Statistics in Medicine: Calculating confidence intervals for relative risks (odds ratios) and standardised ratios and rates

摘要: Gardner and Alunan explained the rationale for using estimation confidence intervals in making inferences from analytical studies described their calculation means or proportions differences.1 In this paper we present methods calculating other common statistics obtained medical investigations. The techniques obtaining estimates of relative risk are described. These can come either an incidence study, where, example, frequency a congenital malformation at birth is compared two defined groups mothers, case-control where group patients with disease interest (the cases) another people without controls). standardised ratios rates incidence, prevalence, mortality Such commonly calculated to enable appropriate comparisons be made between study after adjustment confounding factors like age sex. most frequently used indices ratio (SIR) (SMR). A worked example included each method. calculations have been carried out full arithmetical precision, as recom? mended practice,2 although intermediate steps shown rounded results. Some given large sample approximations not reliable fewer than about 20 cases. Appropriate design principles these types adhered since convey only effects sampling variation on precision estimated cannot control errors such biases due selection inappropriate controls collecting data.