How birth statistics are compiled

There are three main routes through which data about births are collected. These have been described in considerable detail elsewhere ( ) but a brief description and update are given here.

The most frequently used source of data on a national scale is civil registration. In the UK parents are required by law to register a birth with the local Registrar of Births, Marriages and Deaths. As well as issuing a certificate, the registrar passes the information to a central office, which maintains records and, in the past, compiled both national and local statistics. Scotland, Northern Ireland and the Republic of Ireland each have separate General Register Offices. In 1970, the General Register Office for England and Wales was merged with the Government Social Survey to form the Office of Population Censuses and Surveys (OPCS). Then, in April 1996, OPCS merged with the Central Statistical Office and the Labour Market Statistics Group of the former Department of Employment to form the Office for National Statistics, which compiles and publishes a wide range of health, social and economic statistics. In undertook a major review of civil registration and proposed wide-ranging changes, including ‘through-life records’, but these did not get through Parliament ( ; ). On 1 April 2008, the General Register Office for England and Wales was transferred into the Home Office’s Identity and Passport Service, but the Office for National Statistics continues to analyse and publish data derived from civil registration. The General Register Office for Scotland merged with the National Archives of Scotland on 1 April 2011 to become the National Records of Scotland. In Northern Ireland, the General Register Office is part of the Northern Ireland Statistics and Research Agency, which analyses and publishes the data and is itself part of the Northern Ireland Department of Finance and Personnel.

In England, Wales and Scotland, the law originally required all fetal deaths after 28 completed weeks of gestation to be registered as stillbirths. This limit was lowered to 24 weeks on 1 October 1992. All live births at any gestation have to be registered. In Northern Ireland, there is no legal requirement to register a stillbirth, but a system was set up in 1961 to enable next of kin to register them. In the Republic of Ireland there was no system for registration of stillbirths before 1995, although they have been notified to Directors of Community Care since 1957.

The second method of information collection is through birth notification. In the UK all births have to be notified to the local Director of Public Health under a system introduced in 1907 and made compulsory in 1915. This is usually done by midwives immediately after the birth, and must be done within 36 hours. The system was originally devised at an era when most births were at home, so that a health visitor could be informed and then call to see the mother and baby. From the 1960s onwards, local and health authorities developed child health computer systems on which babies’ records were initiated by the birth notification and used to administer vaccination and immunisation programmes and to monitor developmental testing.

Since the introduction in England and Wales of the allocation of National Health Service (NHS) numbers for babies at birth in October 2002, the data flows have changed. As the minimum dataset associated with NHS Numbers for Babies is very limited, many maternity units continue to send the fuller dataset used previously in parallel. At the time of writing the NHS Numbers for Babies service is being moved to the NHS Spine, where it will form part of the Personal Demographics Service. In Wales, child health systems have been redeveloped so that data from birth notifications can be aggregated nationally in the National Community Child Health Database and used to produce national birth and perinatal statistics, amongst other purposes.

Despite its limitations, the NHS Numbers for Babies dataset contains key data items, notably gestational age, baby’s ethnicity, time of day of birth and birth order of stillbirths as well as live births within multiple births, which are not recorded at birth registration. Following a successful pilot project, linking NHS Numbers for Babies and birth registration data for 2005, these two datasets are now linked routinely by the Office for National Statistics, enhancing the scope and range of data available at a national level ( ).

In Northern Ireland, each of the four Health and Social Services Boards had its own child health system. Perinatal data from these were pooled to produce data for the province as a whole, held on the Northern Ireland NHS intranet. These data are now held by the Health and Social Care Board, which superseded the four boards on 1 April 2009. In the Republic of Ireland, a subset of vital statistics data derived from the four-part birth notifications, along with data from birth and death registrations, is analysed and published centrally by the Central Statistics Office. More detailed perinatal statistics, including clinical and sociodemographic data from the third part of the form, are analysed through the National Perinatal Reporting System and published separately by the Economic and Social Research Institute.

