Annals of epidemiology · 2021
Direct modeling of relative and absolute risks in register data: Mortality risk in sarcoidosis.
СтатьяПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Gabriel Erin E, Arkema Elizabeth V, Sachs Michael C
Аннотация
PURPOSE: This paper aims to illustrate the use and interpretation of regression based on pseudo-observations for estimating risks of time-to-event outcomes in epidemiological studies.
METHODS: We use pseudo-observation based regression for estimation of contrasts in the relative and absolute risks at specific times. This relaxes the proportional hazards assumption and directly estimates relative and absolute risks without the need for secondary calculations or standardization. Statistical software is available to use this method, and we demonstrate its use in a reanalysis of the mortality risk in sarcoidosis patients in Sweden.
RESULTS: We report estimated adjusted mortality risk differences and risk ratios by age, and at different years of follow up. Compared to the hazard ratio of 1.62, which is assumed to be time constant, we find risk ratios ranging from 1.7 at 2 years of follow-up to 1.3 at 10 years.
CONCLUSIONS: Pseudo-observation regression is a flexible and powerful tool for censored time-to-event data. The models are easy to run and interpret so they should be considered a standard tool alongside Cox regression and standardization. As with any statistical model, there are assumptions needed for valid inference, which should be assessed on a case-by-case basis.
METHODS: We use pseudo-observation based regression for estimation of contrasts in the relative and absolute risks at specific times. This relaxes the proportional hazards assumption and directly estimates relative and absolute risks without the need for secondary calculations or standardization. Statistical software is available to use this method, and we demonstrate its use in a reanalysis of the mortality risk in sarcoidosis patients in Sweden.
RESULTS: We report estimated adjusted mortality risk differences and risk ratios by age, and at different years of follow up. Compared to the hazard ratio of 1.62, which is assumed to be time constant, we find risk ratios ranging from 1.7 at 2 years of follow-up to 1.3 at 10 years.
CONCLUSIONS: Pseudo-observation regression is a flexible and powerful tool for censored time-to-event data. The models are easy to run and interpret so they should be considered a standard tool alongside Cox regression and standardization. As with any statistical model, there are assumptions needed for valid inference, which should be assessed on a case-by-case basis.