'Preventable' stillbirths and social inequality

There are “twice as many babies stillborn in most deprived 10% of England”, the Guardian has reported. The paper describes research into this “intractable problem” has found that 900 babies' lives could be saved every year if the rate of stillbirths in the poorest areas of England was as low as the rate in the most affluent areas.

This study did not look at data on individuals but analysed stillbirths by geographical areas (approximately 1,500 residents in each) between 2000 and 2007. Overall, the rate of stillbirths was low, with 44 stillbirths per 10,000 single births and no evidence of a change in rates over the course of this period. The authors called the difference in rates of stillbirth across socioeconomic groups “a deprivation gap”. The main contributor to the higher rate was stillbirth due to bleeding from the placenta before labour.

This research points to an important approach for looking at inequalities between and among geographical groups, which will allow planners to focus their attention on populations in need.

Where did the story come from?

The study was carried out by researchers from the Universities of Leicester, Cambridge and London and was funded by the UK’s National Patient Safety Agency. The study was published in the peer-reviewed online medical journal BMJ Open.

What kind of research was this?

The researchers set out to assess the cause-specific stillbirth rates over time and to highlight any differences between small geographic areas in England with different socioeconomic status. They designed a population-based retrospective study measuring deprivation using the UK “index of
multiple deprivation”. From this, they reported the relative deprivation gap (comparing the most and least deprived tenths) in rates of stillbirths, looking at both overall and cause-specific deaths.

This is an appropriate method to look at this sort of question but relies on the average level of deprivation in a geographical area of 1,500 people (slightly bigger than a whole postcode) rather than from individuals

What did the research involve?

This population-based study analysed stillbirths by geographical area between 2000 and 2007.

Researchers analysed information from the Centre for Maternal and Child Enquiries on all single births (not births of twins or multiple babies) born to mothers living in England from 2000 to 2007. Twins and multiples were not examined; this may have been because the numbers were lower and had greater risk of complications and stillbirth, which may skew the overall results. For more information on risk in twins and multiples, see the recent news item Twins 'more likely to die before first birthday'.

Information analysed by the researchers included:

• cause of death
• gestational age
• “super output area” of where the mother lived (defined by the authors as geographical areas with about 1,500 residents)

Causes of death were categorised into nine areas:

• congenital anomalies (defects apparent at birth)
• pre-eclampsia
• antepartum haemorrhage (bleeding before the birth – for example with placenta praevia)
• mechanical event such as cord prolapse, breech presentation or oblique presentation
• maternal disorder such as an infection or maternal hypertension
• miscellaneous, such as infection of the baby
• unexplained and baby small for gestational age (birth weight in the bottom 10%)
• unexplained but baby not small for gestational age
• unclassifiable, including missing data

Socioeconomic differences were measured for each area using an “index of multiple deprivation” score. This measure considered factors relating to:

• income
• employment
• health and disability
• education skills and training
• barriers to housing
• living environment
• crime

All areas were divided into 10 groups with equal numbers, ranging from the least deprived tenth to the most deprived tenth.

Statistical models were used to estimate the relative deprivation gap in rates of stillbirths overall and specific to the cause of death. “Excess mortality” (how many deaths could be prevented by everyone living in ideal conditions) was calculated by applying the rates seen in the least deprived tenth to the entire population at risk.

What were the basic results?

The researchers identified that there were 44 stillbirths per 100,000 and no evidence of change in this rate over the eight-year study period. The key finding of this study was that rates of stillbirth were twice as high in the most deprived tenth compared with the least deprived tenth (rate ratio 2.1, 95% confidence interval 2.0 to 2.2). They found that this rate did not change over the eight-year period for any specific cause of stillbirth.

The widest gap was seen in stillbirths caused by bleeding from the placenta before labour (rate ratio 3.1, 95% confidence interval 2.8 to 3.5). There was a significant deprivation gap for all specific causes of stillbirths except for those caused by mechanical events (rate ratio 1.2, 95% confidence interval 0.9 to 1.5).

How did the researchers interpret the results?

The researchers said that a wide deprivation gap exists in stillbirth rates for most causes and is not diminishing.

It is particularly interesting that unexplained stillbirths accounted for 50% of the deprivation gap. This suggests that a better understanding of the causes of these stillbirths, and why they are linked to deprivation, could lead to taking appropriate action to reduce the stillbirth rate.

Conclusion

This is a well-conducted and well-reported eight-year study. It provides a numerical estimate for the deprivation gap that exists in outcomes of pregnancy care. The authors’ suggestion that more detailed information should be collected is sensible. This could help to identify preventable, avoidable or modifiable risk factors and ways to prevent, avoid or modify them.

Having said that, there are some small limitations to the study acknowledged by the authors, including:

• As individual-level information was not available, the researchers were unable to adjust for factors such as smoking status, which is known to be linked to stillbirth rates.

• Recording cause of death for stillbirths can be complex. There is no consensus on which is the best among the 35 published classifications. Some of these rely on advanced genetic and imaging techniques to identify a cause, which may not be available in all locations or circumstances.

• A two-fold increase in stillbirth rates, at around 44 per 100,000 population, may seem small. However, when extrapolated to the population of England this may be important. The researchers said that if the stillbirth rates seen in the least deprived areas were seen throughout the population, the number of stillbirths in England would decrease by one third, or nearly 900 fewer each year.

Studying routine collected data in this way allows health policymakers to monitor health trends - in this case stillbirths. By highlighting causes of death, particularly those like antepartum haemorrhage that might be preventable, this study will help policy-makers to focus their action.