation with random variable block lengths. Factors within the protocol that were used to stratify randomisation were not disclosed to prevent prediction in this open trial.


Arms


Are arms mutually exclusive 
Yes


Arm title

Intravenous (IV) antibiotics  
Arm description 
14 days IV antibiotics; Ceftazidime and Tobramycin alongside three months of nebulised Colistin  
Arm type 
Experimental  
Investigational medicinal product name 
Ceftazidime


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Injection


Routes of administration 
Intravenous use


Dosage and administration details 
Ceftazidime 150 milligram (mg)/kilogram (kg)/day, in 3 divided doses (maximum of 3 grams (g) three times daily (tds)). Some centres used a once daily continuous infusion (where the maximum daily dose would usually be 6g/day) or twice daily regimen for ceftazidime.


Investigational medicinal product name 
Tobramycin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Injection


Routes of administration 
Intravenous use


Dosage and administration details 
Tobramycin 10mg/kg/day once daily (od) (maximum 660mg /day). Some centres used a
twice daily or three times daily regimen for tobramycin.


Investigational medicinal product name 
Colistin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Nebuliser solution


Routes of administration 
Inhalation use


Dosage and administration details 
3 months (12 weeks) of nebulised colistin in conjunction to the randomised treatment. Colistin dose was as recommended by the UK CF Trust: 1,000,000 units twice daily for children aged ≤ 2 years and 2,000,000 units twice daily for children aged > 2 years and adults. If the colistin is administered via an Ineb a lower dose of 1,000,000 units twice daily for all ages was used.


Arm title

Oral antibiotic therapy  
Arm description 
3 months (12 weeks) of oral ciprofloxacin alongside 3 months (12 weeks) of nebulised colistin.  
Arm type 
Active comparator  
Investigational medicinal product name 
Ciprofloxacin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Granules and solvent for oral suspension, Tablet


Routes of administration 
Oral use


Dosage and administration details 
3 months (12 weeks) oral ciprofloxacin twice daily (bd). (Ciprofloxacin dose was 20 mg/kg twice daily (maximum 750mg twice daily. Some clinicians could use a lower dose of 15mg/kg twice daily for children under 5 years, as used in national CF guidelines.


Investigational medicinal product name 
Colistin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Nebuliser solution


Routes of administration 
Inhalation use


Dosage and administration details 
3 months (12 weeks) of nebulised colistin in conjunction to the randomised treatment. Colistin dose was as recommended by the UK CF Trust: 1,000,000 units twice daily for children aged ≤ 2 years and 2,000,000 units twice daily for children aged > 2 years and adults. If the colistin is administered via an Ineb a lower dose of 1,000,000 units twice daily for all ages was used.




Period 2


Period 2 title 
Followup


Is this the baseline period? 
No  
Allocation method 
Not applicable


Blinding used 
Not blinded  
Blinding implementation details 
See blinding details for the baseline period.


Arms


Are arms mutually exclusive 
Yes


Arm title

Intravenous (IV) antibiotics  
Arm description 
14 days IV antibiotics; Ceftazidime and Tobramycin alongside three months nebulised Colistin  
Arm type 
Experimental  
Investigational medicinal product name 
Ceftazidime


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Infusion


Routes of administration 
Intravenous use


Dosage and administration details 
Ceftazidime 150 milligram (mg)/kilogram (kg)/day, in 3 divided doses (maximum of 3 grams (g) three times daily (tds)). Some centres used a once daily continuous infusion (where the maximum daily dose would usually be 6g/day) or twice daily regimen for ceftazidime.


Investigational medicinal product name 
Tobramycin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Infusion


Routes of administration 
Intravenous use


Dosage and administration details 
Tobramycin 10mg/kg/day once daily (od) (maximum 660mg /day). Some centres used a
twice daily or three times daily regimen for tobramycin.


Investigational medicinal product name 
Colistin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Nebuliser solution


Routes of administration 
Inhalation use


Dosage and administration details 
3 months (12 weeks) of nebulised colistin in conjunction to the randomised treatment. Colistin dose was as recommended by the UK CF Trust: 1,000,000 units twice daily for children aged ≤ 2 years and 2,000,000 units twice daily for children aged > 2 years and adults. If the colistin is administered via an Ineb a lower dose of 1,000,000 units twice daily for all ages was used.


Arm title

Oral antibiotic therapy  
Arm description 
3 months (12 weeks) of oral ciprofloxacin alongside 3 months (12 weeks) of nebulised colistin.  
Arm type 
Active comparator  
Investigational medicinal product name 
Ciprofloxacin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Tablet


Routes of administration 
Oral use


Dosage and administration details 
3 months (12 weeks) oral ciprofloxacin twice daily (bd). (Ciprofloxacin dose was 20 mg/kg twice daily (maximum 750mg twice daily. Some clinicians preferred to use a lower dose of 15mg/kg twice daily for children under 5 years, as used in national CF guidelines.


Investigational medicinal product name 
Colistin


Investigational medicinal product code 

Other name 

Pharmaceutical forms 
Nebuliser solution


Routes of administration 
Inhalation use


Dosage and administration details 
3 months (12 weeks) of nebulised colistin in conjunction to the randomised treatment. Colistin dose was as recommended by the UK CF Trust: 1,000,000 units twice daily for children aged ≤ 2 years and 2,000,000 units twice daily for children aged > 2 years and adults. If the colistin is administered via an Ineb a lower dose of 1,000,000 units twice daily for all ages was used.





