As an public health epidemiologist, I spend a lot of my time figuring how the best way to measure the population burden of an illnesses. Most of the time we have to use which ever data source is most convenient, often administrative hospital or death records, and occasionally larger population surveys. Understanding population health trends helps planners set priorities and give them confidence that public health interventions are working to reduce disease. The biggest problem with trying to link changes in rates of diseases to specific public health interventions is the difficulty in attribution. The causes of non-communicable diseases are very complicated, and extremely difficult to pin on a single factor. For example, rates of teen pregnancy are decreasing. With the data sources we have it’s impossible to tell if that change is related to public health interventions like free condoms and sex education (there are, of course, so many more as well), or changes in any number of underlying population characteristics (improving economy, influx of certain immigrant groups, change in religious beliefs, etc.). Public health practitioners would love to claim credit for the decreasing rates, but the data sources we rely on are simply not powerful enough to determine attribution for trends. In reality, it might be that reality shows like Teen Mom, and 16 and Pregnant are scaring teens being more careful.
Diseases that affect people at the end of their life are even more complicated, because the many of the underlying causal factors occurred much earlier in life. This problem often arises when epidemiologists use cross-sectional data (ie data collected at a single point in time, instead of longitudinally or retroactively over a longer period of time) to try and understand trends. If you are trying to understand the relationship between income and heart disease, knowing the a person’s income at the time of diagnosis or death would not be as helpful as knowing their income level 10 or 20 years earlier, when their income level might have had a much larger impact on their getting heart disease.
Income is widely believed to be one of the single best predictors of disease. It’s far from perfect, but in general, higher income earners are in a better position to access resources which will lead to better health (better housing, healthier food, less stress, more leisure time for physical activity, greater access to physicians and preventative healthcare, and the list goes on…).
The other powerful predictor of health is the effect of early childhood experiences. Poor nutrition during childhood can have lasting biological impacts. Stress and abuse can also leave psychological scars which are even more difficult to detect, and potentially even more damaging. Canadian research Clyde Hertzman developed a theory of biological embedding, which basically describes how early life experiences have a way of literally getting under the skin, and can actually change a person’s DNA. Epigenetics is not really my area, but my understanding is that through biological process such as methylation (help me out here Pat!), the DNA is permanently changed, altering biology and development, and these changes can even be inherited by future generations.
The impact on health is significant. The experiences don’t have to be as violent as abuse, but Dr. Hertzman and others have been able to show that poor child development can have lasting impact on future grades, opportunity, and ultimately long-term health. Canada, unfortunately, ranks last among OECD countries in terms of investing in early child development (usually thought of as before age 5).
Something that really makes me feel encouraged is that interest in investing in early childhood development is an idea that isn’t confined to public health. TD bank has also been supporting investing this area, as a means to foster a healthier and more productive workforce, something that pays big dividends in future economic rewards. I find it so interesting to see other groups addressing this same issue, it’s fascinating to see the issue framed in terms of return on investment instead of improvements in health!