The third route for collecting data about births is through hospital-based systems. Traditionally these have collected data at discharge about hospital inpatient stays. More recently, systems have been developed that gather data about a person’s episodes of care within a given NHS trust. NHS commissioners should hold information about care given to their residents wherever this is provided. The ways in which this is done are changing rapidly with the development of information technology systems within the NHS.

In England, information about inpatient stays in NHS hospitals is aggregated nationally through the Hospital Episode Statistics (HES) system. There is a separate Maternity HES system to collect information about women delivering in, and babies born in, the maternity departments of NHS hospitals. It should also include babies born at home or in private hospitals, but many of these records are missing. Maternity HES records include the standard admitted patient record plus a ‘maternity tail’, with a minimum dataset and clinical options. The items in the minimum datasets were specified by the , chaired by Edith Körner. This was known as the Körner Committee and the datasets it recommended are usually referred to as Körner minimum datasets. The clinical options were set out in publications but were never implemented at a national level.

The HES system started in April 1987 and Maternity HES finally got under way in September 1988, but in the mid-1990s it was still very incomplete. By the financial year 1994–95 the system contained maternity tail records for only 67% of deliveries in England ( ) and 72% in 2002–3 ( ). By 2009–10, after several years of improvement, data were submitted for between 80% and 90% of the items in the maternity tail. Data from the system are published annually on the Information Centre for Health and Social Care’s website ( ). Data are still missing for some units, either because they do not have a computer system in their maternity unit or because maternity systems are not linked to other systems in the hospital so the data in them do not reach Maternity HES ( ). Major changes in data collection have taken place in recent years, with the implementation of the National Care Records Service for individual patient records and of its Secondary Uses Service ( ).

The NHS Information Centre has developed an ambitious new maternity secondary uses dataset for England, with over 1000 data items. It was planned to implement it in stages, with the first starting on 1 April 2011, but the start was delayed by funding decisions ( , ).

In Wales and Northern Ireland, systems similar to Maternity HES were introduced but very few delivery records have data in the maternity tail. Analyses of data about method of delivery and length of stay have been derived from the Patient Episode Database Wales and published in bulletins on paper and on the Statistics Wales website ( ). As described above, the dearth of data from maternity tails has led to decisions to use child health systems for collecting more detailed maternity data in Wales.

In England data about episodes of care in neonatal intensive care units are collected, along with data about other episodes in paediatric departments, in the main part of HES. Unfortunately, these data are not routinely linked, at national level at least, to the record of the baby’s delivery, but the allocation of NHS numbers to babies at birth may make this possible in the future. There is also a lack of consistency in recording levels of special and intensive care in HES and these data are not published routinely. To address this problem a Neonatal Critical Care Minimum Dataset was defined ( ). This is described in Chapter 2 .

Scotland has had a maternity information system working nationally since the mid-1970s and data about mothers have been collected through the SMR2 Maternity Discharge Sheet. In the past, data about babies were collected through the SMR11 record, which from 1996 covered only sick babies, but this has been superseded by the Scottish Birth Record, described below. Data from SMR2 are published annually in Scottish Health Statistics , now published electronically on the Information and Statistics Division Scotland website ( ). Data from SMR2 are also combined with those from the Scottish Perinatal and Infant Mortality Survey and published annually on the Information and Statistics Division website in the Scottish Perinatal and Infant Mortality and Morbidity Report ( ). The report has been administered by a number of different bodies over time. Since 2008, this has been the Reproductive Health Programme of NHS Quality Improvement Scotland, with oversight by the Scottish Perinatal Mortality and Morbidity Review Advisory Group. The report also contains data about the incidence of selected congenital anomalies. The website also has reports on trends over time on topics such as multiple birth and operative vaginal deliveries ( ).

In 2003 Scotland started to implement a completely new system, the Scottish Child Health Information Development project. The first step in this was to implement a web-based Scottish Birth Record and an electronic woman-held record with the later aim of developing links with other systems. The Scottish Birth Record system is a live database, with data entered directly by NHS staff, interfacing with hospital systems where feasible, and available to appropriate NHS staff and to parents.

