Burden of disease, health and population data

For this project, the harmonized health statistics database as held by the World Health Organization was used. This database provides health data specific for each country, health endpoint as defined by the Environmental Burden of Disease -programme, age group and gender. We used data (deaths and DALYs) for the year 2004 (WHO, 2009b; more detailed data available on request)(World Health Organization. The global burden of disease: 2004 update. Geneva: World Health Organization; 2009. Available at: http://www.who.int/healthinfo/global_burden_disease/2004_report_update/en/index.html).

The data were obtained in discounted/age-weighted and undiscounted/un-age-weighted format.

Depending on the type of calculation (see section 2.1), different data were needed. For calculations according to methods 1A and 2A (Table 3-19 shows which methods were used for which calculation), the total YLL and YLD were needed per age group and country. The WHO database provides data for pre-defined age categories (e.g. 0–4; 5–14, etc). Age-specific values were derived assuming an equal distribution of people within the age categories in the WHO data. Table 3-15 shows a sample of the burden of disease as available from the WHO database, aggregated over all ages and for a selection of health endpoints only.

For calculations according to method 2B (see section 2.1 and Table 3-19), no background health data was applied; the incidences were calculated using a unit risk model and the burden of disease was estimated using the WHO disability weights and duration estimates. For these calculations (IQ loss and HTD from lead), no age-weighing was applied due to the lack of information on the age distribution of the effects; however, the impact of the simplification was estimated to be small and to affect only the discounted results. Table 3-16 shows the disability weights and durations that were used.

We have carried out preliminary calculations to investigate the potential effect of lag times on the discounted estimates. The lag times used per health endpoint are provided in Table 3-17.

Population data (number of people in 2004) were used to calculate numbers of DALYs per million people and are provided in Table 3-18.