'When all you have is a hammer, everything has to be a nail': Managing diagnostic uncertainty in urinary tract infection.

Diagnosis of urinary tract infections (UTI) is a routine part of medical work and yet is well recognised to be an area of high clinical uncertainty. Meanwhile, diagnosis of UTI is becoming increasingly important to policymakers globally due to concerns about antibiotic over-prescription. Drawing on Mol's concept of ontological multiplicity in clinical work, I explore how diagnostic uncertainty is co-ordinated into certainty by a UK national diagnostic algorithm for UTI. The diagnosis of UTI is produced or withheld as a post hoc rationalisation of a prior decision whether to prescribe antibiotics or not. Work in the sociology of diagnosis has already noted that diagnostic steps are often re-ordered by health-care professionals taking diverse actions in the best interest of their patients. This article contributes an argument that ordering diagnostic work around antimicrobial stewardship agendas has the effect of narrowing possible actions. Exploring the consequences and effects of doing diagnosis in this way for different groups, I argue that a greater creativity about what could be done to care for painful bladders could be found in a return to more clinical ways of working.

Nitrofurantoin: what is the evidence for current UK guidance?

Three days of nitrofurantoin at 100 mg twice daily is recommended as first-line treatment for uncomplicated urinary tract infection (UTI) in evidence-based guidelines across the UK. A review of international guidelines shows that the evidence base has been interpreted in very different ways. UK guidelines are unusual in promoting short (3 day) courses, and we find little direct evidence to support this. Although 'short' courses of antibiotics for other agents may provide optimum balance between providing effective treatment whilst reducing selective pressure driving resistance amongst colonizing microbial flora, it remains unclear that course lengths can be extrapolated to nitrofurantoin. Three days of nitrofurantoin may indeed be a useful intervention in a large group of patients. However, without supporting evidence and because clinical response should be expected to vary widely, it is unclear that establishing and promoting an antibiotic duration for UTI is the best approach to optimizing prescribing in this important area.