'Premature babies study shows survival rates on rise' is the headline in The Guardian, which is one of many sources reporting the news that survival rates for babies born between 22 and 25 weeks have risen overall since 1995.
This is based on a reliable piece of research that looked at the survival rates and ongoing illnesses or complications affecting babies born extremely prematurely (between 22 and 26 weeks of pregnancy) in England in 2006. The researchers compared these rates with those of equivalent babies born in 1995.
Their main finding was that when comparing survival-to-discharge rates (meaning babies were eventually thought to be well enough to leave hospital) between 1995 and 2006, there was an increase from 40% in 1995 to 53% in 2006.
However, there was no difference in the level of ongoing illnesses or complications affecting these surviving babies, including continued respiratory problems, brain damage and eye disease of prematurity (retinopathy).
Overall, the finding that survival of extremely premature babies has increased, but the proportion of survivors with major health complications has not changed, calls into question the issue of the level of ongoing care and support that extremely premature survivors may require.
This question cannot be answered further by this study alone, as it would need to track the health of the babies involved into childhood and adulthood.
The study was carried out by researchers from Queen Mary University of London, UCL Elizabeth Garrett Anderson Institute for Women's Health and the University of Leicester, and was funded by the Medical Research Council.
The study was published in the peer-reviewed British Medical Journal.
In general, the media reporting represented the research fairly. The Independent raised the question of whether the results of this study add evidence to the debate about any changes to the legal limit for termination of pregnancy (abortion).
This study did not address this question, but that is unlikely to stop the study's findings being used as part of the debate on this issue. The current legal limit for termination is set at 24 weeks of pregnancy.
This was a review of prospective data collected from maternity and neonatal units in England in 1995 and again in 2006.
The review aimed to look at whether the survival and health status of extremely premature babies born between 22 and 25 weeks of pregnancy had changed over this period.
Premature birth (before 37 completed weeks of pregnancy) is known to be associated with increased risk of neonatal death, respiratory problems, cerebral palsy and other neurological problems, as well as the risk of longer term developmental problems.
The more prematurely a baby is born, the higher the risk of complications. Babies born 'extremely premature' (between 22 and 26 weeks of pregnancy) have the highest risk of complications.
Since 1995 medical developments, such as giving the mother steroids to prepare the baby's lungs for premature birth, were expected to have reduced the risk of the premature newborn experiencing complications.
This study aimed to investigate whether these developments have actually reduced the risk of premature infants experiencing medical problems, and if this has increased overall survival rates.
This study used data from two prospective cohort studies: EPICure and EPICure 2. For 10 months in 1995 the first EPICure study collected data on all babies born in the UK and Ireland before 26 weeks of pregnancy (up to 25 weeks and 6 days). Outcomes for surviving children were reported up to the age of 11.
In 2006, EPICure 2 collected similar data for extremely premature babies born England, but slightly extended the cut-off point to babies born in England up to and including 26 weeks (up to 26 weeks and 6 days).
The researchers looked at health outcomes until the time of hospital discharge for babies born in 2006 and compared these with babies born in 1995.
The main health outcomes of interest were survival to the time of hospital discharge, as well as illnesses or complications affecting the premature baby.
Illnesses and complications the researchers were interested in were those known to affect premature babies, including:
Confirmation of the number of weeks of pregnancy in the 1995 study was only available for babies who had been admitted to intensive care. In order to directly compare the two years, the researchers restricted their comparison to babies who in 2006 were admitted to intensive care and had also been born between 22 and 25 weeks, instead of using data from the broader cut-off in 2006, which included babies born at 26 weeks.
EPICure 2 only looked at extremely premature births in England and therefore the researchers also only compared the subset of 1995 babies born in England (excluding the babies born in Ireland).
Looking at complete data for 2006, 3,133 births were confirmed to be between 22 and 26 weeks of pregnancy. The proportion of these babies who were alive at the start of labour ranged from 57% of the babies born at 22 weeks to 81% of the babies born at 26 weeks.
Overall, one-third of these 3,133 babies survived to the time of hospital discharge, with survival rates increasing with the age of the baby:
When looking at illnesses in surviving premature babies in 2006, 68% (705) of survivors had lung immaturity complications and still needed to be on oxygen at 36 weeks, 13% (135) had serious brain abnormalities on ultrasound, and 16% (166) were treated for retinopathy of prematurity.
In order to directly compare to the babies born in 1995, they only looked at 1,115 babies in 2006 who were born between 22 and 25 weeks who had also been admitted to intensive care (ICU). In 1995, 666 babies were born in England between 22 and 25 weeks and admitted to intensive care.
The overall survival to time of discharge was 40% of the 666 ICU babies born in 1995, which increased significantly to 53% of the 1,115 ICU babies born in England in 2006. This corresponded with significant increases in survival rates for each newborn age:
When comparing illnesses in surviving babies between 1995 and 2006, however, there was no difference in the proportion of babies surviving with lung immaturity complications requiring continued oxygen support at 36 weeks. There was also no increase in the proportion of premature babies with serious brain abnormalities on ultrasound. However, there was an increase in the proportion of babies in 2006 who were being treated for retinopathy of prematurity.
Factors associated with risk of newborn death or serious illness were similar in both 1995 and 2006. The most significant factor was that the more prematurely a baby is born, the higher the risk of death or serious complications.
The researchers conclude that the overall survival of babies born between 22 and 25 weeks of pregnancy has increased since 1995, but patterns of illness in the premature newborn have not changed.
They concluded from this that there may be an important increase in the number of extremely premature survivors at risk of health problems in later child and adult life.
This is valuable research that has used reliable maternity and neonatal hospital data to look at the survival rates and ongoing illnesses or complications affecting babies born extremely prematurely, between 22 and 26 weeks of pregnancy.
In 2006, one-third of babies born in England between 22 and 26 weeks survived to the time of hospital discharge. This ranged from 2% of babies born at 22 weeks, increasing to 77% of babies born at 26 weeks.
Comparing this with similar data from 1995 (which required restricting the sample to the babies born between 22 and 25 weeks and admitted to ICU), there was an overall increase in survival-to-discharge rates from 40% in 1995 to 53% in 2006.
However, there was no difference in the ongoing illnesses or complications affecting these surviving babies, including continued respiratory problems, brain damage and eye disease of prematurity.
From this, the researchers suggest that the increased survival rates may correspond with an increased number of extremely premature survivors who have ongoing health problems persisting into later childhood and maturity. This may include an increased need for medical care and help from allied services, such as social care or educational support.
These seem reasonable conclusions but they cannot be further assessed by this study, which has not followed up the premature birth survivors into later life.
The study data does have some limitations, including that for the purposes of comparison between 1995 and 2006 premature birth data, it was not able to look at full datasets for all extremely premature infants born in these years, only a subset of those admitted to intensive care.
The research also does not appear to have had comparative data on the full range of complications that can affect premature infants, including jaundice, anaemia and heart problems.
Overall, this is a valuable study that serves to highlight the level of ongoing care and support that extremely premature babies who survive may require.