Цитата:
Сообщение от Melnichenko
generic inverse variance method ??? Не могу найти наш раздел с помощью в перводе , а нужно срочно !! Если супермодераторы поправят - спасибо !
|
Обычно для мета-анализа - метод объединения данных на основании взвешивания обратно-пропорционально дисперсии величины эффекта в индивидуальном исследовании. Устоявшегося русского эквивалента не знаю, перевел бы как
"[...с использованием] общего метода взвешивания обратных дисперсий".
Пояснения:
The inverse variance method of meta-analysis is a widely applicable approach to meta-analysis. It involves a weighted average of the effect estimates from the separate studies. The weight for each study is taken to be the inverse of the variance (one divided by the square of the standard error) of the effect estimate.
A Generic inverse variance outcome requires two numbers: an effect estimate and its standard error for each of the trials to be combined. The effect estimate summarises the treatment effect in a clinical trial (for example as an odds ratio, a mean difference or a hazard ratio) and the standard error summarises the precision of the estimate (the sampling error within the study). These numbers may be used to perform a meta-analysis to combine results across multiple studies using the inverse variance method.
The Generic inverse variance outcome should be used only when it is impossible or inappropriate to enter data as dichotomous, continuous or individual patient data.
The most likely situations are:
◊ special study designs such as cross-over trials and other matched designs, cluster randomized trials or non-randomized studies
◊ special types of outcome such as time-to-event outcomes, ordinal outcomes or rates (events per unit time)
◊ special effect measures, such as hazard ratios, ratios of means or adjusted estimates.