Test for postnatal depression

“A simple blood test could soon be used to predict if a woman will suffer from postnatal depression” says The Daily Telegraph.

The paper reports on new research, saying that testing levels of a hormone produced by the placenta could be used to predict three-quarters of depression cases in new mothers. According to the newspaper the researchers claim the test could one day become standard practice for expectant mothers.

The report is based on a study of 100 women which found that high levels of the hormone predicted postnatal depression with an accuracy of 75 per cent. The researchers also say the test was even more precise when performed on pregnant women who were already suffering from symptoms of depression.

Identifying pregnant women who will need post-natal mental health support is of great interest, and may lead to new interventions and support. However, at this time we do not know if the test would be accurate enough for use in unselected women, where there is room for misdiagnoses that might cause unnecessary distress. To be suitable for wide-spread use on its own this test would need to be more accurate, or combined with other screening tests.

Where did the story come from?

This research was conducted by Dr Ilona Yim and colleagues from the University of California, Cedars-Sinai Medical Center in Los Angeles and Chapman University in California.

The research was supported by awards from the National Institutes of Health and the National Institute of Child Health and Human Development in the US. It was published in the peer-reviewed medical journal Archives of General Psychiatry.

What kind of scientific study was this?

This was a cohort study looking at the association between levels of hormones and postpartum (postnatal) depression, which is thought to affect more than 70,000 women in Britain each year. Previous research has suggested a possible link between the two.

The researchers recruited 100 expectant mothers from a larger sample of women attending two medical centres in southern California.  They excluded any women who were expecting twins, did not speak English or had alcohol problems or drug abuse within the six months preceding pregnancy. Alcohol and drugs are known to affect hormone levels, and recent use could have affected the results of the test. The women included in the study were mostly married and had an average age of 31.2 years.

The researchers told the women of the purpose of the study and took blood samples about 15, 19, 25, 31, and 37 weeks into the pregnancy. This was to assess levels of  three “stress-related” hormones that have also been studied in non-pregnant depression patients. These hormones were placental corticotrophin-releasing hormone (pCRH), adrenocorticotropic hormone (ACTH) and cortisol.

The researchers assessed depression at two points: during pregnancy and again during the postnatal visit, given nine weeks after birth.

They used a validated scale during pregnancy, a version of the Center for Epidemiological Studies–Depression Scale (CES-D). This scores responses on a four-point scale with participants indicating how often they experienced a depressive symptom during the past week.

At the postnatal visit they used another reliable scale, the Edinburgh Postnatal Depression Scale (EPDS), to assess mothers’ depressive symptoms also using a four-point scale (0 to 3) to record symptoms experienced in the past week.

The researchers also looked at how the results of the hormone test varied when using  blood tests taken at each of the different cut-off points.They used this to calculate the optimum time to take the blood test and the threshold of the hormone they thought best at predicting postnatal depression.

What were the results of the study?

Sixteen women out of 100 in the sample developed symptoms of postnatal depression symptoms. The researchers report that at 25 weeks into pregnancy higher pCRH levels were a strong predictor of these depression symptoms. They say this effect was also significant after controlling for depressive symptoms from before pregnancy. No significant associations were found for the other “stress” hormones, cortisol and ACTH.

Further analysis is reported to show that a pCRH level taken at 25 weeks is a possible diagnostic tool, and the authors express the optimum test’s accuracy in terms of “sensitivity and specificity”:

• Sensitivity is a measure of the probability that a test will correctly identify a person with a condition: in this study the sensitivity of the test was reported as 0.75, meaning the test correctly identified 75% of postnatal depression cases.
• Specificity is a measure of the probability that those who do not have a condition will be correctly identified by a test. The specificity level reported in this test was 0.74, meaning it correctly identified 74% of subjects without the condition.

The researchers say this is moderate discrimination, meaning that about three quarters of women who have a pCRH level above the cut-off (56.86 pg/mL) do develop mild depression and three quarters below this level will not develop depression.

Conversely, about one quarter of women who test negative, below the cut-off, will also develop mild depression (the false negatives) and they and their carers may be falsely reassured by the negative test.

What interpretations did the researchers draw from these results?

The researchers conclude that at the “critical period in mid-pregnancy (25 weeks), pCRH is a sensitive and specific early diagnostic test for PPD symptoms.” They claim that if replicated, these results have implications for the identification and treatment of pregnant women at risk for postnatal depression.

What does the NHS Knowledge Service make of this study?

This is a small cohort study with selected women who all successfully reached the end of their pregnancies (full term). It used self-reported questionnaires to detect depression, rather than a clinical diagnosis.

The authors acknowledge some strengths and limitations to this study:

• It is known that pCRH predicts the length of a baby’s gestation period in the womb. It was important for this study that this potential “confounding” factor, was controlled for.  If women with low levels of the hormone were to have been included in this study and delivered early this could have introduced bias into the study.
• The use of a self-reported questionnaire to define mild depression will have reduced the accuracy of the diagnosis compared to a clinical examination, however the authors say that it is likely that this would have had only a limited affect on the accuracy of the results.
• During their analysis researchers were unable to control for a history of  “lifetime” depression, depression occurring  outside of pregnancy, as this information was not available. It is unclear how this may have affected the results.

Overall this is a study which points the way to further research into the role of this hormone, but the false positive and negative rates implied by the results suggest that it is not yet a suitable test for widespread use.