Maternal perinatal mental illnesses and adverse pregnancy outcomes: population-based studies using data from United Kingdom primary care

Background: Perinatal mental illness, especially depression, is a leading cause of maternal morbidity and mortality in high-income countries. In the United Kingdom (UK), mental illness commonly presents to and is treated at primary care level; however there are no up-to-date estimates of the burden of different mental illnesses in women in and around pregnancy. The potential impact of mental illness with or without psychotropic medication on the risk of non-live pregnancy outcomes is unclear. In this context, the safety of psychotropic drugs, especially antidepressants, remains controversial. Aim and objectives: To estimate the clinical burden of depression, anxiety and serious mental illness (defined as bipolar disorder, schizophrenia and other related psychotic disorders) presenting to and/or being treated in UK primary care, and to investigate the effects on pregnancy outcomes while trying to differentiate the effects of psychotropic medication from mental illness itself. Methods: Women aged 15-45 years from 1990 to 2009 were identified from The Health Improvement Network, a UK primary care database. Coding of mental illness diagnoses and psychotropic drug prescriptions were examined by separately assessing the proportions of women with recordings of diagnoses, symptoms, and drug prescriptions over the study period. Three separate studies were then carried out. A cross-sectional study was firstly conducted to estimate the prevalence and diagnostic overlap of mental illnesses before, during and after pregnancy and the variation by maternal age, socioeconomic status and other maternal factors. The second study examined the risks of non-live pregnancy outcomes (defined as perinatal death, miscarriage, and termination) in women with no history of depression and anxiety, a diagnosis of such illness prior to pregnancy, illness during pregnancy or illness during pregnancy with use of medication (stratified by medication type). Multinomial logistic regression models were used to compare risks of non-live outcomes across these groups, adjusting for important socio-demographic and lifestyle characteristics. The third study examined the risks of major and system-specific congenital anomalies in children born to women with depression or anxiety that was untreated or treated with psychotropic medication. Logistic regression with a generalised estimating equation was used to compare risks of major congenital anomalies in children exposed and unexposed to psychotropic medication during the first trimester of pregnancy, adjusting for important socio-demographic, lifestyle and chronic comorbidity in the mother. Results: There were 344,042 women who had one or more singleton pregnancies identified between age 15 and 45 from 1990 to 2009. Recording of mental illness and prescriptions of psychotropic drugs increased considerably over the study period. There was high prevalence and overlap of different maternal mental illnesses, especially depression and anxiety, during and after pregnancy, and the prevalence was generally highest in younger, socioeconomically deprived women who had smoked before childbirth, were outside the normal range of BMI and had other chronic medical conditions, such as diabetes. Socioeconomic deprivation was associated with increased risk of all mental illnesses, although the impact of deprivation was more marked in older women. Those aged 35-45 in the most deprived group had 2.63 times the odds of antenatal depression (95% confidence interval [CI] 2.22-3.13) compared with the least deprived; in women aged 15-25 the increased odds associated with deprivation was more modest (odds ratio [OR]=1.35, 95%CI 1.07-1.70). Similar patterns were found for anxiety and serious mental illness. Women with antenatal exposure to antidepressant or anti-anxiety drugs showed the greatest increased risks for non-live pregnancy outcomes, relative to those with no history of depression or anxiety, although women with prior (but currently un-medicated) illness also showed modest increased risks. Compared with un-medicated antenatal morbidity, there was weak evidence of an excess risk in women taking tricyclic antidepressants (TCAs), and stronger evidence for other medications. The absolute risks of major and system-specific congenital anomalies were small in the general population (269 per 10,000 children for major congenital anomalies). Compared with un-medicated antenatal depression or anxiety (278 per 10,000 children for major congenital anomalies), the use of antidepressants during early pregnancy was associated with excess risks, especially for selective serotonin reuptake inhibitors (SSRIs) (290 per 10,000 children for major congenital anomalies). Compared with children born to women with no depression or anxiety, there was an increased risk of heart anomalies in children with antenatal exposure to SSRIs (adjusted OR=1.25, 95% 95%CI 1.02-1.53), particularly in those exposed to paroxetine (adjusted OR=1.89, 95%CI 1.24-2.88). Children exposed to sertraline and escitalopram also had similar increased risks, although fewer women were exposed to these drugs. No increased risks of major congenital anomalies were found in children exposed to TCAs or benzodiazepines; however, the risks of right ventricular outflow tract anomalies were notably higher for all drug classes. Conclusion: Strong socioeconomic inequalities in perinatal mental illnesses occur and persist with increasing maternal age. Women with depression or anxiety have higher risks of miscarriage, perinatal death and therapeutic terminations than women without these diagnoses and the risks are even higher if prescribed psychotropic medication during early pregnancy than if not. There is also an increased risk of congenital heart anomalies in children exposed to paroxetine and other SSRIs during the first trimester compared with those who are unexposed, although the absolute risk is small. There could be other associated factors also related to depression, anxiety or use of medications, which yet unlikely fully explain the observed excess risks. Whilst medicated depression or anxiety could be a marker of more severe illness than un-medicated ones, my findings indicate there may be some specific drug effects Targeting detection and effective interventions to women at risk of mental illness during pregnancy may reduce inequity and avoid substantial psychiatric morbidity, and subsequently reduce the need for further psychotropic treatment. GPs and other health care professionals should take a cautious approach when managing mental illness in pregnant women. The findings in this thesis provide vital information for this purpose, namely helping communicate the magnitude of risk of major congenital anomalies to women with the use of different psychotropic drugs in the context of the baseline risk in the general population