'Cat ladies' suicide risk probed
“Cat ladies are more likely to commit suicide,” The Daily Telegraph states, evoking images of sad and lonely self-harming spinsters surrounded by hordes of meowing kitties. The reality is somewhat different.
The Telegraph’s story is based on a study that found that Danish mothers were more likely to have developed antibodies to the parasite Toxoplasma gondii if they self-harmed at a later date. T. gondii is a parasite that can cause toxoplasmosis, and hygiene is key to preventing it. It can be caught from cat faeces, unwashed vegetables, undercooked meat and contaminated water. It can also pass from a woman to her unborn baby if she is infected during pregnancy.
Toxoplasmosis reportedly affects approximately a third of people worldwide. In most people it causes no symptoms, but it can cause severe problems in pregnant women and in people with a weakened immune system. Previous studies have linked T. gondii infection to schizophrenia and self-harm. The current study investigated this link.
Just over a quarter of a cohort of 45,788 women who gave birth had T. gondii. Of the 45,271 mothers who had not previously self-harmed, only 1% later self-harmed. Women with antibodies against T. gondii were 53% more likely to go on to self-harm. However, this study does not prove that T. gondii infection actually caused the women to self-harm. There may have been various mental health, medical, personal or social causes which this study has not explored. Furthermore, this study looked at the association between self-harm and women with antibodies against T. gondii rather than those who owned cats. There is no need to get rid of Tiddles just yet.
Where did the story come from?
The study was carried out by researchers from Aarhus University and Statens Serum Institut, Denmark and the University of Maryland School of Medicine, US. It was funded by US mental health research institutes. It was published in the peer-reviewed journal, Archives of General Psychiatry.
The Telegraph has generalised the findings of this study to all women who own cats and used an inappropriate stereotype in its headline. This is at odds with the study, which only included mothers. It looked at mothers with antibodies against T. gondii, rather than at women with cats. Although cat faeces is a common source of T. gondii infection, the infection can also commonly be caught from:
- unwashed vegetables
- undercooked meat
- cross-contamination through the reuse of kitchen utensils that have been used with raw meat without thorough washing
- contaminated water sources
What kind of research was this?
This was a prospective cohort study carried out in Denmark. It investigated the relationship between the levels of antibodies against T. gondii when women gave birth and subsequent self-harm and suicides.
Antibodies are produced in response to an infection. Because they target specific infections, they can be used by researchers to see whether someone has caught the infection – and in this case the researchers used T. gondii antibodies as a marker for the disease. Although toxoplasmosis does not pass between people, a woman who is infected while she is pregnant can pass the infection to her unborn baby. Some newborns can experience severe complications from toxoplasmosis. For this reason, some infants may be screened for toxoplasmosis, as the infants in this study were.
This is an appropriate study design to answer this question, although it has some limitations. Although prospective cohort studies can show associations, they cannot show causation. It is not known whether other factors are responsible for any observed link between having antibodies against T. gondii and self-harm. However, it would be difficult and unethical to perform a randomised controlled trial to answer this question. Also, when assessing a psychological outcome such as self-harm there is the possibility that not all outcomes have been identified. This study used registry-based data, but there are many reasons why women who self-harm may not report this to a health professional, and only self-harm which was serious enough to require medical treatment would have been captured in this study.
What did the research involve?
The study included a total of 45,788 mothers who gave birth in Denmark between 1992 and 1995 who had their child screened for T. gondii infection at birth. The women were followed up until 2006. The presence of antibodies against T. gondii was assessed in newborns, and this was used as a marker of their mother’s antibody production. This is because infected newborn children do not begin to produce their own antibodies against T. gondii until three months of age, so the antibodies in the blood of newborns must have come from the mother.
Researchers then analysed the Danish Cause of Death Register to identify any of the mothers who had died from suicide, and The Danish National Hospital Register and the Danish Psychiatric Centre Research Register to identify mothers treated following a suicide attempt or deliberate self-harm.
The researchers then calculated the relative risk of self-directed violence in women with antibodies against T. gondii depending on whether they had a pre-pregnancy history of self-harm.
What were the basic results?
Of the 45,271 mothers in this study with no history of self-harm, only 1% (488) later self-harmed. Of these 488, antibodies against T. gondii were found in 34% (168). The remaining 320 did not have the antibodies. Mothers with antibodies against T. gondii were 53% more likely to self-harm than mothers without antibodies (relative risk 1.53, 95% confidence interval 1.27 to 1.85). This increase was statistically significant. However, the overall number of new self-harm cases was low, with a rate of 8.2 new cases of self-harm per 10,000 person-years. The risk of self-harm also increased with increasing antibody levels.
Analysis of the results found that mothers who had T. gondii antibodies but no history of mental illness, were 56% more likely to self-harm than those without T. gondii antibodies and no history of mental illness (relative risk 1.56, 95% confidence interval 1.21 to 2.00). When the researchers looked at the mothers who had a past history of mental health illness it was found that mothers with antibodies to T. gondii had a 25% increased risk of self-harm (relative risk 1.25, 95% confidence interval 0.94 to 1.66), although this increase was not significant.
Mothers who had a history of self-harm were 54% more likely to have a repeated episode of self-harm if they had antibodies against T. gondii. However, the increase was not statistically significant (relative risk 1.54, 95% confidence interval 0.98 to 2.39).
There were 18 suicides during the study (more than 604,844 person years of follow-up). Mothers who had antibodies against T. gondii were twice as likely to commit suicide (relative risk 2.05, 95% confidence interval 0.78 to 5.20), although it should be noted that there was no statistically significant difference in suicide risk when comparing infected with non-infected mothers. When violent suicide attempts were analysed, it was found that mothers with antibodies against T. gondii had an 81% increased risk (relative risk 1.81, 95% confidence interval 1.13 to 2.84).
How did the researchers interpret the results?
The researchers say, “women with a T. gondii infection have an increased risk of self-directed violence”. They did not come close to saying “Cat ladies are more likely to commit suicide”, as the Telegraph put it.
Cat owners need not fear today’s news, but pregnant women and people with impaired immunity do need to continue to exercise sensible precautions. This prospective cohort study has found an association between having antibodies against T. gondii and self-harm in mothers in Denmark.
However, this cannot show whether T. gondii infection caused the women in this study to self-harm. Self-harm can have various mental health, medical, personal or social causes, and this study has not explored all of these. Significantly, when the researchers analysed whether the women had a past history of mental health illness, they found no significant association between T. gondii and self-harm. As the authors of the study do acknowledge, T. gondii infection is unlikely to be a random event. For example, the results seen could be explained if people who go on to self-harm exhibit behaviour that makes it more likely for them to become infected with T. gondii (for example, if they are less careful with hygiene).
These findings cannot be generalised to men, or to women who have not had children, as only women who had a baby between 1992 and 1995 were included in this study.
T. gondii infection is common – nearly a third of people are infected. In most cases it has minimal or no symptoms. However, in rare cases in pregnant women it can cause miscarriage or stillbirth. Pregnant women can pass the infection on to their baby in the womb (congenital toxoplasmosis), which can lead to brain damage, epilepsy and blindness.
To minimise this risk, pregnant women and people with a weakened immune system should:
- wear gloves when gardening
- cook meat thoroughly
- wash fruit and vegetables thoroughly
- avoid cat faeces in cat litter or soil
- Pedersen MG, Mortensen PB, Norgaard-Pedersen B, Postolache TT. Toxoplasma gondii Infection and Self-directed Violence in Mothers. Archives of General Psychiatry. Published online July 2012