Introduction Adequate follow up is a key element of COPD care bundles (CB). COPD nurse specialists responsible for completing follow up consultations may be able to utilise clinical judgment and measures of health status to predict which patients are at greater risk of readmission.

Objective We explored whether COPD nurse specialists working in the REspiratory Discharge Service (REDS), who delivered the CB, could predict whether patients would readmit within15 days post discharge. We also explored levels of health and psychological status for those patients who the REDS team thought were and were not at risk of readmission.

Methods This was a retrospective audit of patients who received a COPD discharge CB from April 2013 to March 2014. Readmission likelihood was recorded by the REDS team after completion of a 2 day post-discharge phone consultation. Patients also completed the COPD Assessment Test (CAT), MRC breathlessness scale and the Hospital Anxiety and Depression Scale (HADS). Mean between-group differences for the ‘will admit’ and ‘will not admit’ groups were analysed using independent t-tests.

Results Readmission risk was recorded for 1003 patients who received the CB prior to discharge. A total of 100 patients of these 1003 readmitted (readmission rate of 9.7%). The REDS team correctly predicted that 39 of these 100 patients would be readmitted. There were statistically significant between-group differences for MRC, CAT and HADS scores (see table): Patients placed in the ‘will readmit’ group had significantly worse CAT, HADS anxiety and depression scores compared those placed in the will not readmit group.

Conclusions COPD nurse specialists were unable to correctly predict the majority of readmissions. Patients deemed at risk of readmision had worse levels of psychological and health status than those who were not thought to be at risk of readmission. This suggests other factors must be important in predicting COPD readmissions.