Baseline characteristics reporting groups


Reporting group title 
Intravenous (IV) antibiotics


Reporting group description 
14 days IV antibiotics; Ceftazidime and Tobramycin alongside three months of nebulised Colistin  
Reporting group title 
Oral antibiotic therapy


Reporting group description 
3 months (12 weeks) of oral ciprofloxacin alongside 3 months (12 weeks) of nebulised colistin.  



End points reporting groups


Reporting group title 
Intravenous (IV) antibiotics


Reporting group description 
14 days IV antibiotics; Ceftazidime and Tobramycin alongside three months of nebulised Colistin  
Reporting group title 
Oral antibiotic therapy


Reporting group description 
3 months (12 weeks) of oral ciprofloxacin alongside 3 months (12 weeks) of nebulised colistin.  
Reporting group title 
Intravenous (IV) antibiotics


Reporting group description 
14 days IV antibiotics; Ceftazidime and Tobramycin alongside three months nebulised Colistin  
Reporting group title 
Oral antibiotic therapy


Reporting group description 
3 months (12 weeks) of oral ciprofloxacin alongside 3 months (12 weeks) of nebulised colistin. 


End point title 
Primary efficacy assessment  Successful eradication of P. aeruginosa three months after the start of treatment, remaining infection free through to 15 months after the start of treatment  
End point description 

End point type 
Primary


End point timeframe 
3 months after the start of treatment through to 15 months after the start of treatment .




Statistical analysis title 
Successful eradication from 3 months to 15 months  
Statistical analysis description 
The number of patients who were classified as (a) a success and (b) failure for the primary outcome (and percentages) were presented for each treatment arm. The relative risk together with 95% confidence interval was reported along with a twosided pvalue from a chisquared test.


Comparison groups 
Oral antibiotic therapy v Intravenous (IV) antibiotics


Number of subjects included in analysis 
255


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.184  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.84


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.65  
upper limit 
1.09 


End point title 
Primary efficacy assessment  Successful eradication of P. aeruginosa three months after the start of treatment, remaining infection free through to 15 months after the start of treatment  
End point description 

End point type 
Primary


End point timeframe 
From 3 months after the start of treatment to 15 months after the start of treatment.




Statistical analysis title 
Unsuccessful eradication at 3 months  
Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
226


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.037  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
2.74


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.01  
upper limit 
7.44 


End point title 
Primary efficacy assessment – Sensitivity analysis 6: no T15 window (Post hoc)  
End point description 

End point type 
Primary


End point timeframe 
From 3 months after the start of treatment to 15 months after the start of treatment.




Statistical analysis title 
Sensitivity analysis 6  
Statistical analysis description 
The relative risk with 95% CI was presented. A chisquared test will be used to calculate a pvalue for this relative risk.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
280


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.317  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.88


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.68  
upper limit 
1.13 


End point title 
Time to reoccurrence of original P.aeruginosa infection  unknown strains assumed to be same as baseline  
End point description 

End point type 
Secondary


End point timeframe 
3 months to 24 months. Only patients who have a baseline sample and at least one sample post three months will be included in this analysis.




Statistical analysis title 
Time to reoccurrence of original P.aeruginosa  
Statistical analysis description 
Difference between the two treatment arms was tested using the logrank test. Cox proportional hazards regression was used to calculate a hazard ratio with 95% CI, comparing the IV treatment group to oral, – unknown strains assumed to be same as baseline.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
285


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.061  
Method 
Logrank  
Parameter type 
Hazard ratio (HR)  
Point estimate 
1.37


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.99  
upper limit 
1.91 


End point title 
Time to reoccurrence of original P.aeruginosa infection  unknown strains assumed to be different to baseline  
End point description 

End point type 
Secondary


End point timeframe 
From 3 to 24 months.




Statistical analysis title 
Time to reoccurrence of original P.aeruginosa  
Statistical analysis description 
Difference between the two treatment arms was tested using the logrank test. Cox proportional hazards regression was used to calculate a hazard ratio with 95% CI, comparing the IV treatment group to oral, – unknown strains assumed to be differnt than baseline.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
285


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.075  
Method 
Logrank  
Parameter type 
Hazard ratio (HR)  
Point estimate 
1.85


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.94  
upper limit 
3.64 


End point title 
Reinfection with a different strain of P.aeruginosa  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 15 months.




Statistical analysis title 
Reinfection with a different strain  
Statistical analysis description 
The relative risk together with 95% confidence interval was reported along with a twosided pvalue from a chisquared test.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
42


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.733  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.82


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.3  
upper limit 
2.25 


End point title 
Post hoc analysis – Genotyping results  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




No statistical analyses for this end point 


End point title 
Lung function  FEV1% predicted  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [1]  At baseline n=67, T3 n=25, T15 n=32, T24 n= 11 [2]  At baseline n=70, T3 n=28, T15 n=32 , T24 n=17 

Statistical analysis title 
FEV1 % predicted  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.008  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
3.75


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.97  
upper limit 
6.53  
Statistical analysis title 
FEV1 % predicted  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.184  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
2.08


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.99  
upper limit 
5.14  
Statistical analysis title 
FEV1 % predicted  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures will be fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.705  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
0.82


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.45  
upper limit 
5.1  
Statistical analysis title 
FEV1 % predicted  treat difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.019  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
3.16


