Pneumonia is one of the most frequent causes of hospitalisation, accounting for many deaths each year. Elderly patients, especially those in extended care facilities, are at particular risk for pneumonia and have a higher mortality rate than younger patients. The cost of treating patients with lower respiratory tract infections (LRTIs) is staggering, especially for patients who require hospitalisation. Less extensive diagnostic testing may be utilised in the future to minimise the cost of LRTIs, although this in turn might compromise our knowledge of the pathogens involved and their resistance patterns. Currently, the prevalence of various pathogens is known, and varies on the basis of underlying risk factors such as age, structural or functional lung disease, mental status, immune system function and geographical region. However, resistance patterns of commonly implicated pathogens are ever-changing. For example, Streptococcus pneumoniae, which is the most frequent cause of community-acquired pneumonia, has become resistant to benzylpenicillin (penicillin G) in recent years. This is especially disturbing because cross-resistance with other classes of antibiotics frequently occurs. Many antibiotics have been used in the treatment of LRTIs. Cephalosporins are popular because of their broad spectrum of activity and excellent safety profiles. Penicillins have also been popular, although resistant strains of S. pneumoniae now pose a serious threat. The macrolides have recently enjoyed increased popularity because of their activity against atypical pathogens. Although the fluoroquinolones are second-line agents for community-acquired pneumonia, they have a place in the treatment of LRTIs encountered in the nursing home or hospital setting, and even have activity against atypical bacteria. A variety of innovative programmes have been developed in recent years to control the cost of treating LRTIs. Although limited formulary choices have been used in the hospital setting for years, and are now becoming popular in managed care, there is no proof that this mechanism saves money when looking at the overall picture. A rational approach is to conduct a rigorous pharmacoeconomic evaluation of treatment options, thus identifying the therapies that provide the best value in each setting. Equally important are various programmes that encourage the cost-conscious use of the antibiotics chosen. Some of the methods evaluated in the literature include: notifying prescribers of the true cost of treatment alternatives, notifying prescribers whether or not third-party coverage is available for the prescription, streamlining from combination therapy to a single agent, early switching from parenteral to oral therapy, initiating treatment with oral agents, administering parenteral antibiotics at home from the outset of therapy, and antibiotic streamlining programmes that are partnered with infectious disease physicians. For the most part, these programmes have not been rigorously evaluated. Newer, more innovative ways to provide cost-conscious treatment of LRTIs will undoubtedly be developed. The basic premise for these programmes should be rigorous, well-designed pharmacoeconomic evaluations. Such studies will help ensure that all facets of therapy are evaluated and should prevent choices being made simply on the basis of the lowest acquisition cost.