Ambulatory Oxygen in Fibrotic Lung Disease (FLD) (AmbOx)

The planned non-commercial study is a randomised, controlled crossover trial of ambulatory oxygen against no ambulatory oxygen over a four week period (two weeks on ambulatory oxygen and two weeks on air, with no portable devices), to evaluate the effects of ambulatory oxygen on health status in patients with ILD. A short crossover study in this context has many advantages, since ambulatory oxygen has immediate effect, with no wash-out period needed after use. The primary outcome of the study will be the change in health status, as measured by the K-BILD questionnaire (Thorax 2012 67: 804-810).

At the start of the trial, the effects of ambulatory oxygen on 6 Minute Walk Test (6MWT) parameters will be evaluated on oxygen and on air-filled canisters, with the patient blind to the contents of the canister, to assess whether oxygen-induced improvements in 6MWT parameters can predict its effectiveness in day to day life. The 6MWT is a well established and highly reproducible test validated in ILD patients, with significant prognostic implications.

The study design does not include a placebo arm because:

1. The intervention is a combination of possible benefits from oxygen and the disadvantage of canister weight. These cannot be separated. Placebo control is impossible because there is no means of providing placebo weight. Attempts to control solely for oxygen use without taking canister weight into account are clinically meaningless. In a recent COPD, chronic obstructive pulmonary disease study, cylinder weight was reported as a barrier to use by 93% of study participants.
2. A positive result against an air-filled canister arm would be clinically uninterpretable. This is a study design in which the 'placebo' would be actively harmful to study participants. Carriage of an air-filled cylinder would be expected to lead to earlier desaturation and reduced exercise tolerance. Such a design would not inform the real life comparison between oxygen plus cylinder and no intervention.
3. All this aside, blinding is legally impossible (UK health and safety regulations require that oxygen cylinders for home use must be clearly labelled).

It should also be stressed that objective measures of change are evaluated as secondary end-points: the investigators expect to explore correlations between these variables and the primary end-point to exclude the possibility that an observed treatment benefit on the primary end-point might be confounded by a placebo effect.

A more in depth qualitative assessment of the impact of ambulatory oxygen will be undertaken via a semi-structured interview, in a subset of 20 patients, to investigate patients' and their carers' personal perspective on how the ambulatory oxygen has affected their day to day life. The interview will be conducted within 2 weeks of the end of treatment visit. Patients from the Royal Brompton Hospital site will be approached by the researcher at the end of treatment visit and asked if they will participate in the qualitative assessment. If the patient agrees a date will be agreed for the patient to be interviewed at a venue convenient to them (usually the patient's home).

During the interviews a semi-structured topic guide will address practical barriers to optimal oxygen usage, practical, social and psychological difficulties encountered, concerned about dependency, and views on the information required prior to ambulatory oxygen prescription. The interviews will also explore patients' (and carers') experience of participating in the trial. Notes will be written after each interview to aid reflective analytical processes. Individual feed back on how the system and the service could be improved will assess how the needs of patients and their future involvement can be incorporated into the design of more patients and their future involvement can be incorporated into the design of more patients centred devices.

Qualitative interview analysis: Interviews will be transcribed verbatim. Interview transcripts will be analysed thematically using a framework approach. Atlas/ti computer software (http://atlasti.com) will be used to manage and index the data prior to charting, mapping and interpretation.