Life is lived on a bell curve. Many attributes of a population – height, for example - are distributed on a bell-shaped curve, with the average at the centre and then decreasing numbers of people as we get further from the centre.
At each end of the curve are the small number of people who are either extremely tall or short. This pattern is found throughout nature, and is one of the most important concepts in biology, medicine and public health.
Understanding the bell curve is important for the work of public health. For example, we know that being overweight or obese increases the risk of developing diabetes. An example from a Canadian population health primer notes that those who are very obese have a 32 percent chance of developing diabetes over the next 10 years, while those who are obese have a 21 percent risk. But people who are overweight but not obese have only a 10 percent risk, and those with a normal healthy weight or who are underweight have only a 3 to 7 percent risk.
So you might think it would make sense to focus our prevention efforts on those who are obese – and you would be wrong. Because in doing so you would miss 61 percent of those who develop diabetes. Forty percent of cases would occur in the overweight population and an additional 21 percent of cases would occur in the low-risk normal weight population.
This is known as Rose’s Paradox, identified by the noted British epidemiologist Sir Geoffrey Rose. He pointed out that while the people at one end of the bell curve have a higher risk of getting a disease, more cases are actually found in the population with moderate or low risk. This is because there are far more people in these categories. For this reason, it is better to try to shift the curve for the entire population a bit.
Moreover, this doesn’t just apply to individuals, but to entire neighbourhoods. My friend and colleague, the late Clyde Hertzman, established and led the Human Early Learning Partnership (HELP) at UBC. He led pioneering work in BC on early child development, and as a result, BC became the first jurisdiction in the world with maps of early development for every neighbourhood and school district in the province. These maps helped to show the relationships between patterns of vulnerability in young children and their socio-economic conditions.
As would be expected, lower incomes and more impoverished living conditions and neighbourhood resources were linked to worse outcomes. But importantly, HELP also showed that “although the highest risk of vulnerability is found in the poorest neighbourhoods of town, the largest number of children at risk is spread across middle-class neighbourhoods”.
This has important implications for public health policy and programs. It is tempting to focus only on the small number of high-risk people, groups and communities – so called ‘targeted’ interventions – because it seems as if that would be cheaper. But its not a very effective strategy because it misses most of the cases. For example, BC’s Nurse-Family Partnership provides regular visits by a public health nurse throughout a woman’s first pregnancy, and those visits continue until the child reaches two years of age.
But it is only available to a select group of women; those under 19, or those aged 20 – 24 who are lone parents, or have low income and education or are experiencing social, financial, or housing challenges, including being homeless. Nobody would argue that this is not a high-risk group, but Rose’s Paradox and Clyde Hertzman’s work suggest the program may be missing most of the cases that need support.
If we want to have the greatest impact, we need to affect the entire population, What is needed, as Clyde and his colleagues at HELP point out in the BC Atlas of Child Development, is a combination of civil society interventions that “create family-friendly environments across class and ethnic divides”; universal interventions, with barriers to vulnerable people removed, and targeted interventions.
In the UK, this is known as ‘proportionate universalism’; everyone gets the intervention, but those with the greatest need get more. It's the best way to shift the curve towards health.
© Trevor Hancock, 2019