November 04, 2005
First Born Girls 3 Times More Likely To Live To 100

First born daughters live longest. (same article here)

Predictors for exceptional human longevity may include birth order, place of birth and early-life living conditions, according to a recent Society of Actuaries (SOA) study that suggests there are several factors linked to one’s longevity. The data indicates that first-born daughters are three times more likely to survive to age 100 compared to later-born daughters. The chances for exceptional longevity are minimal for sons having a birth order of four to six compared to those born earlier or later.

The research, developed by the SOA in partnership with researchers at the Center on Aging and the National Opinion Research Center at the University of Chicago, evaluated detailed family data for nearly 1,000 centenarians born in the U.S. between 1875-1899. Drs. Natalia Gavrilova and Leonid Gavrilov collected data from publicly available computerized genealogies of 75 million individuals identified in previous studies and validated ages and birth dates by linking records to the Social Security Administration Death Master File and reviewing U.S. censuses for years 1900, 1910 and 1920.

“The study supports the idea that early childhood conditions might be important for survival to advanced ages,” said Dr. Natalia Gavrilova. “Limited access to parental care, including attention and supervision, may result in less attention being paid to the health and safety of later-born children, resulting in a higher risk of infections and malnutrition during early childhood.”

The data further suggests that children born to parents who are farmers and childhood residence in the Western region of the U.S. may be indicators for subsequent survival to age 100. The study determined that children of farming parents who lived in the Mountain Pacific and West Pacific regions of the U.S. have a greater chance of surviving to age 100 than those from the Midwest and Northeast areas of the country.

“Without the type of food processing that’s currently available, living on a farm 100 years ago meant fresher food with more nutrient value,” said Thomas Edwalds, Fellow of the SOA and chairman of the project oversight committee. “This very well might correlate to prenatal and perinatal nutrition as factors of exceptional longevity.”

People living to age 100 and beyond represent one of the fastest-growing age groups of the American population, increasing at a rate of about 4.1 percent each year.

“Actuaries are skilled at measuring risks, and this research helps us better understand the predictors of longevity and quantify the implications on society and business,” continued Edwalds. “This research also illustrates that studies on human longevity could be modernized and advanced further by using new computerized data resources such as genealogies.”

Can one use this information in any way? Without knowing the mechanism which cause these effects it is hard to say. Some possible causes of this phenomenon could be managed for better results. For example, one could imagine ways that a woman's body could treat an earlier pregnancy differently than a later pregnancy. During a first pregnancy a woman could have a larger store of minerals and vitamins to give to a fetus. So a woman could compensate with vitamin supplements. But alternatively the immune system of a woman's body could could respond to a succession of pregnancies differently and become poorer at avoiding immune responses to a pregnancy. That'd be harder to do something about.

Infection is also hard to manage. A first born child typically grows up in a home with fewer children than a later born child. So the first born does not have the later children as vectors to give the first born infections. But the later born children have the older earlier born children as sources of pathogens. When the 5 or 6 year old goes off to school, gets infected, and then brings home the infectious pathogen to infect the 1 and 2 year olds then the younger later birth children end up getting hit by more infections which each exact their toll. How to compensate for this? One could imagine that home schooling might reduce the infectious disease risk posed by older children to younger children in the same family. If the kids do not go to school to get dosed with pathogens by other kids then the kids won't bring those diseases home to infect their siblings.

Suppose a big factor in causing the above results is frequency of infections. Let me take this as an opportunity to pitch for one of my favorite public health ideas: Vaccines should be developed for more diseases including the many rhinoviruses (which cause some cases of the common cold). Development of technologies that allow much larger numbers of pathogens to be targetted by vaccines could reduce the frequency of infection during early childhood and later in life and reduce the aging caused by infection. It'd also eliminate many miserable days of being sick.

Share |      Randall Parker, 2005 November 04 02:24 PM  Aging Studies

CS said at November 4, 2005 4:34 PM:

You don't mention the possibility that the effect of birth order is in part simply a proxy for the age of parents. Sperm and egg quality, as well as maternal health, decline with age. Of course, there may also be independent birth order effects (given the stresses of pregnancy it seems very plausible.)

If both parental age and birth order are separately important for lifespan, then the effects of the demographic transition on lifespan will be ambiguous: women are having fewer children (lower average birth ranks) at later ages. Which effect will dominate?

Randall Parker said at November 4, 2005 4:48 PM:


You make two excellent points.

I wonder how much age of mother affects life expectancy. Has this been studied previously I wonder?

Demographic transition: Yes, I've wondered about the effect of this on average health as well. Certainly it must be increasing the rate of birth defects. But even in those who have no obvious birth defects probably are at greater risk of having hidden birth defects when they belong to cohorts that have higher rates of birth defects.

