The Status of Baby Boomers' Health in the United States:  The Healthiest Generation?

This is an interesting and thought provoking paper, and the overall message of more prevention for the health of the baby boomers generation is welcomed. Nevertheless, while the authors control for several demographic factors, they forgot to control for the most relevant factor, namely survivorship. 

With every new generation, the proportion of people surviving longer increases. People who earlier would have died by diseases, which can now be treated more effectively, are surviving to reach higher ages. This means that the baby boomer generation is more heterogeneous in terms of health, because more frail persons have survived until now. This is different from previous cohorts, where higher mortality levels executed a selection to old age process. 

Given the rich historical data on mortality available, for instance in the Human Mortality Database, the exact survivorship of cohorts can be calculated. These data show that the survival of the generation of baby boomers up to the age of the NHANES surveys is 5% higher than that of previous cohorts, e.g. those born in the 30s. This increased survival observed in the baby boomer generation implies that people with a higher level of frailty, who likely would have died if mortality had not reduced, are now alive. The increased survivorship of the baby boomer generation has an impact on the many weak disparities presented in the paper by King et al., for which they have not accounted. The baby boomers’ health is negatively affected by the survival of the more frail population segments. A simple way of examining this would be to exclude the lowest 5% of NHANES’ results for health of the population of baby boomers and test if the results still hold. Amazingly the lower smoking prevalence found for the baby boomers, and noted in the paper, will be even more extreme in their favor. 

The authors write in their comment “Despite their [baby boomer’s] longer life expectancies over previous generations…”. However, these are cohorts that are still alive today and it is not possible to calculate life expectancy for the whole life span, but only a “temporary life expectancy from birth to age 50 or 60”. To calculate those life expectancies, it is necessary to first have the survival of the cohorts.  If the authors had made these steps, they would have faced the result of having to control for the extra survival of the baby boomer generation.

An additional comment is that there seems to be an overlap between the two generations studied: the “previous cohort”, those reaching ages 46 to 64 in NHANES III (1988-1994), were born from 1924 to 1948; the baby boomers’ years of birth as defined by the authors are 1943 to 1964, or those reaching ages 46 to 64 in NHANES (2007-2010). Those born 1943 to 1948 are present in both cohorts, which seems problematic if not properly defined. I agree with the authors that more prevention is needed to improve the health of baby boomers, and of all Americans!