In this multicentre randomised controlled trial from the USA, the efficacy of a comprehensive care management programme (CCMP) in preventing hospitalisations for chronic obstructive pulmonary disease (COPD) was compared with usual care. This approach has shown favourable results in other chronic conditions but evidence is lacking in COPD. An unexpected increase in mortality resulted in the trial being terminated before enrolment was complete.

Patients with known COPD at risk of hospitalisations were randomised to receive either usual care or a CCMP. This constituted a series of educational sessions, personalised action-plans and prescriptions to target exacerbations and regular telephone contact with case managers. The primary end-point was time-to-first hospitalisation due to COPD.

As the trial was performed within the Veteran's Affairs health service, a data and safety monitoring committee was required. As a result, at the 2-year mark, with 467 patients enrolled out of a target of 960, a significant increase in all-cause mortality was noted in the intervention group, with a non-significant trend towards increased deaths due to COPD.

The authors surmised that unrecorded differences between the two groups may be responsible. The absence of comparative scores for body mass index and exercise tolerance between the two groups suggests that BODE scores (a composite index including: Body Mass Index, Airflow Obstruction, Dyspnoea and Exercise tolerance, which has been seen to predict mortality in COPD) were not generated and compared, potentially masking important differences. As well as increased mortality, the available information suggests that the intervention group saw no significant improvement in the primary outcome of increased time-to-next hospitalisation due to COPD.

Further studies of this type in the COPD population will require careful data monitoring and a rigorous characterisation of enrolled patients to try and avoid the outcome seen here.