An ethnographic study of diagnosis of physical illness in people with mental health conditions in the emergency department

People living with mental health conditions face reduced life expectancy, largely associated with under-diagnosed and under-treated physical illnesses. Inequalities in the way physical symptoms are diagnosed may be implicated in these outcomes, but, to date, studies have primarily understood these inequalities in terms of ‘diagnostic overshadowing’: the misattribution of physical symptoms to mental health conditions. In this paper, we use the candidacy framework to offer an extended analysis of the influences on diagnosis of physical health symptoms in people with mental health conditions presenting to the emergency department (ED)—a crucial node in the diagnostic pathway. We conducted a multi-site ethnography in three English EDs, including 284 h of non-participant observation and 43 interviews with clinicians, patients with mental health conditions, and accompanying persons. We found that, although the ED was seen as an open door, patients often struggled to have their physical symptoms recognised as legitimate concerns. Some delayed seeking care, feeling less deserving or anticipating dismissal; others tried to enhance their candidacy by downplaying their mental health history. The ED's operating conditions—throughput targets, overcrowded bays, and stretched staffing—favoured presentations that were clear-cut, straightforward, and urgent. This left little room for exploring more complex cases, such as those involving overlapping mental and physical health symptoms. Patients, clinicians and others recognised the risk that implicit bias might affect how physical symptoms in people with mental health conditions were interpreted, and they often sought to mitigate the impacts of such biases. This, however, could create its own risks, as different parties framed and reframed patients' presentations without full knowledge of the adjustments made by others—potentially impacting risks of both under-diagnosis and over-investigation in ways that were difficult to gauge. By applying the candidacy framework and examining the entire ED diagnostic pathway, our study illuminates a range of influences on diagnosis that extend beyond diagnostic overshadowing.