search

ITHIM / ITHIM-R

Development of the ITHIM-R, also known as ITHIM version 3.0. Started in January 2018.
https://ithim.github.io/ITHIM-R/
GNU General Public License v3.0
19 stars 11 forks source link

Background burden of disease data (deaths, YLL, YLD and DALY) #21

Open markotainio opened 6 years ago

markotainio commented 6 years ago

Discussion on where and how to get background burden data. Burden meaning background:

-- The main global source for burden data is the IHME Global Burden of Disease project which provides age, gender and disease specific background burden for all countries. Most recent data (2016) can be downloaded from GBD Results Tool: http://ghdx.healthdata.org/gbd-results-tool

Population data for countries is here: http://ghdx.healthdata.org/record/global-burden-disease-study-2016-gbd-2016-population-estimates-1950-2016

More information on data is available here: http://ghdx.healthdata.org/gbd-2016

-- I have not examined if IHME provides possibility to download data automatically. Even if this would be possible, I would build model assuming that user provides background burden data. Otherwise ITHIM-R will be dependent on functionality of the IHME data tool.

-- Currently ITHIM estimates background burden for study area based on population differences between study area and country. In some settings this could be improved e.g. using local mortality data.

-- For injuries we also need deaths to YLL, YLD and DALY conversation ratios by assuming that in most setting we have data only on fatalities, which we will then use to estimate total burden, including injuries.

JDWoodcock commented 6 years ago

We also need to add about downscaling to city level using mortality rates. We should also discuss how we deal with the two potential risks of overestimates a) YLLs assuming maximum life expectancy, b) YLLs assuming lived in full health

markotainio commented 6 years ago

It would also be informative to see how much better our predictions are if we use more refined downscaling methods? So far we have used (in UK) mainly population numbers by age and gender, by assuming that differences between country and city are not large enough to change the conclusions.

gotom22 commented 6 years ago

Regarding downscaling please see #38

markotainio commented 5 years ago

Case city GBD data is available here (in v-drive): V:\Studies\MOVED\HealthImpact\Research\TIGTHAT\Case cities data

Note! Endometrial cancer is in GBD IHME data "Uterine cancer".