Health uptake model

[Robinlovelace]

Age-gender are most important variables.

Baseline level vs new level -> Marginal METs, one way of measuring all physical.

[Robinlovelace]

Marginal METS method is based on total physical activity, meaning including sport and commuting exercise, may not include occupation (may be relevant in low income countries).

[Robinlovelace]

You need a baseline level of physical activity.

[Robinlovelace]

ITHIM did not have Monte Carlo methods as of last year, long-term aim is to introduce.

[Robinlovelace]

Regarding the air pollution part, there are 2 main target populations, by which we can mean PM2.5 (which is more dispersed and has many more sources):

• When you change transport, it changes air pollution exposure for everyone
• When people change mode, they will be exposed to different levels of air pollution

An issue with PM2.5 is that it is produced by local stoves, industry, many other things. PM2.5 is not so local. NO2 can become HNO3.

Dose-response functions are in ITHIM. PM2.5 dominates the results in many developing contexts. ITHIM assumes that PM2.5 is a single value per city with a factor of 2 for active transport users.

[Robinlovelace]

The best available data varies by place. A key parameter is the % of PM2.5 that is due to local traffic, that would reduce when local traffic. Accra was the one that had most done on it.

• Global burden of disease data takes data from GBD and converts it into the right format - check inside the CSV files

[Robinlovelace]

Marko's idea was to:

• Generate a synthetic population for each city
• We account for physical activity in the individual data
• Travel data also accounted for

My thinking was to allocate individuals at the OD level and calculate responses accordingly.

[Robinlovelace]

Chat to ITHIM guys to look at way to interface.