The other three countries of the UK include data about neonatal care in statistics collected about activity in paediatric departments, but have not so far been able to identify them separately in published data. This gap has led to the implementation of dedicated neonatal systems, linked to the establishment of neonatal networks. The information from them is used in the neonatal audits described in Chapter 2 . The Neonatal Data Analysis Unit was set up to oversee these systems and analyse the data.

Trends and variations in birth rates

Definitions

• •
  

  General fertility rate: the number of live births per 1000 women aged 15–44 years living in the same area.

  
  
• •
  

  Age-specific fertility rate: the number of live births to mothers of each age group per 1000 women in the age group in the same population.

  
  
• •
  

  Total fertility rate: the average number of live children that a group of women would have if they experienced the age-specific fertility rates of the calendar year in question throughout their childbearing lives.

The numbers of live births registered in recent years in each of the four countries of the UK and in the Republic of Ireland are shown in Table 1.1 . This shows that, in the late 1980s, the numbers of births rose everywhere except in the Republic of Ireland, before falling again in the early 1990s. After a slight increase in the mid-1990s followed by a further decline, numbers started to rise in 2002 and 2003. This rise continued until 2008, with a slight falling off in 2009, except in the Republic of Ireland. These figures are useful as a measure of the workload of the maternity and paediatric services but shed very little light on the reasons for the increases and decreases. Fluctuations can arise either as a result of changes in the size and age structure of the childbearing population or as a consequence of changes in the birth rate within each age group.

Table 1.1

Live births in England, Wales, Scotland and Ireland 1975–2010

YEAR	ENGLAND	WALES	ENGLAND AND WALES *	SCOTLAND	NORTHERN IRELAND †	REPUBLIC OF IRELAND
1975	568 900	38 030	603 445	67 943	26 130	67 178
1980	618 371	37 357	656 234	68 892	28 582	74 064
1985	619 301	36 771	656 417	66 676	27 427	62 388
1990	666 920	38 866	706 140	65 973	26 251	53 044
1995	613 257	34 477	648 138	60 051	23 693	48 787
1996	614 188	34 894	649 489	59 296	24 382	50 655
1997	608 202	34 520	643 095	59 440	24 087	52 775
1998	602 111	33 438	635 901	57 319	23 668	53 969
1999	589 468	32 111	621 872	55 147	22 957	53 924
2000	572 826	31 304	604 441	53 076	21 512	54 789
2001	563 744	30 616	594 634	52 527	21 962	57 854
2002	565 709	30 205	596 122	51 270	21 385	60 503
2003	589 851	31 400	621 469	52 432	21 648	61 529
2004	607 184	32 325	639 721	53 957	22 318	61 972
2005	613 028	32 593	645 835	54 386	22 328	61 372
2006	635 748	33 628	669 601	55 690	23 272	65 425
2007	655 357	34 414	690 013	57 781	24 451	71 389
2008	672 809	35 650	708 711	60 041	25 631	73 996
2009	671 058	34 937	706 248	59 046	24 910	74 728 §
2010	687 007	35 952	723 165	58 791	25 315	73 724 §

 

(Data from Office for National Statistics, General Register Office, Scotland, Northern Ireland Statistics and Research Agency, Northern Ireland and Central Statistics Office, Ireland.)

* Including births in England and Wales to women normally resident outside England and Wales.

† Live birth figures from 1981 are resident births only

§ Births for Ireland for 2009 and 2010 are year of registration figures and are provisional.

One of the most long-standing measures of birth rate is the general fertility rate. In this the number of live births is expressed as a rate per 1000 women aged 15–44 or, in some cases, 15–49 years. Figure 1.1 shows the general fertility rate for England and Wales since 1838, the first full year after civil registration began in July 1837. The rates for the mid-19th century are probably an underestimate, as birth registration did not become compulsory in England and Wales until 1874. Shortly after this the fertility rate began to decline, a trend that continued steadily until the 1930s. This was interrupted only by a trough during the First World War and a short-lived peak after the war ended. A similar peak followed the Second World War. After this there was a longer term rise in the 1950s and 1960s, followed by a decline through most of the 1970s. After the rate reached a minimum in 1977 it fluctuated, gradually increasing in the late 1980s and decreasing through the 1990s. After rising from 2003 to 2008, it fell slightly in 2009 but rose again in 2010, except for women aged under 20.