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.51  
upper limit 
5.8 


End point title 
Lung function  FVC% predicted  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [3]  At baseline n=67, T3 n= 25, T15 n=32 and T24 n=11 [4]  At baseline n=70, T3 n=28, T15 n=32 and T24 n=17 

Statistical analysis title 
FVC % predicted  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.008  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
3.86


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.03  
upper limit 
6.69  
Statistical analysis title 
FVC % predicted  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.0396  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
3.14


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.15  
upper limit 
6.14  
Statistical analysis title 
FVC % predicted  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.216  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
2.61


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.52  
upper limit 
6.73  
Statistical analysis title 
FVC % predicted  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
137


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.009  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
3.56


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.9  
upper limit 
6.21 


End point title 
Lung function  FEF2575% predicted  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [5]  At baseline n=44, T3 n=18, T15 n=26 and T24 n=7 [6]  At baseline n=53, T3 n=20, T15 n=21 and T24 n=12 

Statistical analysis title 
FEV2575% predicted  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
97


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.274  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
3.76


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.99  
upper limit 
10.52  
Statistical analysis title 
FEV2575% predicted  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
97


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.345  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
3.46


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.74  
upper limit 
10.66  
Statistical analysis title 
FEV2575% predicted  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
97


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.505  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
3.23


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.29  
upper limit 
12.76  
Statistical analysis title 
FEV2575% predicted  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
97


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.269  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
3.64


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.82  
upper limit 
10.11 


End point title 
Oxygen saturation  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [7]  At baseline n=118, T3 n= 42 , t15 n=48 and T24 n=17 [8]  At baseline n=133, T3 n=53, T15 n=46 and T24 n=26 

Statistical analysis title 
O2 Saturation  3 month treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
251


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.728  
Method 
Mixed models analysis  
Parameter type 
3 Month Treatment difference  
Point estimate 
0.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.21  
upper limit 
0.3  
Statistical analysis title 
O2 Saturation  15 month treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
251


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.709  
Method 
Mixed models analysis  
Parameter type 
15 Month Treatment difference  
Point estimate 
0.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.2  
upper limit 
0.29  
Statistical analysis title 
O2 Saturation  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
251


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.686  
Method 
Mixed models analysis  
Parameter type 
Treatment difference  
Point estimate 
0.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.18  
upper limit 
0.27  
Statistical analysis title 
O2 Saturation  24 Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
251


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.79  
Method 
Mixed models analysis  
Parameter type 
24 month Treatment difference  
Point estimate 
0.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.3  
upper limit 
0.4 


End point title 
Growth and nutritional status  Height zscores  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [9]  At baseline n=125, T3 n=32, T15 n=36 and T24 n=13 [10]  At baseline n=131, T3 n=47, T15 n=32 and T24 n=16 

Statistical analysis title 
Height zscores  3 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures will be fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.952  
Method 
Mixed models analysis  
Parameter type 
3 Month treatment difference  
Point estimate 
0


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.06  
upper limit 
0.05  
Statistical analysis title 
Height zscores  15 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.572  
Method 
Mixed models analysis  
Parameter type 
15 Month treatment difference  
Point estimate 
0.03


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.13  
upper limit 
0.07  
Statistical analysis title 
Height zscores  24 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.543  
Method 
Mixed models analysis  
Parameter type 
24 Month treatment difference  
Point estimate 
0.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.21  
upper limit 
0.11  
Statistical analysis title 
Height zscores  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.939  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.06  
upper limit 
0.05 


End point title 
Growth and nutritional status  Weight zscores  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [11]  At baseline n=102, T3 n=30, T15 n=27, T24 n=12 [12]  At baseline n=109, T3 n=37, T15 n=23, T24 n=11 

Statistical analysis title 
Weight zscores  3 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
211


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.988  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
0


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.06  
upper limit 
0.06  
Statistical analysis title 
Weight zscores  15 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
211


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.794  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
0.02


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.15  
upper limit 
0.11  
Statistical analysis title 
Weight zscores  24 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
211


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.78  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
0.03


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.24  
upper limit 
0.18  
Statistical analysis title 
Weight zscores  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
211


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.987  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.06  
upper limit 
0.06 


End point title 
Growth and nutritional status  BMI zscores  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [13]  At baseline n=125, T3 n=32, T15 n=36 and T24 n=13 [14]  At baseline n=131, T3 n=47, T15 n=32 and T24 n=16 

Statistical analysis title 
BMI zscore  3 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.767  
Method 
Mixed models analysis  
Parameter type 
3 Month Treatment Difference  
Point estimate 
0.01


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.08  
upper limit 
0.06  
Statistical analysis title 
BMI zscore  15 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.912  
Method 
Mixed models analysis  
Parameter type 
15 Month Treatment Difference  
Point estimate 
0.01


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.14  
upper limit 
0.16  
Statistical analysis title 
BMI zscore  24 Month Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.854  
Method 
Mixed models analysis  
Parameter type 
24 Month Treatment Difference  
Point estimate 
0.02


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.22  
upper limit 
0.26  
Statistical analysis title 
BMI zscore  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
256


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.774  
Method 
Mixed models analysis  
Parameter type 
Treatment Difference  
Point estimate 
0.01


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.08  
upper limit 
0.06 


End point title 
Growth and nutritional status  BMI  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 24 months.




