BACKGROUND
Though medical decision making is ostensibly only a task of health professionals, many – if not most – and also critical health related decisions are made by laypersons. For example, the first step to receiving the right health care requires the patient to determine whether and where too seek health care (triage). Overcautious self-assessments (ie, overtriaging) may lead to overutilization of health care facilities (eg, overcrowded emergency departments) while imprudent decisions (ie, undertriaging) constitute a risk to the patient’s health. Patient-facing decision support system providing guidance in patients’ self-appraisal of their complaints, commonly known as symptom checkers, might mitigate both dangers. Their use might, however, also constitute a hazard on its own.
OBJECTIVE
To provide evidence on where symptom checkers might prove beneficial, we assessed the capability to triage of laypersons, and explored where triage errors commonly occur.
METHODS
We analyzed publicly available data on 91 laypersons appraising 45 short fictitious patient descriptions (case vignettes;n = 4095 appraisals). Applying signal detection theory and descriptive statistics, we explore whether the type of decision, laypersons’ confidence in their decision, and sociodemographic factors influence their triage accuracy and the type of error they make.
RESULTS
Sensitivity for detecting emergencies was lower (M = 67.5%, SD = 16.4%) than the specificity (M = 89.6%, SD = 8.6%). Sensitivity for deciding whether medical care is required was higher (M = 90.5%, SD = 8.3%) than the specificity (M = 46.7%, SD = 15.95%). Sociodemographic variables were not associated with risk-averseness despite female participants overtriaging more often than male ones. Participants’ triage accuracy was higher when they were certain about their appraisal (62.5%, 2114/3381) than when being uncertain (52.9%, 378/714). However, most errors occurred when participants were fond of their decision (79%, 1267/1603).
CONCLUSIONS
Laypersons’ confidence in their own triage assessment might prove an inadequate criterion to judge whether decision support would benefit them. While they seem overcautious in deciding whether they require medical care, they miss a considerable portion of emergencies. Symptom checkers able to correct laypersons’ stand-alone triage errors may therefore aid the patient by avoiding unsafe urgency assessments. Concerning potential benefits to the health care system, symptom checkers might facilitate disburdenment of low-acuity care more than that of emergency care facilities.