Skip to content

Proportional multi state life table

Belen edited this page Mar 28, 2018 · 4 revisions

Proportional multi-state life table

The proportional-multi state life table (PMSLT) is a comparative risk assessment method to simulate changes in health burden of changes in exposure to health risk factors. In the ITHIM R project the PMSLT is being developed, in this first instance, as a long script, and we aim to develop a package in the future. The PMSLT is presented as an alternative to the ITHIM original approach of estimating the change in burden of disease by directly applying potential impact fractions (PIFs) to population health measures (e.g. years of life lost, years lived with disability) attributable to the risk factors of interest (e.g. physical inactivity, air pollution and road trauma).

The PMSLT may be used when the user is interested in including trends (e.g. mortality, incidence) and time lags between changes in exposure to health risk factors and health outcomes. The PMSLT also allows for a life time approach, rather than only estimating a yearly change in the burden (or summing years) as done on ITHIM. In this manner, the PMSLT allows to take into account the effect of changes in life years on the burden of disease.

Briefly, the PMSLT consists of two main components: a general life table and a life table for each of the included disease.

Inputs

  • General life table: mortality rates per one year intervals, population numbers (depends on the age cohorts of interest, for example, 5 years) and total prevalent years lived with disability. Details on data sources are provided in the documentation for the PMSLT development.

  • Disease life tables: incidence and cases fatality. Such data is rarely available, and for a large scale project, global burden of disease data generated by the Institute of Health Metrics and Evaluation and Dismod II can be used to derive them.

Clone this wiki locally