Notes [15]  At baseline n=7, T3 n=2, T15 n=2 and T24 n=2 [16]  At baseline n=10, T3 n=2, T15 n=3 and T24 n=3 

Statistical analysis title 
BMI  3 Month Treat Diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
17


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.067  
Method 
Mixed models analysis  
Parameter type 
3 Month Treatment Difference  
Point estimate 
0.48


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1  
upper limit 
0.03  
Statistical analysis title 
BMI  15 Month Treat Diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
17


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.029  
Method 
Mixed models analysis  
Parameter type 
15 Month Treatment Difference  
Point estimate 
0.73


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.39  
upper limit 
0.08  
Statistical analysis title 
BMI  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a continuous variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
17


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.02  
Method 
Mixed models analysis  
Parameter type 
Treatment Difference  
Point estimate 
0.56


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.03  
upper limit 
0.09  
Statistical analysis title 
BMI  24 Month Treat Diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a continuous variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
17


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.093  
Method 
Mixed models analysis  
Parameter type 
24 Month Treatment Difference  
Point estimate 
0.92


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.01  
upper limit 
0.16 


End point title 
Number of pulmonary exacerbations  median number of exacerbations during the 15 months following treatment commencement  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 15 months.




Statistical analysis title 
Mannâ€“Whitney U test  
Statistical analysis description 
A MannWhitney Utest will test whether the distribution of number of exacerbations is the same in each treatment arm


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
283


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.09  
Method 
Wilcoxon (MannWhitney)  
Confidence interval 


End point title 
Number of pulmonary exacerbations  Number of participants experiencing at least one exacerbation during the first 15 months of follow up  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 15 months.




Statistical analysis title 
Chisquared test  
Statistical analysis description 
The number and percentage of patients experiencing at least one exacerbation in each treatment arm was reported. Treatment groups were compared using the chisquared test.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
283


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.155  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.78


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.55  
upper limit 
1.1 


End point title 
Admission to hospital  Number of participants experiencing at least one hospital stay during the first 3 months of treatment  
End point description 

End point type 
Secondary


End point timeframe 
First 3 months of treatment.




Statistical analysis title 
Admission to hospital  3 months of treatment  
Statistical analysis description 
Number of participants experiencing at least one hospital stay during the first 3 months of treatment


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
278


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.057  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
1.77


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.97  
upper limit 
3.2 


End point title 
Admission to hospital  Number of participants experiencing at least one hospital stay during the 12 months following treatment  
End point description 

End point type 
Secondary


End point timeframe 
12 months following treatment.




Statistical analysis title 
Admission to hospital  12 months  
Statistical analysis description 
Number of participants experiencing at least one hospital stay during the 12 months following treatment.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
265


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.02  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.69


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.5  
upper limit 
0.95 


End point title 
Admission to hospital  Number of participants experiencing at least one hospital stay between 15 and 24 months  
End point description 
Number of participants experiencing at least one hospital stay between 15 months and 24 months.


End point type 
Secondary


End point timeframe 
Between 15 months and 24 months.




Statistical analysis title 
Admission to hospital  1524 months  
Statistical analysis description 
Number of participants experiencing at least one hospital stay between 15 months and 24 months


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
212


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.293  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.82


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.57  
upper limit 
1.19 


End point title 
Number of days spent as an inpatient in hospital  During treatment phase  
End point description 

End point type 
Secondary


End point timeframe 
From baseline to 3 months.




Statistical analysis title 
Length of stay in days during the treatment phase  
Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
278


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.066  
Method 
Wilcoxon (MannWhitney)  
Confidence interval 


End point title 
Number of days spent as an inpatient in hospital  During the 12 months post treatment phase  
End point description 

End point type 
Secondary


End point timeframe 
12 months post treatment.




Statistical analysis title 
Length of stay during 12 months post treatment  
Statistical analysis description 
Length of stay in days during the 12 months post treatment phase.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
265


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.005  
Method 
Wilcoxon (MannWhitney)  
Confidence interval 


End point title 
Number of days spent as an inpatient in hospital  between 15 and 24 months follow up  
End point description 

End point type 
Secondary


End point timeframe 
Between 15 and 24 months.




Statistical analysis title 
Length of stay 1524 months  
Statistical analysis description 
Length of stay in days, between 15 and 24 months follow up.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
201


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.261  
Method 
Wilcoxon (MannWhitney)  
Confidence interval 


End point title 
CFQ  Physical functioning – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Basline to 24 months




Notes [17]  The number of patients at base line was 53, 49 at 3 months, 50 at 15 months and 44 at 24 months. [18]  The number of patients at baseline was 61, 56 at 3 months, 50 at 15 months and 45 at 24 months 

Statistical analysis title 
Physical functioning (Selfreport)  3 Month Diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups at T3 (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.056  
Method 
Mixed models analysis  
Parameter type 
Mean difference at 3 months  
Point estimate 
6.63


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.18  
upper limit 
13.44  
Statistical analysis title 
Physical functioning (Selfreport)  15 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups at T15 (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.292  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
3.63


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
10.41  
upper limit 
3.16  
Statistical analysis title 
Physical functioning (Selfreport)  24 Month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups at T24 (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.793  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
0.94


