Depression is common during the perinatal period (from conception to one year post-partum), while suicide continues to be a leading cause of maternal mortality. Perinatal mental health conditions are more common in particular groups, including younger women, migrant women and women with a history of trauma, including intimate partner violence. A study of 2 million women in England also found increased obstetric (pre-term birth) and neonatal (small for gestational age) risks among women who had had a secondary mental health service contact before pregnancy, with risks higher where contacts were more recent or more intensive (i.e. crisis resolution/home treatment team input or inpatient admission). Women with mental health conditions are known to be at increased risk of relapse or worsening during the perinatal period.
Unique to the National Health Service (NHS) in England, £365 million was invested in 2016 for perinatal mental health services, followed by further funding in 2019. The authors of this recent study set out to determine whether the gradual roll-out of community perinatal mental health teams was associated with increased access to secondary mental healthcare or reduced postnatal relapses.
NHS England invested £365 million in perinatal mental health services in 2016.
Methods
This cohort study analysed data from the NHS England national dataset of secondary mental healthcare. This linked all mental healthcare ‘episodes’ from 01/04/2006 to 31/03/2019 (except 01/12/15 to 31/03/16) to Hospital Episode Statistics (all general hospital contacts) and Personal Demographic Service birth notifications.
The authors examined the records of women aged 18 years and above with a pregnancy starting between 01/04/2016, having a single baby up to 31/03/2018, at 24 weeks’ gestation or more. Women with a pre-existing mental health condition (defined as contact with any secondary mental health service in the 10 years before their current pregnancy) were included in this study.
They then determined whether the Clinical Commissioning Group (CCG) responsible for healthcare in the region where the woman lived did or did not provide a community perinatal mental health team (defined as the presence of at least one dedicated psychiatrist, psychologist, and specialist nurse in post) from the date of her pregnancy onwards.
The authors calculated adjusted odds ratios and 95% confidence intervals using logistic regression, adjusting first for monthly time trends and then for maternal demographic characteristics and regional differences in socio-economic deprivation.
Results
Out of the 780,026 eligible women, 70,323 (9.0%) had a pre-existing mental health condition. Availability of community perinatal mental health teams increased from 81 CCGs (39%) in April 2016 to 130 (63%) in June 2017 (when women giving birth in March 2018 became pregnant). Out of the 70,323 included women, 31,276 (44.5%) lived in a region with a community perinatal mental health team and 39,047 (55.5%) did not.
A smaller proportion of women had an acute postnatal relapse (inpatient admission or crisis resolution/home treatment team) in regions with a community perinatal team than in regions without a team (n=1117, 3.6% vs. n=1,745, 4.5%; aOR=0.77, CI=0.64 to 0.92). There was no statistically significant difference in relapses during pregnancy.
A higher proportion of women received secondary mental healthcare (admission, crisis resolution/home treatment team or community mental health team) during the perinatal period (both during pregnancy and within one year post-partum) in regions with a community perinatal team than in regions without a team (n=9,888, 31.6% vs. 10,033, 25.7%; aOR=1.35, CI=1.23 to 1.49).
The authors also found that a higher proportion of women had a stillbirth or neonatal death in regions with a community perinatal team than in regions without a team (n=165, 0.5% vs. n=151, 0.4%, aOR=1.34, CI=1.09 to 1.66). They found the same pattern for babies born small for gestational age (n=2,777, 7.2% vs. n=2,542, 6.6%, aOR=1.1, CI=1.02 to 1.20). The opposite was true for pre-term birth: a lower proportion of women in regions with a community perinatal team had a premature baby than in regions without a team (n=3,167, 10.1% vs 4,341, 11.1%; aOR=0.86, CI=0.74 to 0.99).
Differences in obstetric and neonatal outcomes were found in women with mental health conditions living in regions with and without community perinatal teams.
Conclusions
As expected, the presence of community perinatal mental health teams was associated with increased access to secondary mental healthcare in the perinatal period. Encouragingly, they were also associated with reduced risk of post-partum relapse (requiring hospital admission or crisis resolution/home treatment team support) and pre-term birth.
Unexpectedly, the authors found higher rates of stillbirth, neonatal death and small for gestational age babies in regions where community perinatal mental health teams were provided, despite controlling for potential confounders. Potential explanations for these unforeseen findings include:
- Focus on perinatal mental health overshadowing recognition of modifiable behavioural and obstetric risk factors by physical healthcare professionals.
- Highlighting mental health conditions could lead to discrimination (diagnostic overshadowing) when women access physical healthcare.
- Increased use of psychotropic medication. However, the authors note that there is no current evidence linking psychotropic medication to stillbirth.
The existence of community perinatal mental health teams was associated with higher access to secondary mental healthcare in the perinatal period.
Strengths and limitations
- Due to substantial missing data, the authors did not identify women with pre-existing mental health conditions using clinically recorded diagnoses. They used mental health service contacts as a proxy, increasing the number of women who could be included in their analyses.
- Using regional provision of community perinatal teams avoided confounding by clinical indication, but may have reduced the estimated effect size (because not all women accessed the team).
- Because the authors could not access adolescent mental health records, fewer than 10 years of psychiatric history could be captured for younger women, who may be at greater risk of perinatal mental ill-health.
- The authors are conducting a realist evaluation that will explore the mechanisms of women’s engagement with community perinatal teams as well as changes to patterns of service use and costs over time, which are likely to illuminate some of these findings.
It would be helpful to capture history of adolescent mental health to identify women at risk of perinatal mental health difficulties.
Implications for practice
Clinicians and policy makers can be encouraged that provision of community perinatal mental health teams is associated with increased mental health service access and reduced post-partum relapse, as well as reductions in pre-term birth. However, higher rates of stillbirth and neonatal death in regions where such teams are provided show that investment in mental healthcare alone cannot be assumed to influence the pregnancy risks known to be higher in women with mental health conditions. Clinicians in psychiatry, obstetrics and general practice must be attentive to the risk of diagnostic overshadowing and work closely together to provide joined up care at all stages of the perinatal period.
Professionals need to work in a multidisciplinary capacity and provide high quality of care during the perinatal period to prevent pregnancy risks.
Statement of interests
My PhD second supervisor was Professor Louise Howard (one of the authors), but I had no involvement with this study.
