Understanding missed opportunities for more timely diagnosis of cancer in symptomatic patients after presentation

Abstract

The diagnosis of cancer is a complex, multi-step process. In this paper, we highlight factors involved in missed opportunities to diagnose cancer more promptly in symptomatic patients and discuss responsible mechanisms and potential strategies to shorten intervals from presentation to diagnosis. Missed opportunities are instances in which post-hoc judgement indicates that alternative decisions or actions could have led to more timely diagnosis. They can occur in any of the three phases of the diagnostic process (initial diagnostic assessment; diagnostic test performance and interpretation; and diagnostic follow-up and coordination) and can involve patient, doctor/care team, and health-care system factors, often in combination. In this perspective article, we consider epidemiological 'signals' suggestive of missed opportunities and draw on evidence from retrospective case reviews of cancer patient cohorts to summarise factors that contribute to missed opportunities. Multi-disciplinary research targeting such factors is important to shorten diagnostic intervals post presentation. Insights from the fields of organisational and cognitive psychology, human factors science and informatics can be extremely valuable in this emerging research agenda. We provide a conceptual foundation for the development of future interventions to minimise the occurrence of missed opportunities in cancer diagnosis, enriching current approaches that chiefly focus on clinical decision support or on widening access to investigations.

Figure 1

A model for defining missed diagnostic opportunities. Adopted from Singh, 2014.

Figure 2

Epidemiological evidence suggestive of likely missed opportunities. Incidence rate ratios (IRR) for general practitioner consultations before the diagnosis of cancer compared with age- and sex-matched ‘control' patients (without a diagnosis of cancer). Data from Christensen et al (2012); n (women)=63 362 cancer patients and 633 620 controls; n (men)=63 848 cancer patients and 638 480 controls. Note very narrow 95% confidence intervals that exclude parity (i.e., 1.00); and excess risk spanning a 12-month period, including −6 to −4 months.