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.15  
upper limit 
8.03  
Statistical analysis title 
Physical functioning (Selfreport)  Treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable . The following was reported: mean (SD) for each treatment group; mean (95% CI) and difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.538  
Method 
Mixed models analysis  
Parameter type 
Treatment Difference  
Point estimate 
1.62


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.57  
upper limit 
6.8 


End point title 
CFQ  Role/school functioning – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
baseline to 24 months




Notes [19]  At baseline n=14, at T3 n=13, at T15 n=11 and at T24 n=12 [20]  At baseline n=18, T3 n= 14, T15 n=15 and T24 n=12 

Statistical analysis title 
Role/school functioning (Selfreport)3 Month diff  
Statistical analysis description 
A mixedeffects model for repeated measures will be fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.181  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
8.92


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.42  
upper limit 
22.26  
Statistical analysis title 
Role/school functioning (Selfreport)15 treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures will be fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.267  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
7.66


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.21  
upper limit 
21.52  
Statistical analysis title 
Role/school functioning (Selfreport)24 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.328  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
6.97


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.39  
upper limit 
21.33  
Statistical analysis title 
Role/school functioning (Selfreport) treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.204  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
8.09


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.7  
upper limit 
20.89 


End point title 
CFQ  Vitality – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [21]  At baseline n=14, T3 n=13, T15 n=11 and T24 n=12 [22]  At baseline n=18, T3 n=14, T15 n=15 and T24 n=12 

Statistical analysis title 
Vitality (Self Report)  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.182  
Method 
Mixed models analysis  
Parameter type 
3 Month Treatment Difference  
Point estimate 
9.07


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.56  
upper limit 
22.69  
Statistical analysis title 
Vitality (Self Report)  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.373  
Method 
Mixed models analysis  
Parameter type 
15 Month Treatment Difference  
Point estimate 
5.4


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.88  
upper limit 
17.69  
Statistical analysis title 
Vitality (Self Report)  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.132  
Method 
Mixed models analysis  
Parameter type 
24 Month Treatment Difference  
Point estimate 
10.07


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.27  
upper limit 
23.41  
Statistical analysis title 
Vitality (Self Report)  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.191  
Method 
Mixed models analysis  
Parameter type 
Treatment Difference  
Point estimate 
6.89


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.67  
upper limit 
17.45 


End point title 
CFQ  Emotional functioning – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [23]  At baseline n=53, T3 n=49, T15 n=50 and T24 n=44 [24]  At baseline n=56, T3 n=49, T15 n=50 and T24 n=45 

Statistical analysis title 
Emotional functioning (Selfreport)  3 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.977  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
0.07


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
5.06  
upper limit 
4.91  
Statistical analysis title 
Emotional functioning (Selfreport) 15 month diff  
Statistical analysis description 
A mixedeffects model for repeated measure was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.589  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
1.59


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.39  
upper limit 
4.22  
Statistical analysis title 
Emotional functioning (Selfreport) 24 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.964  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
0.14


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.19  
upper limit 
5.91  
Statistical analysis title 
Emotional functioning (Selfreport)  Treat Diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.807  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0.53


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.78  
upper limit 
3.73 


End point title 
CFQ  Social functioning – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
baseline to 24 months




Notes [25]  At baseline n=53, T3 n=49, T15 n=49 and T24 n=44 [26]  At baseline n=61, T3 n=56 , T15 n=50 and T24 n=45 

Statistical analysis title 
Social functioning (Selfreport)  3 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.311  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
2.98


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.82  
upper limit 
8.78  
Statistical analysis title 
Social functioning (Selfreport)  15 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.498  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
2.11


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.03  
upper limit 
8.25  
Statistical analysis title 
Social functioning (Selfreport)  24 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.159  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
4.31


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.72  
upper limit 
10.34  
Statistical analysis title 
Social functioning (Selfreport)  Treatment Diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. . The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.138  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
3.16


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.03  
upper limit 
7.36 


End point title 
CFQ  Body Image – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [27]  At baseline n=52, T3 n= 49, T15 n=50 and T24 n=43 [28]  At baseline n=61, T3 n=56, T15 n=50 and T24 n=45 

Statistical analysis title 
Body image (Selfreport)  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.443  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
2.23


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.51  
upper limit 
7.97  
Statistical analysis title 
Body image (Selfreport)  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.309  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
4.01


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
11.78  
upper limit 
3.77  
Statistical analysis title 
Body image (Selfreport)  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.986  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
0.07


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.58  
upper limit 
7.45  
Statistical analysis title 
Body image (Selfreport)  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable.The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.931  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0.2


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.43  
upper limit 
4.48 


End point title 
CFQ  Eating Problems – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [29]  At baseline n=53, T3 n=49, T15 n=50 and T24 n=44 [30]  At baseline n=61, T3 n=56, T15 n=50 and T24 n=45 

Statistical analysis title 
Eating problems (Selfreport)  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.453  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
2.52


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.12  
upper limit 
9.16  
Statistical analysis title 
Eating problems (Selfreport)  15 month treat dif  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.903  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
0.39


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.78  
upper limit 
6  
Statistical analysis title 
Eating problems (Selfreport) 24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.31  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
3.67


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.47  
upper limit 
10.81  
Statistical analysis title 
Eating problems (Selfreport) treatment diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.506  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
1.66


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.27  
upper limit 
6.58 


End point title 
CFQ  Treatment Burden – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [31]  At baseline n=52, T3 n=49, T15 n=50 and T24 n=43 [32]  At baseline n=61, T3 n=56, T15 n=50 and T24 n=45 

