Kevin Lomangino is the managing editor of HealthNewsReview.org. He tweets as @KLomangino.
Many everyday things have been proposed to cause Alzheimer’s disease over the years — deodorant, dental fillings, diet soda, and flu shots, to name just a few.
But none of these proposed causes has stood up to sustained scientific scrutiny, and the alarm bells raised by early research have turned out to be false alarms.
Now here comes NBC News with a new factor that apparently “may affect” Alzheimer’s risk: motherhood.
Specifically, according to NBC, “Women who had three kids had a lower lifetime risk of Alzheimer’s than women who had one child, researchers found.”
The implications for women who have fewer children is clear: This decision may increase the risk of Alzheimer’s disease later in life.
But is this new culprit–based on an observational study of about 15,000 women reported at the Alzheimer’s Association conference in Chicago–any more likely to be guilty than the previous suspects?
Stephen Soumerai, ScD, Professor of Population Medicine at Harvard Medical School and Harvard Pilgrim Health Care Institute, says it’s way too soon to know.
“This unpublished abstract should be considered extremely preliminary, but the scientists and news reports would almost have women believe that having three children will protect them from senile dementia decades later,” he wrote via email. “The causal words imply this but the data do not begin to show it.”
He said that the influence of “hormones” on the body — how researchers explain the apparent link to motherhood — isn’t a solid justification for the findings, because household income and dozens of other factors could be responsible for the observed results. Not to mention that a woman’s overall health might affect not just how many children she has, but also her dementia risk later in life.
“This is yet another example of a correlational analysis that could influence important lifestyle decisions in ways that adversely affect quality of life,” he said. In other words, people might actually plan their families based on the idea that having more kids is protective against Alzheimer’s. But this study simply isn’t capable of telling us whether there’s any benefit from this approach.
Susan Molchan, MD, a psychiatrist and former NIH clinical researcher who studied Alzheimer’s disease, agreed that there’s “no real information” here.
“Many correlations were done and one is bound to find what seems like significant associations that may superficially appear to be causal if one does enough correlations,” she wrote via email. In other words, if you look at enough different outcomes, you’re bound to find some that look interesting due purely to the play of chance.
She noted that the results seem to conflict with those of another recently reported study which found that having 5 or more children supposedly increases Alzheimer’s risk. NBC mentions those findings but doesn’t seem to consider their implications.
“One study finds 3 children associated with a lower risk, another finds that 5 children is a higher risk,” Molchan said. “It’s hard to fit this into biological plausibility and we’re left with the fact that dementia is multi-factorial, that women live longer, and that age is the greatest risk factor for dementia.”
Surprisingly, given this context, you won’t find a single note of caution about the limits of this kind of research in NBC’s coverage. The only source interviewed for the story is a representative of the Alzheimer’s Association, which has a clear stake in amplifying the importance of this study presented at its conference.
Three other quotes are ripped from news releases, one of which floats “hormone-based preventive strategies” as a potential intervention to reduce risk. In another instance readers aren’t told the source of the cribbed quote.
This kind of incomplete and imbalanced reporting may help NBC News generate more clicks for its coverage, but it’s doing a real disservice to readers who may make decisions based on this information.
Consumers, journalists, publicists, and researchers who want to understand how to better communicate results of these types of studies should read our primer: Observational studies: Does the language fit the evidence? Association vs. causation.