General fertility rate, England and Wales, 1838–2010.

(Data from Office for National Statistics, Birth Statistics.)

This overall rate masks changes since 1977 within age groups. Rates for England and Wales are set out in Table 1.2 , which shows birth rates among women in their late teens and 20s rising slightly in the late 1980s as the ‘bulge’ of women born in the mid-1960s entered the childbearing age range ( ) and then falling through the 1990s and the early years of the 21st century. In contrast, rates among women in their early 30s rose before levelling off in the mid-1990s, while rates for women in their late 30s and 40s have risen consistently. Since 2003, rates have risen in all age groups except for women aged under 20. They declined in 2009 for all groups aged under 30 while in 2010, they rose for all age groups except women aged under 20.

Table 1.2

Age-specific fertility rates, England and Wales, 1964–2010

YEAR	LIVE BIRTHS	LIVE BIRTHS PER 1000 WOMEN IN AGE GROUP								TOTAL FERTILITY RATE
		15–44	Under 20	20–24	25–29	30–34	35–39	40–44	45 and over
1964 (max.)	875 972	92.9	42.5	181.6	187.3	107.7	49.8	13.0	0.9	2.9
1977 (min.)	569 259	58.1	29.4	103.7	117.5	58.6	18.2	4.1	0.3	1.7
1979	638 028	63.3	30.3	111.3	131.2	69.0	21.3	4.3	0.4	1.8
1985	656 417	61.0	29.5	94.5	127.6	76.4	24.1	4.6	0.4	1.8
1990	706 140	64.3	33.3	91.7	122.4	87.3	31.2	5.0	0.3	1.84
1995	648 138	60.5	28.5	76.4	108.7	88.2	36.4	6.5	0.3	1.72
2000	604 441	55.9	29.3	70.0	94.3	87.9	41.4	8.0	0.4	1.72
2001	594 634	54.7	28.0	69.0	91.7	88.0	41.5	8.4	0.5	1.70
2002	596 122	54.7	27.1	69.1	91.5	89.9	43.0	8.6	0.5	1.65
2003	621 469	56.8	26.9	71.3	96.0	95.0	46.4	9.3	0.5	1.73
2004	639 721	58.2	26.8	72.7	97.7	99.7	48.8	9.9	0.5	1.78
2005	645 835	58.3	26.3	71.5	97.9	100.8	50.3	10.3	0.6	1.79
2006	669 601	60.2	26.6	73.0	100.7	104.9	53.9	10.8	0.6	1.86
2007	690 013	62.0	26.0	73.3	103.8	110.2	56.9	11.4	0.7	1.91
2008	708 711	63.8	26.0	74.4	108.1	113.1	58.4	11.9	0.7	1.97
2009	706 248	63.7	25.3	73.9	107.2	113.1	59.0	12.1	0.8	1.96
2010	723 165	65.4	24.2	74	108.0	117.8	61.5	12.5	0.9	2.00

 

The rates for women of all ages, under 20, and 40 and over are based upon the female population aged 15–44, 15–19 and 40–44 years, respectively.

(Data from Office for National Statistics Birth Statistics).

These age-specific rates can be summed up in a statistic called the ‘total fertility rate’. This is a standardised measure that gives the total number of children who would be born to each woman if she experienced the age-specific fertility rates for the year in question throughout her childbearing life. As Table 1.2 shows, the rate for England and Wales rose gradually in the latter half of the 1980s, before falling gradually again, followed by a rise from 2003 to 2008 which tailed off in 2009. Age-specific fertility rates for women aged 30 and over continued to rise in 2009, while those for women aged under 30 fell slightly. Figure 1.2 shows trends over time in total fertility rates in England and Wales, Scotland and the two parts of Ireland, with Scotland having the lowest rates. Rates for local areas, published in national statistical publications for Scotland and Northern Ireland and in the Office for National Statistics Vital Statistics (VS) tables for England, vary considerably between regions and areas. Trends for each local authority area in England can also be seen in interactive maps on the Office for National Statistics website (Office for National Statistics 2011a).