Statistical analysis title 
Treatment burden (Selfreport) 3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.244  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
4.92


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.42  
upper limit 
13.27  
Statistical analysis title 
Treatment burden (Selfreport)15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.532  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
2.86


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.19  
upper limit 
11.92  
Statistical analysis title 
Treatment burden (Selfreport)24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.343  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
4.68


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
5.07  
upper limit 
14.43  
Statistical analysis title 
Treatment burden (Selfreport) Treatment diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
113


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.167  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
4.23


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.8  
upper limit 
10.27 


End point title 
CFQ  Health Perceptions – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [33]  At baseline n=14, T3 n=13, T15 n=11 and T24 n=12 [34]  At baseline n=18, T3 n=14, T15 n= 15 and T24 n=12 

Statistical analysis title 
Health perceptions (Selfreport)3 month treat dif  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.045  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
18.41


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.4  
upper limit 
36.42  
Statistical analysis title 
Health perceptions (Selfreport)15 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.583  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
5.06


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
13.72  
upper limit 
23.84  
Statistical analysis title 
Health perceptions (Selfreport)24 month diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.544  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
6.33


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
14.89  
upper limit 
27.54  
Statistical analysis title 
Health perceptions (Selfreport)treatment diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
32


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.13  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
12.46


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.96  
upper limit 
28.88 


End point title 
CFQ  Weight – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [35]  At baseline n=14, T3 n= 12, T15 n=11 and T24 n=12 [36]  At baseline n=17, T3 n=14 , T15 n=15, T24 n=12 

Statistical analysis title 
Weight (self report)  3 Month treatment diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Oral antibiotic therapy v Intravenous (IV) antibiotics


Number of subjects included in analysis 
31


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.725  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
2.51


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
12.06  
upper limit 
17.08  
Statistical analysis title 
Weight (self report)  15 Month treatment diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
31


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.888  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
1.4


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
19.02  
upper limit 
21.83  
Statistical analysis title 
Weight (self report)  24 Month treatment diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
31


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.842  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
2.68


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
24.8  
upper limit 
30.17  
Statistical analysis title 
Weight (self report)  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
31


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.603  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
2.5


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.29  
upper limit 
12.29 


End point title 
CFQ  Respiratory Symptoms – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 Months




Notes [37]  At baseline n=53, T3 n=48, T15 n=49 and T24 n=44 [38]  At baseline n=61, at T3 n=48 , at T15 n=50 and T24 n=44 

Statistical analysis title 
Resp Symp Self Report  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.606  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
1.79


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
5.06  
upper limit 
8.64  
Statistical analysis title 
Resp Symp Self Report 15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.374  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
2.82


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.44  
upper limit 
9.08  
Statistical analysis title 
Resp Symp Self Report 24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.53  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
2.08


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.47  
upper limit 
8.64  
Statistical analysis title 
Resp Symp Self Report treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures wsa fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.344  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
2.27


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.47  
upper limit 
7.01 


End point title 
CFQ  Digestive Symptoms – Self Report  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 Months




Notes [39]  At baseline n=53, T3 n=47, T15 n=48 and T24 n=42 [40]  At baseline n=61, T3 n=56, T15 n=50 and T24 n=45 

Statistical analysis title 
Digestive (Self report)  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.551  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
2.72


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.3  
upper limit 
11.75  
Statistical analysis title 
Digestive (Self report) 15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.998  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
0.01


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.95  
upper limit 
9.93  
Statistical analysis title 
Digestive (Self report)  treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.987  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.23  
upper limit 
6.33  
Statistical analysis title 
Digestive (Self report)  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
114


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.581  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
2.49


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
11.44  
upper limit 
6.45 


End point title 
CFQ  Physical functioning – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 Months




Notes [41]  At baseline n=37, T3 n=38 , T15 n= 42, T24 n=32 [42]  At baseline n=42, T3 n=39, T15 n=32 and T24 n=33 

Statistical analysis title 
Phys Fn  Parent/Carer  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.033  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
6.18


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.51  
upper limit 
11.85  
Statistical analysis title 
Phys Fn  Parent/Carer  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.223  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
5.17


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
13.55  
upper limit 
3.22  
Statistical analysis title 
Phys Fn  Parent/Carer  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.342  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
4.63


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
14.29  
upper limit 
5.03  
Statistical analysis title 
Phys Fn  Parent/Carer  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.191  
Method 
Mixed models analysis  
Parameter type 
treat difference  
Point estimate 
3.62


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.85  
upper limit 
9.09 


End point title 
CFQ  Role/school functioning – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [43]  At baseline n= 36, T3 =37, T15 n=42 and T24 n=32 [44]  At baseline n=42, T3 n=39, T15 n=32 and T24 n=33 

Statistical analysis title 
Role/School  P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.238  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
5.47


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.69  
upper limit 
14.64  
Statistical analysis title 
Role/School  P/C  treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.845  
Method 
Mixed models analysis  
Parameter type 
treat diff  
Point estimate 
0.67


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.49  
upper limit 
6.15  
Statistical analysis title 
Role/School  P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.738  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
1.47


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
10.22  
upper limit 
7.28  
Statistical analysis title 
Role/School  P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.318  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
5.06


