Diagnostic accuracy of symptoms for an underlying disease: a simulation study

Abstract

Symptoms have been used to diagnose conditions such as frailty and mental illnesses. However, the diagnostic accuracy of the numbers of symptoms has not been well studied. This study aims to use equations and simulations to demonstrate how the factors that determine symptom incidence influence symptoms’ diagnostic accuracy for disease diagnosis. Assuming a disease causing symptoms and correlated with the other disease in 10,000 simulated subjects, 40 symptoms occurred based on 3 epidemiological measures: proportions diseased, baseline symptom incidence (among those not diseased), and risk ratios. Symptoms occurred with similar correlation coefficients. The sensitivities and specificities of single symptoms for disease diagnosis were exhibited as equations using the three epidemiological measures and approximated using linear regression in simulated populations. The areas under curves (AUCs) of the receiver operating characteristic (ROC) curves was the measure to determine the diagnostic accuracy of multiple symptoms, derived by using 2 to 40 symptoms for disease diagnosis. With respect to each AUC, the best set of sensitivity and specificity, whose difference with 1 in the absolute value was maximal, was chosen. The results showed sensitivities and specificities of single symptoms for disease diagnosis were fully explained with the three epidemiological measures in simulated subjects. The AUCs increased or decreased with more symptoms used for disease diagnosis, when the risk ratios were greater or less than 1, respectively. Based on the AUCs, with risk ratios were similar to 1, symptoms did not provide diagnostic values. When risk ratios were greater or less than 1, maximal or minimal AUCs usually could be reached with less than 30 symptoms. The maximal AUCs and their best sets of sensitivities and specificities could be well approximated with the three epidemiological and interaction terms, adjusted R-squared ≥ 0.69. However, the observed overall symptom correlations, overall symptom incidence, and numbers of symptoms explained a small fraction of the AUC variances, adjusted R-squared ≤ 0.03. In conclusion, the sensitivities and specificities of single symptoms for disease diagnosis can be explained fully by the at-risk incidence and the 1 minus baseline incidence, respectively. The epidemiological measures and baseline symptom correlations can explain large fractions of the variances of the maximal AUCs and the best sets of sensitivities and specificities. These findings are important for researchers who want to assess the diagnostic accuracy of composite diagnostic criteria.

Mental Illness Diagnoses May Not Cause All Mental Symptoms: A Simulation Study for Major Depressive Episodes, Dysthymic Disorder, and Manic Episodes

“…Third, if the disorder does exist, it is important to accurately diagnose the disorder. However, the diagnostic accuracy of symptoms is far from perfect [ 3 ]. Although the diagnostic accuracy of symptoms, sensitivity and specificity, can increase gradually when using more symptoms to detect a disease cause, it is unlikely to reach perfect diagnostic accuracy even with more than 30 symptoms [ 3 ].…”

“…However, the diagnostic accuracy of symptoms is far from perfect [ 3 ]. Although the diagnostic accuracy of symptoms, sensitivity and specificity, can increase gradually when using more symptoms to detect a disease cause, it is unlikely to reach perfect diagnostic accuracy even with more than 30 symptoms [ 3 ]. We do not identify evidence to show incremental increases in diagnostic accuracy for a mental illness using more psychiatric symptoms [ 3 ].…”

“…Based on the latest guideline on mining innovative composite measures [30] and the recent introduction of the perspectives of patients or the public [7], we think it possible to improve the diagnostic criteria and recommend researchers to collect epidemiological evidence (prevalence and incidence of psychiatric symptoms or preferably biomarker data), assess the relationships between symptoms by consulting patients or members of the public [7], create interpretable diagnostic criteria based on the literature [30], avoid data processing steps that introduce biases to the diagnoses [27], verify the diagnostic accuracy of symptoms for well-defined diagnoses [3], and examine the likelihood of falsely defining cases in populations with nearzero prevalence [27]. When researchers are confident the identified psychiatric symptoms can be sensitive to cases and be specific enough to exclude those who are not likely to have the condition, the newly developed diagnostic criteria will become less vulnerable to the critics that the currently used DSM system is subject to.…”

“…Ideally, medical diagnoses represent underlying pathological changes or health conditions. The symptoms that are caused by these conditions and occur more frequently can be used to diagnose the underlying conditions [ 3 ]. Based on this principle, psychiatric symptoms have been identified by professionals to diagnose mental illnesses [ 4 ].…”

“…The diagnostic criteria of the above-mentioned three mental illnesses use different methods to aggregate information on the input symptoms [ 5 ]. For example, major depressive disorder can be confirmed in individuals with major depressive episodes according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR, American Psychiatric Association, Philadelphia, PA) [ 3 , 5 ]. To confirm each major depressive episode, individuals need to present five out of the nine symptoms, including depressed mood, feelings of worthlessness, and fatigue [ 5 ].…”