The use of benzodiazepines is not associated with community-acquired pneumonia

We read with interest the recent study done by Obiora et al 1 about the incidence and mortality with pneumonia for benzodiazepine users verses non-benzodiazepine users. We were interested to see whether these results would be generalisable to the Taiwanese population, therefore, we used the National Health Insurance Database of 22 million (2002) population in order to investigate the association of benzodiazepines and community-acquired pneumonia (CAP). We selected window size of 1, 2 and 3 months to compute odds ratios of the diagnosis of pneumonia and presence of benzodiazepine prescription filling in all age and sex groups. We took patients having CAP identified through ICD-9-CM (480–486) codes (International Classification of Diseases, Ninth Revision, and Clinical Modification) and identified medications from their prescription using ATC (Anatomical Therapeutic Chemical) drug classification codes (N05BA01, N05BA02, N05BA06) system for benzodiazepines.

However, we found no significant association between benzodiazepines use with CAP for 30, 60 and 90 days while computing the odds ratios,2 as shown in table 1. We checked benzodiazepines separately and in combination, but still, there was no association found for benzodiazepines being associated with CAP. Although there are not much studies about benzodiazepines in relation to pneumonia; however, added to our results, Dublin et al 3 found in geriatrics that there is no risk for pneumonia while using benzodiazepines. Trifirò et al 4 studied specifically typical and atypical antipsychotic drugs use, and risk for pneumonia in the elderly population; they also did not find any significant relationship with regard to benzodiazepines. Our result supports Dublin and Trifiro, which are contradictory to the Obiora et al study, that there is no significant relationship found between using benzodiazepines and having incidence risk for pneumonia.

We believe that our findings would contribute to the discussion on this topic. It is important, however, to emphasise the need for randomised controlled trials for benzodiazepines use and incidence of CAP in patients to demonstrate any causal relationship in terms of morbidity and mortality of having pneumonia while using benzodiazepines.