Prof. Dr. F. Perry Wilson is an associate professor of medicine and public health and director of the Clinical and Translational Research Accelerator at Yale. His writing on science communication can be found in the Huffington Post, NPR, and here at Medscape.
In his current blog, Wilson presents a study on determinants of life expectancy. Read the Transcript of English Originalvideos on Medscape.com. It has been edited for better readability.
Welcome to Impact Factor your weekly dose of commentary on new medical studies. I am Dr. F. Perry Wilson from the Yale School of Medicine.
What is it about today? 2 people in the United States are celebrating their 30th birthday. It is a good day. They are in the prime of their lives. One is a married white woman with a college degree. The other is an unmarried white man with a high school diploma.
How many more years of life can these two people look forward to? There is a pretty dramatic difference. The man will live to about 67 years old. The woman can expect to live to 85 years old. That’s an 18-year discrepancy in life expectancy based solely on gender, education and marital status.
I use this example to illustrate extremes in life expectancy based on 4 key social determinants of health: gender, ethnicity, marital status, and education. We all have some idea of how these factors affect health. A new study suggests it’s actually a bit more complicated than we thought.
I confess to being biased. As a clinical researcher, I sometimes find it difficult to evaluate the relevance of actuarial studies that address life expectancy (or any other metric) such as marital status.
Married people live longer, that’s the conclusion. OK, but what am I supposed to do as a doctor? Encourage my patients to settle down and commit? I can’t really put studies that show that women live longer than men, or that white people live longer than black people, to much practical use. I can use my time more wisely to get people to stop smoking or eat healthier.
But studies examining such groups are a reasonable starting point for asking important questions: Why do women live longer than men? Is it behavioral (men live riskier lives and visit the doctor less often)? Or is it hormonal (estrogen has many protective effects that testosterone does not)? Or is it something else?
Relating these social determinants of health is a little more difficult than it may seem. This is shown by a study in BMJ Open published Study.
Each person in the study can be clearly classified into one of 54 groups. The people can be male or female. They can be black, white or Hispanic. They can have a high school diploma or less, an associate degree [ein Abschluss zwischen dem High-School-Abschluss und dem Bachelor] or have a college degree. They may be married, have been married previously, or have never been married.
Of course, this does not capture the entire diversity of people in a population, but let’s read on. The researchers are working with data from the American Community Survey, which includes 8,634 people. An even finer grid would cause statistical problems due to the small sample size.
The survey may be population-weighted so that the results fairly represent all people in the United States.
As part of the survey, data were collected on the four major categories of gender, ethnicity, education and marital status and compared with the CDC dataset [Centers of Disease Control and Prevention] on causes of death. From this point on, it is a fairly simple task to arrange the 54 categories in order from longest to shortest life expectancy.
But that’s not really the interesting part of this study. Sure, there are a lot of differences; it’s important that these 4 factors explain about 18 years of difference in life expectancy. What strikes me here is that there is no clear statement at this point.
Let me show the 2nd figure in this paper because it illustrates the surprising heterogeneity very nicely.
At first glance, symbols higher up the Y-axis represent the groups with a longer life expectancy. White men who have never been married and have a low level of education are at the bottom of the axis. Married Hispanic women, on the other hand, fare quite well in terms of life expectancy, even with lower levels of education.
The authors quantify this phenomenon by creating a mortality risk score. 0 represents the average morality in the United States. The score looks like this:
As you can see, you get a lot of points for being female, but you lose a lot of points if you are not married. Education plays a big role, with a minus for people who have a high school diploma or less, and a bonus for those with a college degree. Ethnicity is of rather minor importance.
This is all very interesting. But as I said, as a doctor, this knowledge is of little use to me. What is more important is to find out why these differences exist. And there are some clues in the study data, especially when we look at the causes of death.
The following figure ranks all 54 groups again, from married white women with a college education to unmarried white men with a high school diploma. The boxes show how much more or less likely this group is to die from a particular disease than the general population.
Looking at the lower groups, there is a dramatically increased risk of death from unintentional injuries/accidents, heart disease, and lung cancer. There is also an increased risk of suicide. In the upper groups, the risk only seems to be higher than expected in the category of “other cancers,” which reminds us that many cancers do not follow socioeconomic status.
One can even update the risk scoring system to account for risk for different causes of death. You can see here that white people, for example, have a higher risk of death from unintentional injuries/accidents than other populations, despite having lower mortality overall.
So perhaps the cause of death will bring us a little closer to answering the question of why. But this study is really just a start.
What is surprising to me is that in a wealthy country like the United States, such dramatic differences exist that, with the exception of gender, are not biological. This means that we may need to turn from physiology to sociology to explore causes.
This article originally appeared on Medscape.com As part of the translation process, our editorial team may also use text editing software including AI.
