Jump to comment:

• Absolute Risk Difference is what matters
  Peter A West
  Published on: 9 August 2017

• Published on: 9 August 2017

  Absolute Risk Difference is what matters

  • Peter A West, Health Economist Peter West Associates Ltd.

  This paper makes a number of claims about health in the North relative to the South of England using comparisons of relatively low death rates. When the denominator in such calculations is a very low rate of death, the size of the difference can appear large. However, if we compare the absolute risk of dying, it is relatively close in the North and South and if we were to divide the rate of survival in the South by the rate of survival in the North each year, we would have a very small comparative statistic.

  Abstracts and conclusions can easily be taken out of context and authors of papers like this one should be careful to present appropriate information. For example, the conclusion "...1.2 million northern excess deaths under age 75 over five decades.." implies very high potential death rates, a million! But this figure is presented with no population and reflects experience over 50 years. If we divide by 50, we get 24,000 deaths a year. A further weakness is that no measure of population is provided to put this total number of deaths in context. Using a plausible estimate of 20 million, for example, implies excess deaths at a rate of about 1.2 per 1,000 people. I wonder how many residents of the North are planning to migrate South today to reduce their risk of an early death by just over 1 in 1,000. Yes we should be concerned about all differences in health across regions and social groups but by inflating them with misleading divisions of one small num...

  Show More

  This paper makes a number of claims about health in the North relative to the South of England using comparisons of relatively low death rates. When the denominator in such calculations is a very low rate of death, the size of the difference can appear large. However, if we compare the absolute risk of dying, it is relatively close in the North and South and if we were to divide the rate of survival in the South by the rate of survival in the North each year, we would have a very small comparative statistic.

  Abstracts and conclusions can easily be taken out of context and authors of papers like this one should be careful to present appropriate information. For example, the conclusion "...1.2 million northern excess deaths under age 75 over five decades.." implies very high potential death rates, a million! But this figure is presented with no population and reflects experience over 50 years. If we divide by 50, we get 24,000 deaths a year. A further weakness is that no measure of population is provided to put this total number of deaths in context. Using a plausible estimate of 20 million, for example, implies excess deaths at a rate of about 1.2 per 1,000 people. I wonder how many residents of the North are planning to migrate South today to reduce their risk of an early death by just over 1 in 1,000. Yes we should be concerned about all differences in health across regions and social groups but by inflating them with misleading divisions of one small number by another, researchers are contributing to the sensationalisation of such differences.

  The paper citation is Buchan IE, et al.,North-South disparities in English mortality 1965– 2015: longitudinal population study, J Epidemiol Community Health 2017;0:1–9. doi:10.1136/jech-2017-20919

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top