Table 3 Assumptions and the assessments of the simulated symptoms.
From: Diagnostic accuracy of symptoms for an underlying disease: a simulation study
Assumptions | ||
1 | 2 diseases of interest: one disease directly related to the symptoms and the other associated with the disease only (unrelated to symptoms) | |
2 | Similar baseline incidence rates among those not diseased and similar risk ratios for the symptoms | |
3 | Accurate disease statuses; symptoms reported accurately by patients | |
4 | The products of baseline incidence rates and risk ratios less than or equal to 1 | |
5 | Similar baseline correlations between symptoms among those diseased or not diseased | |
Epidemiological measures of symptom occurrence in simulations | ||
1 | Population sizes | 10,000 |
2 | Number of symptoms that can be caused by the disease | 40 |
3 | Correlations between the disease that caused symptoms and the other associated disease that did not cause symptoms | 0, 0.3, and 0.7 |
4 | Prevalence rates or proportions of the population with the disease that caused symptoms | 0.05, 0.1, 0.2, 0.4, and 0.8 |
5 | Baseline incidence rates of the symptoms, similar to all symptoms | 0.05, 0.1, 0.2, 0.4, and 0.8 |
6 | Risk ratios of developing symptoms if diseased | 0.5, 1.0, 2.0, 10.0, and 25.0 |
7 | Correlations between symptoms | 0, 0.4, and 0.8 |
8 | Number of simulations for each combination of the above measures 3 to 7 | 10 |
Statistics for assessment | ||
Correlations between symptoms | ||
Diagnostic test accuracy (sensitivities, specificities, and AUCs) of the symptoms for the detection of the disease | ||
Diagnostic test accuracy (sensitivities, specificities, and AUCs) of the symptoms for the detection of the unrelated disease | ||
