The perinatal mental health ‘blind spot’: recommendations to improve routine data 


by Sarah Masefield

Routine health data (i.e. data recorded in health records) is valuable for producing estimates of the prevalence of health conditions and exploring variation relating to sociodemographic characteristics (such as socio-economic status) or geography (i.e. differences between one area and another). In the field of perinatal mental health (PMH) it could help us to see if any groups of women, such as from minoritised ethnicities, are less likely to have poor mental health detected/recorded and managed during the period from pregnancy to two years after the birth, known as the perinatal period. Without accurate data these inequalities cannot be addressed and can grow. These data are also important in understanding whether local services have the capacity for the PMH demand in their area at all levels of need (e.g. low-moderate, moderate-severe, critical) and for different PMH conditions (e.g. depression, anxiety, trauma, psychosis). 

Our project, Inequalities in identification and treatment of perinatal mental health, aimed to explore inequalities in identification and treatment of PMH conditions and describe the differences across the West Yorkshire region. As part of this work, we tried to see if the prevalence of PMH conditions could be identified from publicly available routine health data, and whether these datasets could identify variation in prevalence by sociodemographic characteristics. To do this, we looked at the usefulness of publicly available sources of routine PMH data for estimating the prevalence of poor PMH and how this data might vary at the local level in England (e.g. by NHS Trust or clinical commissioning group). 

We reviewed five data sources (hosting four data sets) of publicly available routine data, although one (iViewPlus) was decommissioned during our investigation. None of the data sets included estimates of the prevalence of PMH conditions or adequate data for producing estimates. For example, the Maternity Services Data Set includes the sole PMH data item ‘PMH prediction and detection’. This item indicates whether the PMH screening tool (Whooley) was asked or not, and if previous mental health conditions were identified. It does not record whether a mother has a current PMH condition. The Mental Health Services Data Set (MSDS) provides information on the number of women (aged 16 or over) who received a mental health referral to a secondary care NHS Trust between their pregnancy booking appointment and 12 or 24 months post-pregnancy. However, it only provides information on women who are referred and take up care from secondary care services, so is only indicative of moderate to severe and complex PMH conditions. As such it is likely to underestimate PMH need in the maternal population but could also overestimate it in certain groups that are known to be overrepresented in acute mental health services, such as minority ethnic groups.

Information reviewed in the Public Health England PMH Data Catalogue indicated that there may be plans in place to improve PMH data collection by NHS Digital, such as data items for PMH assessment scales (e.g. the Edinburgh Postnatal Depression Scale, Patient Health Questionnaire (PHQ)-9 and Whooley questions). Likewise, a PMH data item may also be added to the Community Services Data Set for when Health Visitors ask about PMH health. However, we were unable to find information on when these data items might be added and if sharing this data with NHS Digital will be required.

In every data set there was insufficient sociodemographic information to look at potential differences in PMH conditions by geographical area and/or by social determinants of health, such as ethnicity or socioeconomic status. The data sources and technical specifications for the data sets were often hard to find and the supporting information (e.g. for key definitions and data quality) was insufficient. 

Our six recommendations would ensure that accurate area-based estimates of the prevalence of PMH difficulties could be made with publicly availability routine PMH data:

1. Mandate the recording of: a) the outcomes of routine screening and assessments of PMH conditions, and any referrals made for these conditions; and b) inequality characteristics (e.g. ethnicity, socioeconomic status, sexuality etc.) at all stages of the PMH pathway in the maternity and community services routine health record systems.

2. Require the above data to be shared via open access routine datasets (e.g. MSDS, CSDS)).

3. Enhance the interoperability of all PMH NHS data systems (e.g. midwifery, health visiting and general practitioners) so that information on PMH (and other areas of concern/need) is consistently collected across the healthcare pathway and collated into a single data source.

4. Include all perinatal data (including for PMH assessment and sociodemographic information) in one publicly available platform, with the option to extract the data for further analysis.

5. Include prevalence estimates for PMH conditions adjusted for area-based sociodemographic differences and include population reference values to make the data more comparable between geographical areas.

6. Include clear and comprehensive descriptions of data items and statements of data limitations that are easy to find and signposted from within data sources, including information about how the data were collected and how the data items were derived.

For more information, read the paper:  Can we identify the prevalence of perinatal mental health using routinely collected health data?: A review of publicly available perinatal mental health data sources in England. Learning Health Systems, 2023, e10374.

Related Blogs

Securing the future of child health and development: evaluating what works in the early years using a novel interventional family cohort study.

Early childhood provides the building blocks for a child's future, but there isn't enough research evidence about how to best support their parents during this time. In this blog, Kate Mooney describes how the BSB Innovation hub are planning to evaluate a series of early years interventions, providing evidence on how to give every child the best possible start in life."

My day at BiBFest; a day of great food, great music, and great science!

Evan Seaman, Sports Journalism Student at Leeds Trinity University and Social Media Manager at Ilkley Town AFC, details his experiences covering Day 1 of BiBFest 2023, Born In Bradford's science festival. Day 1 (for Bradford's young people) brought something entirely new to the fold, as all activities, talks, stalls, and entertainment were led by young people themselves, giving young people a platform to express what’s most important to them, and share their thoughts on how we can all work together to support health and wellbeing across the district.

The BaBi Network – Building Research Capacity

Sam Debbage, Research Capacity Building Lead for the YH ARC explains how The Born and Bred In (BaBi) Network has increased research capacity in a number of Trusts over the last year, and what this means for Yorkshire and Humber families.