Bob Badour said at November 4, 2005 5:25 PM:

I can think of so many plausible factors, I do not know why you would even think infectious disease is one of them. How does infectious disease explain the increased longevity of boys born 7th or later?

Randall Parker said at November 4, 2005 5:32 PM:


Infections do cause lasting damage. Cells get killed by viruses and other pathogens and by immune system cells reacting to the pathogens. Put the infections far enough back in development and they interfere with development. That causes much larger effects.

As for patterns of infections with boys versus girls: They play differently. Perhaps little boys or little girls have larger circles of friends, more physical contact, and other differences in risk factors for passing along pathogens. How do parents treat little boys versus little girls? How much does their treatment change on successive girl babies versus successive boy babies? They might change more for girls than for boys or vice versa.

Brett Bellmore said at November 5, 2005 3:04 AM:

Which brings up another factor, that's shown up in educational accomplishment: Firstborns spend their developmental years surrounded by a higher average number of adults, and tend to mature socially much faster, whereas the later born mature socially slower due to having more children to mimic in their enviroment. Maybe it's partially a behavioral matter, with firstborns more completely adopting the adult habits of risk avoidance.

Bob Badour said at November 5, 2005 6:02 AM:


I don't see how you addressed my issue regarding the later born boys. The chance of longevity decreases for the first few boys and, if I am reading the article correctly, increases again somewhat for later born boys. I wasn't referring to the contrast with girls but questioned the plausibility that infectious disease accounts for this pattern.

I can immediately think of two things unrelated to infectious disease that can better explain such a phenomenon.

1. Let's assume older siblings leave a few cells around in the womb and that these cells affect the developmental environment in the womb. Suppose these cells act as antagonists to fetuses with a differing genotype and as protagonists for fetuses with a similar genotype. Initially, for the 2nd to 6th the chance of developing among antagonists and no protagonists might increase and after the 6th the chance of developing among protagonists might increase because perhaps there are only 4 or 5 possible polymorphisms.

2. Parenting and stress. Youngest sons with lots of older sisters might receive as much attention and care as first born children. Perhaps these later born sons are born much later and do not have to compete with brothers who are only a few years older thereby reducing stress. Chronic elevated cortisol is bad. Mmmmm'kay?

My point is there are so many plausible explanations I am not sure we have sufficient information to favour any explanation over others. I speculate the stress response in particular, which to a certain extent includes infectious disease, may have a greater influence on longevity.

Randall Parker said at November 5, 2005 8:11 AM:


Yes, I was thinking about the left-over cells from previous pregnancies when I mentioned immune response to a fetus. Fetal cells get into the maternal bloodstream and become cell types around a woman's body and perhaps even fix things (and I think I did a post about this or at least meant to). But those cells have some sort of immunological impact that might effect later pregnancies.

As for the later born boys: I do not doubt that several mechanisms are happening here to cause the observed results. Patterns of behavior of first and later born children is probably one. Immunological differences encountered by successive fetuses, nutritional differences on fetuses, stress on mom, and stll other mechanisms seem plausible at least.

BTW, I vaguely recall a paper recently reporting that later siblings are at greater risk of accidents as parents become more complacent/tired as the later siblings come along.

I agree we have a great many plausible explanations. But my intuitive guess is that infectious agents are underrated in terms of their long term impacts on health.

Carlotta said at November 5, 2005 10:49 PM:

There are numerous advantages of home-schooling, but I doubt that avoiding most kinds of infection in later-born children is one of them. Home-schoolers not only socialise, but they often socialise intensively, and with a huge number of different people. Immediately lethal infections such as bird flu might change that, but predictions about longevity most probably won't.

Warren D said at November 6, 2005 8:51 AM:

I was also always under the impression that children after the first born were larger and therefore healthier. The same effect is repeated in animals as the second litter of any young dog, cat, mouse ect. is larger and the mother takes better care of them. I would figure, barring any other problems, the mother's womb would be in better shape.

As far as fetal cell types posing problems to later children, I would think that if fetal cells are incorporated stably into the maternal body that they then would contribute to the deletion of immune cells that would be auto reactive.

One interesting thing I see about infections affecting health is the idea that as countries become more hygienic the incidences of auto-immune diseases and allergies increases in children.

momochan said at November 7, 2005 2:21 PM:

I remember a study on the presence of grandmothers affecting the survival rates of grandchildren. Again, birth order may be a proxy -- children born earlier are more likely to have a functioning grandmother present.

Regarding the Western US having greater longevity -- the first thing that leaps to my mind is that this region was industrialized later. Life expectancy impacted by industrial pollutants, in other words.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©