Parameters used (Table 1, page 198, Hyder, 1998):
(1) incidence rate per 1,000 people in the population per year
(2) average age of onset, in years
(3) average age at the time of death, in years
(4) life expectancy in years at the time of disease onset
(5) case fatality ratio (proportion of those developing the disease who die from the disease), expressed as a decimal fraction from 0.00 to 1.00
(6) case disability ratio (proportion of those developing the disease who have disability from the disease), expressed as a decimal fraction from 0.00 to 1.00
(7) extent of disability, expressed as a decimal fraction from 0.00 to 1.00, where 1.00 indicates very severe disability
(8) average duration of disability for those disabled by the disease, expressed in years
HeaLYs per 1,000 people per year =
= (incidence of disease in 1,000 people) * ((morbidity factor) + (mortality factor))
morbidity factor =
= (case disability ratio) * (extent of disability) * (average duration of disability)
mortality factor =
= (case fatality ratio) * ((life expectancy at time of onset) - ((average age of death in those affected) - (average age of onset)))
NOTE: On page 198, Hyder et al (1998), the authors mention that Excel spreadsheets for their HeaLY programs are available from the authors.
Limitations:
• Robine (1998) feels that attempting to use the HeaLY to express the impact of disease in a single value may not be justified.
• Trying to base cost savings for health care based on HeaLY needs to consider the additional health care costs incurred by having individuals live longer and surviving to more cost-intensive ages.
• Precise results require that the calculations be somewhat complex. To simplify the calculations runs the risk of oversimplifying variables and giving a false sense of understanding very difficult problems.
Discounting
Discounting is a controversial subject in the calculation of HeaLY and other disease burden measures. A discussion is given in Barnum (1987). Discounting may make sense when discussing the currency savings that can be expected over time, but I do not understand its use on lives or years of life.
Robine (1998) points out that the life-expectancy gains resulting from the suppression of a cause of premature mortality are generally greater the lower the general mortality. Discounting may serve to adjust an ideal gain to more realistic levels when there are other causes for increased mortality in the society that a survivor might then be exposed to.
The proposed discounts range from 0 to 20%, with 3-6% being common. The discounts are applied per annum in HeaLY to the life expectancy at age of disease onset and the duration of disability. The discounted HeaLY may be fairly close to the undiscounted value, but it can be only a half or a third of undiscounted value for some diseases, like malaria.