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
15.09  
upper limit 
4.97 


End point title 
CFQ  Vitality– Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [45]  At baseline n=37, T3 n=38, T15 n=42 and T24 n=31 [46]  At baseline n=41, T3 n=39 , T15 n=32 and T24 n=33 

Statistical analysis title 
Vitality  P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.059  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
6.53


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.26  
upper limit 
13.32  
Statistical analysis title 
Vitality  P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.923  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
0.44


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.36  
upper limit 
8.48  
Statistical analysis title 
Vitality  P/C  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable.The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.208  
Method 
Mixed models analysis  
Parameter type 
treat diff  
Point estimate 
3.7


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.11  
upper limit 
9.5  
Statistical analysis title 
Vitality  P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.913  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
0.57


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.84  
upper limit 
10.98 


End point title 
CFQ  Emotional functioning – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [47]  At baseline n=36, T3 n=37, T15 n= 42, t24 n=32 [48]  At baseline n=42, T3 n=39, T15 n=32 adn T24 n=33 

Statistical analysis title 
Emotional  P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.069  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
5.97


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.47  
upper limit 
12.4  
Statistical analysis title 
Emotional  P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.339  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
3.32


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.56  
upper limit 
10.2  
Statistical analysis title 
Emotional  P/C  treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.166  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
3.47


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.48  
upper limit 
8.41  
Statistical analysis title 
Emotional  P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.668  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
1.69


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.55  
upper limit 
6.16 


End point title 
CFQ  Body Image – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 Months




Notes [49]  At baseline n=36, T3 n=36 , T15 n=42 and T24 n=32 [50]  At baseline n=42, T3 n=39, T15 n=32, and T24 n=33 

Statistical analysis title 
Body Image  P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Oral antibiotic therapy v Intravenous (IV) antibiotics


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.555  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
2.66


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.28  
upper limit 
11.59  
Statistical analysis title 
Body Image  P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.922  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
0.56


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
12.03  
upper limit 
10.9  
Statistical analysis title 
Body Image  P/C  treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.644  
Method 
Mixed models analysis  
Parameter type 
treat diff  
Point estimate 
1.94


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.4  
upper limit 
10.28  
Statistical analysis title 
Body Image  P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.679  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
2.58


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.81  
upper limit 
14.97 


End point title 
CFQ  Eating Problems– Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [51]  Baseline n=37, T3 n=34 , T15 n=39 and T24 n=32 [52]  At baseline n=41, T3 n=39, T15 n=32 and T24 n=32 

Statistical analysis title 
Eating Problem P/C  3 Month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.3  
Method 
Mixed models analysis  
Parameter type 
3 month treatment difference  
Point estimate 
4.6


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.19  
upper limit 
13.39  
Statistical analysis title 
Eating Problem P/C  15 Month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.174  
Method 
Mixed models analysis  
Parameter type 
15 month treatment difference  
Point estimate 
6.15


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
2.78  
upper limit 
15.08  
Statistical analysis title 
Eating Problem P/C  24 Month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.732  
Method 
Mixed models analysis  
Parameter type 
24 month treatment difference  
Point estimate 
1.94


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
13.18  
upper limit 
9.31  
Statistical analysis title 
Eating Problem P/C  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable . The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.138  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
4.05


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
1.33  
upper limit 
9.44 


End point title 
CFQ  Treatment Burden – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [53]  At baseline n=37, T3 n=37, T15 n=42, T24 n=32 [54]  At baseline n=42, T3 n=39, T15 n=32, T24 n=33 

Statistical analysis title 
Treatment Burden P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.718  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
1.99


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
8.94  
upper limit 
12.92  
Statistical analysis title 
Treatment Burden P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.96  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
0.28


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
11.25  
upper limit 
10.69  
Statistical analysis title 
Treatment Burden P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.966  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
0.23


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
10.12  
upper limit 
10.57  
Statistical analysis title 
Treatment Burden P/C  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.862  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0.68


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.08  
upper limit 
8.44 


End point title 
CFQ  Health Perceptions – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [55]  At baseline n=36, T3 n=36, T15 n=42, T24 n=32 [56]  At bseline n=42, T3 n=39, T15 n=32, T24 n=33 

Statistical analysis title 
Health Perceptions P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.988  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
0.07


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.37  
upper limit 
9.23  
Statistical analysis title 
Health Perceptions P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.07  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
7.25


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
15.09  
upper limit 
0.6  
Statistical analysis title 
Health Perceptions P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.297  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
6.76


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
19.6  
upper limit 
6.07  
Statistical analysis title 
Health Perceptions P/C  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.115  
Method 
Mixed models analysis  
Parameter type 
treat diff  
Point estimate 
4.39


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
9.88  
upper limit 
1.1 


End point title 
CFQ  Weight – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [57]  At baseline n=37, T3 n=37, T15 n=41 and T24 n=32 [58]  At baseline n=42, T3 n=37, T15 n=32 and T24 n=33 

Statistical analysis title 
Weight  P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.294  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
7.66


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
6.78  
upper limit 
22.1  
Statistical analysis title 
Weight  P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.886  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
1.19


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
17.63  
upper limit 
15.26  
Statistical analysis title 
Weight  P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.492  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
7.03


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
13.27  
upper limit 
27.33  
Statistical analysis title 
Weight  P/C  treatment difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.247  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
6.37


