30-Day readmission rates after hospitalisation have risen and the UK Department of Health plans to introduce financial penalties for early readmissions. The surgical argument is reasonably clear but for chronic conditions such as COPD the presumption that early readmission is either avoidable or due to poor hospital care is far from clear. We used Hospital Episode Statistics data to examine the characteristics of patients readmitted within 30 days of an index COPD hospitalisation during 2006–2007. Mortality data were obtained from ONS death registration. There were 124 834 COPD admissions in 2006–2007 with 78 693 different patients discharged from hospital and we followed the first admission of each. 28 533 (36%) patients were readmitted within 1-year and 8439 (11%) within 30 days. Patients readmitted within 30 days were more likely to be male (53% vs 50%; p<0.001), older (72.7 vs 71.6 yrs; p<0.001), live in a more deprived area (55% vs 51% highest quintile; p<0.001) and much more likely to die (30-day mortality 11% vs 3%; p<0.001; and 1-year mortality 40% vs 20%; p<0.001). For the index admission, patients readmitted within 30 days were more likely to have been admitted via A&E rather than via their GP (74% vs 70%; p<0.001), more likely to have had a LRTI (88% vs 83%; p<0.001), had a longer median length of stay (6 vs 5 days; p<0.001) and were more likely to have been discharged from the care of a chest physician (44% vs 40%; p<0.001). Hospitals with the lowest inpatient mortality during the first admission had the highest 30-day readmission rate (r=0.22; p<0.01). Early readmission would appear to be driven by demographics, severity of the index admission and disease severity which cannot be influenced by care delivery or organisation of care. It's possible that better care—keeping more patients alive during their index admission—may lead to those hospitals being penalised for a higher 30-day readmission rate.