Total fertility rates, UK and Ireland, 1938–2010.

(Data from Office for National Statistics, General Register Office Scotland, Northern Ireland Statistics and Research Agency and Central Statistics Office, Ireland.)

For planning services it would be useful to have some idea of future trends in births, but these are notoriously difficult to predict. Nevertheless, government statisticians attempt to make such projections, combining analyses of past trends with replies to surveys about people’s intentions to have children. Population projections for the countries of the UK, formerly produced by the Government Actuary’s Department, are now produced by the Office for National Statistics. The 2008-based projections, published in 2010, assumed a peak in fertility in 2008, followed by a fall in 2009 and succeeding years and then a levelling off ( ).

Definitions

• •
  

  Birthweight: the first weight of the newborn or fetus obtained after birth. The actual weight should be recorded to the degree of accuracy to which it is measured.

  
  
• •
  

  Low birthweight: less than 2500 g (up to and including 2499 g).

  
  
• •
  

  Very low birthweight: less than 1500 g (up to and including 1499 g).

  
  
• •
  

  Extremely low birthweight: less than 1000 g (up to and including 999 g).

  
  
• •
  

  Gestational age: the duration of gestation is measured from the first day of the last normal menstrual period. Gestational age is expressed in completed days or completed weeks.

  
  
• •
  

  Preterm: less than 37 completed weeks or 259 completed days of gestation.

In England and Wales, birthweight data have been collected since the mid-1950s through the birth notification system. From 1953 to 1973, each local authority, and from 1974 to 1986 each health authority, submitted a form to central government giving the numbers of low-weight births to women living in their area. Data from this source have been used for these years in Figure 1.3 , which shows that the percentage of liveborn babies weighing 2500 g and less, the original definition of ‘low birthweight’, remained at a similar level of between 6% and 7% from the mid-1950s to the mid-1980s.

Incidence of low birthweight, England and Wales, 1953–2010.

(Data from LHS 27/1 low birthweight returns and Office for National Statistics mortality statistics. Updated from Birth counts, Tables A3.4.1 and A3.4.2. )

The system changed in the early 1980s, when arrangements were made for health authorities to extract birthweights from babies’ birth notifications and send them to local registrars of births and deaths to be added to the birth registration. Data flows started to change from 2006 onwards when local register offices moved to registration online. This was an intermittent process, with many teething problems, which could have affected data quality. By July 2009, it was implemented in all register offices in England and Wales. A byproduct of this is that birthweight is now passed electronically to the Office for National Statistics from the notification by the midwife or doctor in attendance at the birth.

These data are analysed using the current definition of low birthweight, which was changed in the ninth revision of the International Classification of Diseases ( ) to ‘under 2500 g’. Babies weighing under 1500 g at birth are now categorised as ‘very low birthweight’, and those weighing under 1000 g are described as ‘extremely low birthweight’ ( ). These birthweight groups are used in Table 1.3 and in the data for the mid-1990s onwards in Figure 1.3 .

Interpreting trends

Both show recent trends in the incidence of low birthweight in England and Wales. Although the percentage of liveborn babies weighing under 2500 g has fluctuated since 1983, the general trend was upwards, followed by a levelling off from 2000 onwards, followed by an apparent drop in 2007. This could be a discontinuity resulting from the change of method of data collection. There were similar patterns in all groups of babies weighing under 2000 g, except that, in the under-1000-g group, the discontinuity in 2007 took the form of an increase.

Between 1983 and 1988 there was no clear trend in the very small proportion of liveborn babies for whom birthweight was missing, and who are known to include a high proportion of small and immature babies ( ). In the middle of 1989, financial constraints in the OPCS led to a decline in the completeness of recording of birthweight on birth registration records. Birthweight was missing on up to 4% of records from 1989 to 1994, making the data for these years difficult to interpret. As shown later in Table 1.18 , the mortality rate among babies with missing birthweights was well above the overall rate, suggesting that the group included a relatively high proportion of low-birthweight babies. By 1995, the numbers of missing birthweights had declined markedly and the data for 1995 onwards used in Tables 1.3, 1.9, 1.11, 1.13 and 1.18 (see below) and Figure 1.3 became much more reliable than those for the early 1990s. Unfortunately, however, the move to registration online was accompanied by a slight increase in the percentage of missing birthweights, from under 0.2% between 2000 and 2004 to 1.1% in 2007 and 0.74% in 2009.