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.5  
upper limit 
17.25 


End point title 
CFQ  Respiratory Symptoms– Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [59]  At baseline n=37, T3 n=37, T15 n=41 and T24 n=32 [60]  At baseline n=42, T3 n=39, T15 n=32 and T24 n=33 

Statistical analysis title 
Resp Symp P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.39  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
4.11


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
5.36  
upper limit 
13.58  
Statistical analysis title 
Resp Symp P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.432  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
3.33


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
11.71  
upper limit 
5.06  
Statistical analysis title 
Resp Symp P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.728  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
2.11


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
14.15  
upper limit 
9.93  
Statistical analysis title 
Resp Symp P/C  treament difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Oral antibiotic therapy v Intravenous (IV) antibiotics


Number of subjects included in analysis 
79


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.885  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
0.47


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
7.01  
upper limit 
6.06 


End point title 
CFQ  Digestive symptoms – Parent/Carer  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 24 months




Notes [61]  At baseline n=37, T3 n=34 , T15 n=39 and T24 n=32 [62]  At baseline n=41, T3 n=39, T15 n=32 and T24 n=32 

Statistical analysis title 
Digestive  P/C  3 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.474  
Method 
Mixed models analysis  
Parameter type 
3 month treat diff  
Point estimate 
2.66


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.71  
upper limit 
10.03  
Statistical analysis title 
Digestive  P/C  15 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.405  
Method 
Mixed models analysis  
Parameter type 
15 month treat diff  
Point estimate 
3.54


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
4.9  
upper limit 
11.98  
Statistical analysis title 
Digestive  P/C  24 month treat diff  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group, timepoint as a categorical variable and an interaction term (treatment group*time). The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.725  
Method 
Mixed models analysis  
Parameter type 
24 month treat diff  
Point estimate 
1.75


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
8.12  
upper limit 
11.62  
Statistical analysis title 
Digestive  P/C  Treatment Difference  
Statistical analysis description 
A mixedeffects model for repeated measures was fitted, with an unstructured covariance matrix. The baseline measurement of the domain was fitted as a covariate along with treatment group and timepoint as a categorical variable. The following was reported: mean (SD) for each treatment group; mean (95% CI) difference between treatment groups (derived from the model); and a pvalue of the treatment effect


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
78


Analysis specification 
Posthoc


Analysis type 
superiority  
Pvalue 
= 0.389  
Method 
Mixed models analysis  
Parameter type 
treatment difference  
Point estimate 
2.61


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
3.4  
upper limit 
8.61 


End point title 
Number of patients with at least one positive result of MRSA by 3 months  
End point description 
No analysis was possible due to there being 0 events in the IV group.


End point type 
Secondary


End point timeframe 
Baseline to three months




No statistical analyses for this end point 


End point title 
Number of patients with at least one positive result of Burkholderia cepacia complex (BC) by 3 months  
End point description 
No analysis was possible due to there being 0 events in the Oral group.


End point type 
Secondary


End point timeframe 
Baseline to three months




No statistical analyses for this end point 


End point title 
Number of patients with at least one positive result of Candida by 3 months  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 3 months




Statistical analysis title 
Secondary Outcome  Relative Risk  
Statistical analysis description 
For each microorganism, the number and percentage with at least one positive result were presented split by treatment arm, and a relative risk and 95% confidence interval calculated. Chisquared test (or if necessary Fisher’s exact test) was used to test for a difference between treatment groups.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
283


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.917  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
1.03


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.63  
upper limit 
1.67 


End point title 
Number of patients with at least one positive result of Aspergillus by 3 months  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 3 months




Statistical analysis title 
Secondary Outcome  Relative Risk  
Statistical analysis description 
For each microorganism, the number and percentage with at least one positive result waspresented split by treatment arm, and a relative risk and 95% confidence interval calculated. Chisquared tests (or if necessary Fisher’s exact test) was used to test for differences between treatment groups.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
280


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.686  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
1.27


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.4  
upper limit 
4.07 


End point title 
Number of patients with at least one positive result of MRSA by 15 months  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 15 months




Statistical analysis title 
Secondary Outcome  Relative Risk  
Statistical analysis description 
For each microorganism, the number and percentage with at least one positive result was presented split by treatment arm, and a relative risk and 95% confidence interval calculated. Chisquared tests (or if necessary Fisher’s exact test) was used to test for differences between treatment groups.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
275


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.441  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
2.07


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.39  
upper limit 
11.14 


End point title 
Number of patients with at least one positive result of Burkholderia cepacia complex (BC) by 15 months  
End point description 

End point type 
Secondary


End point timeframe 
Baseline to 15 months




Statistical analysis title 
Secondary Outcome  Relative Risk  
Statistical analysis description 
For each microorganism, the number and percentage with at least one positive result was presented split by treatment arm, and a relative risk and 95% confidence interval calculated. Chisquared tests (or if necessary Fisher’s exact test) was used to test for differences between treatment groups.


Comparison groups 
Intravenous (IV) antibiotics v Oral antibiotic therapy


Number of subjects included in analysis 
274


Analysis specification 
Prespecified


Analysis type 
superiority  
Pvalue 
= 0.684  
Method 
Chisquared  
Parameter type 
Risk ratio (RR)  
Point estimate 
0.51


Confidence interval 

level 
95%  
sides 
2sided


lower limit 
0.1  
upper limit 
2.76 


End point title 
Number of patients with at least one positive result of Candida by 15 months  
End point description 