Helping a woman to be as healthy as she can be before pregnancy has benefits for her at the time, during her pregnancy and birth and has positive health effects for her and her baby in the longer-term. Our aim is to carry out studies to better understand how women of reproductive age used their GP services before becoming pregnant, and how good pre-pregnancy health and care affects outcomes for women and their babies and the cost-effectiveness of pre-pregnancy care.
We will use patient data collected by a network of GP practices across England and linked to information about hospital admissions and visits where relevant. This information is recorded routinely by GPs during consultations. The information that identifies each person is then removed, and the de-identified data are made available for research. These data have previously been linked to create mother-baby pairs, so that we can also look at the health of the baby. We will describe women’s characteristics such as age, ethnic group, smoking, and whether they have pre-existing health conditions. We will describe how, why and how often they use their GP practice generally and in the year prior to pregnancy and explore whether we can identify specific pre-pregnancy care or advice using these routine records. Finally, we will look at how the health and care that women received before becoming pregnant, with or without pre-existing conditions, affects their own health during and after pregnancy, the health of their baby, and the economic consequences for the NHS.
The aim of this study is to examine the impact of good pre-pregnancy health and care on outcomes for women and their children.
Stage 1: First, it is necessary to understand the characteristics of women registered with a GP and to explore how they engage with their GP. Women of reproductive age (18-48yrs) will be studied over a 2 year period (2017-2018) prospectively, and their demographic characteristics, lifestyle factors, and pre-existing conditions described. Engagement with GP services will be described in terms of number and type of consultations. Diagnostic, symptom and therapy codes, plus health prevention entity types relating to preventative health advice, pre-pregnancy care, pregnancy advice or fertility concerns will be extracted to explore the feasibility of identifying specific pre-pregnancy care.
Stage 2: Next, we will focus on women who became pregnant, to assess the effects of pre-pregnancy health on selected maternal and infant outcomes. Examples of outcomes for women include pregnancy loss, gestational diabetes, pre-eclampsia, and postpartum mental health conditions, while examples of baby outcomes include gestational age at birth, birthweight, birth injuries and other serious neonatal morbidity. Regression methods will be used to explore the association between sociodemographic and lifestyle factors, pre-existing morbidity, poor pregnancy history and outcomes of interest. We will examine how receipt of pre-pregnancy care in the year prior to pregnancy affects the outcomes. Population attributable fractions will be estimated for different aspects of poor pre-pregnancy health.
Stage 3: Finally, an economic evaluation will be conducted to estimate potential NHS costs and outcome differences between women receiving different pre-pregnancy care. Costs will be estimated from an NHS perspective and will include a thorough account of primary and secondary care in both groups. The outcome measure of the economic evaluation will be subsequent adverse events averted between women receiving and not receiving pre-pregnancy care.
Health Outcomes to be Measured:
This is a staged analysis:
For Stage 1: the analysis is descriptive.
For Stage 2: outcomes that have a substantial impact on maternal and infant health have been selected based on evidence from literature, such as the MBRRACE-UK reports of national confidential enquiries into maternal deaths and morbidity,1-4 and include the three categories as follows. Depending on the quality of the data, we may not be able to analyse some outcomes if the prevalence estimated from our data is considered to be unreliable after being checked against published estimates. Examples of outcomes in each group are provided:
i) Pregnancy outcome: recorded termination of pregnancy, pregnancy loss (miscarriage, stillbirth). Note that CPRD pregnancy register has a proportion of pregnancies classified as unknown outcome, we may need to treat these as a group in their own right.
ii) Maternal outcomes: hypertension in pregnancy, (pre)-eclampsia, gestational diabetes, induction of labour, mode of birth, sphincter injury, duration of postnatal hospital stay, postpartum incontinence, painful intercourse and perineal or pelvic pain, wound problems and infection, postpartum mental health conditions.
iii) Neonatal and infant outcomes
Preterm birth (gestational age at birth <37 weeks), small for gestational age, low birthweight (birthweight <2500 grams), very low birthweight (birthweight <1500 grams), macrosomia (birthweight >4000 grams), birth injuries and fracture including shoulder dystocia, and neonatal resuscitation. We may define a ‘composite’ neonatal/infant outcome for serious morbidity/mortality outcomes with small numbers of events.
For Stage 3: maternal and baby health care resource use and costs associated with different forms of pre-pregnancy care or without pre-pregnancy care will be estimated from an NHS perspective. A cost-effectiveness analysis over the study period will be conducted to demonstrate value for money of pre-pregnancy care expressed as cost per adverse event averted.
CPRD Mother-Baby Link;HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation;Pregnancy Register