There are limitations to the current study. The control group was of above average cognitive ability, thereby potentially overestimating cognitive differences. However, significant differences were still evident between the COPD groups after controlling for covariates. While our study had a prospective element, we do not know the state of the patients prior to the exacerbation that brought them into the study.

The patients in the COPD-E group had poorer scores than the stable patients across a range of measures of COPD severity, but we could not separate out acute effects of the disease Cheap Viagra online, including severity of exacerbation and other inpatient events from those that are more chronic. That would require preadmission measurement when the patients were in a stable state. Such a study would need to be large and resource intensive, since with annual hospitalization rates of 0.2 admissions per year and a 1-year mortality typically > 20%, it would be necessary to follow > 140 patients for 1 year to capture 30 episodes.

However, the absence of any evidence of recovery over 3 months suggests that the observed impairment is a relatively stable characteristic of these patients rather than the effect of the single acute event that we documented. We chose 3 months for our follow up point, since health status should have improved significantly. The interpretation of data from a longer follow up period would be problematic because of attrition of the most ill patients due to recurrent exacerbations in this high risk group and to survival bias. Finally, it is probable that there was an acquisition bias in the recruitment of exacerbating patients that may have excluded more severe patients. Those who were subjectively more frail were less likely to agree to participate, and in addition, 10% of the screened population was considered by the investigator to be too frail to participate; a further 10% had clinically apparent dementia and so were not recruited.

The clinical implications of these findings are quite significant. Over recent years there has been a drive for admission avoidance and early or supported discharge schemes. At an individual patient level, success will depend on the patient’s ability to function at home, and it is likely that cognitive ability will be a contributing factor to success.