These methods use frequentist approaches to estimate prevalence and confidence limits, assuming a fixed pool size and a test with known values for sensitivity and specificity, as described below.
Este método (Método 4 deCowling et al. (1999) assumes a fixed pool size and that test sensitivity and specificity are known exactly (no uncertainty about their values). Confidence limits are based on a normal approximation and may be <0 for low prevalence values.
La prevalencia se estima como:
y el error estándar (SE (p)) se estima como la raíz cuadrada de la varianza, dada por:
donde:
Los límites de confianza asintóticos se calculan utilizando la aproximación normal:
donde es la variable normal estandarizada correspondiente al límite de confianza deseado.
Las entradas requeridas para este método son:
Pool size, number of pools and number of pools positive must be positive integers and the number of positive pools must be less than the number of pools tested. Sensitivity and specificity must be >0 and <=1 and upper and lower confidence limits must be >0 and <1.
Las salidas incluyen:
Estimates are only valid if the proportion of positive pools is greater than the false positive rate (1 - Specificity) and less than or equal to the true positive rate (Sensitivity). Invalid results are indicated by NA in the results table.
Este método (Método 5 deCowling et al. (1999)) asume un tamaño de grupo fijo y que la sensibilidad y especificidad de la prueba se conocen exactamente (sin incertidumbre acerca de sus valores). Los límites de confianza exactos se calculan en base a la teoría binomial, de modo que los límites de confianza nunca son <0 o >1.
La prevalencia y la varianza se estiman como paraMétodo 3:
y:
donde:
Exact confidence limits are estimated by calculating the corresponding binomial confidence limits for the proportion of positive pools and then transforming these back to individual-level prevalence values using the equation for estimating prevalence from Método 3.
Las entradas requeridas para este método son:
Pool size, number of pools and number of pools positive must be positive integers and the number of positive pools must be less than the number of pools tested. Sensitivity and specificity must be >0 and <=1 and upper and lower confidence limits must be >0 and <1.
Las salidas incluyen:
Estimates are only valid if the proportion of positive pools is greater than the false positive rate (1 - Specificity) and less than or equal to the true positive rate (Sensitivity). Invalid results are indicated by NA in the results table.