In a Psych. 101 class many years ago, I remember learning that some mental disorders were more common among people born in winter and spring. The reason was unclear. I pretty much forgot about this interesting factoid until reading about a recent study that found evidence of a link between some mental illness and the seasons: the flu.

The new study, published online this month in the journal JAMA Psychiatry, found that in utero exposure to maternal influenza was associated with a four-fold increase in the likelihood of developing bipolar disorder. The disorder, once known as manic depression, causes dramatic shifts in mood and energy, and can inhibit the ability to carry out day-to-day tasks. It affects an estimated 2.6 percent of American adults each year.

The research analyzed data from the Child Health and Development Studies, a large cohort of 19,044 people born in Alameda County, Calif., between 1956 and 1966. All of the study participants’ mothers were members of the Kaiser Permanente insurance plan when they gave birth—providing the researchers with fairly comprehensive data on maternal health and information on whether or not they had the flu while pregnant.

To test the hypothesis that in utero exposure to maternal influenza is associated with developing bipolar disorder as an adult, the researchers first had to identify all members of the Child Health and Development Study who developed the disorder after childhood. To do this, they reviewed Kaiser's medical database and records from Alameda County Behavioral Health Care Services, and sent questionnaires to all study participants and their mothers. After all possible cases were identified, a structured clinical interview was conducted with each person to independently confirm bipolar disorder. There were 92 cases of the disorder.

For each person with bipolar disorder, they then identified up to eight other study participants who did not have the disorder but were the same sex and were born within 30 days of the person with bipolar. These 722 people served as a matched control group that was as identical as possible to the group of people with bipolar disorder. At this point in the analysis, the researchers did not know any participants' family histories of the flu. If maternal influenza played no role in the development of adult bipolar disorder, then the rate of maternal influenza would theoretically be equal in both groups. They then reviewed medical records of all the mothers' pregnancies.

The finding: 8.7 percent of mothers of the adults with bipolar disorder had had the flu at some point while they were in the womb compared to 2.6 percent of the mothers of adults without bipolar disorder. This strong association persisted after controlling for factors such as maternal mental health, age, race, educational level, and gestational age of birth. Maternal exposure to influenza was more common among people with bipolar disorder across all three trimesters of pregnancy and the period just before and just after conception.

Why might having the flu while pregnant increase the risk of a child developing bipolar disorder as an adult? The researchers are not sure, but some evidence suggests that a deficit of reelin, a glycoprotein secreted by the brain, might have something to do with it. Post-mortem examinations of the human brain have found that people with bipolar disorder have substantially fewer reelin-positive cells than people without the disorder. Animal studies also provide evidence of a link between reelin and the flu: When mice were injected with influenza virus while pregnant, scientists found that their offspring had lower reelin counts. Other research has identified a relationship between maternal influenza, reelin, and schizophrenia in humans.

The evidence from the recent study is not sufficient to say that the flu causes bipolar disorder. Its results are compelling, however—especially when considered within the larger body of research on maternal influenza and adverse mental health outcomes.

The bottom line: get the flu vaccine this fall if you’re pregnant or might become pregnant (6% of the mothers in the study cohort had been vaccinated.) The vaccine lowers but does not erase the likelihood of contracting the flu. So the more people that receive the vaccine, the less likely it is that an expectant mother will get the flu from them. Lastly, Philadelphia City Council and Mayor Nutter ought to consider this evidence next time the paid sick leave bill comes up. People go to work with the flu if they don’t have paid sick leave—passing the virus on to pregnant mothers and potentially shaping the mental health trajectories of their unborn children.

Read more about The Public's Health.