The reported incidence of low-weight births in 1995 was well above that for 1988 and rose markedly after 1995, as Figure 1.3 shows. Analyses of birthweight data for both England and Wales and Scotland identified two separate trends, however. Although the percentages of low-weight births had increased during the 1980s, there had also been an increase in the proportion of heavier babies ( ; ; ). From the 1996s onwards in England and Wales, the proportion of heavier babies levelled off, fluctuating between 11.3% and 11.6% for singleton births.

Although the World Health Organization has published a definition of preterm birth, it has no definition of very preterm or extremely preterm, as a result of the lack of consensus mentioned in Chapter 2 . It does, however, recommend groups to be used when publishing data about gestational age. These are under 28, 28–31, 32–36, 37–41 and 42 or more completed weeks and the corresponding numbers of days ( ). The under-28-week category is often subdivided into less than 24 and 24–27 completed weeks, as has been done in this chapter.

When OPCS made the arrangements in the 1970s to acquire the information about babies’ birthweights from birth notification, it also requested gestational age. For reasons that are long forgotten, access to this was refused by clinical organisations. Data about gestational age of babies born in England and Wales only became available at a national level in England and Wales for 2005 onwards via the linkage with NHS Numbers for Babies data ( ; ). The data available so far are summarised in Table 1.4 and are published in fuller detail in Health Statistics Quarterly and elsewhere on the Office for National Statistics website ( ; ) As a result, the only data about long-term trends in preterm birth are for Scotland ( ; ; ). A reported increase in the proportion of preterm births in Scotland can be seen in Figure 1.4 and Table 1.5 , which are derived from information in the SMR2 system. Although multiple births have contributed to the rise, preterm singleton births have also increased since the mid-1970s as a proportion of all live births. Table 1.5 also shows a decrease in the rate of postterm births in Scotland and comparison with Table 1.4 suggests that this rate is lower than in England and Wales.

Percentage of live births born before 37 weeks of gestation, by multiplicity, Scotland, 1976–2010.

(Data from Information and Statistics Division Births in Scottish Hospitals.)

The data for both England and Wales and Scotland suggest that there has been an increase in the reported incidence of very small and very preterm babies. Although the rising incidence of multiple birth, discussed later, has made a major contribution to this rise, it is certainly not the only factor. It has been suggested that an increasing tendency to admit smaller and iller babies to neonatal nurseries has also contributed to the rise. By law all live births should be registered but there is a subjective element in distinguishing between a live birth and a miscarriage, particularly if the baby dies very shortly after birth. In the past some of these very tiny babies would have been regarded as miscarriages and would not therefore have been registered as live births. The lowering of the gestational age limit for registering fetal deaths as stillbirths in all countries of the UK in October 1992 may well have reinforced changes in people’s perceptions of which births should be registered as live births.

Another factor that may have contributed to the increase in registration is the growing recognition of parents’ need to mourn an unsuccessful outcome of pregnancy. The formalities of registration can sometimes form part of this, together with the process of holding a funeral.

Interpreting differences between areas and countries

The incidence of low birthweight varies both between geographical areas and between sectors of the population, including internationally. Considerable differences were seen in the late 1980s between the countries and parts of countries that took part in the International Collaborative Effort on birthweight, plurality and perinatal and infant mortality ( ). Similar differences were seen in the incidences of low birthweight in 1990 ( ), in the Peristat perinatal indicator project which used data for the years around 2000 ( ), and again in the Europeristat project using data for 2004, shown in Table 1.6 and Figure 1.5 ( ). These studies used data collected routinely through the participating countries’ vital statistics systems or medical birth registers ( ; ; ).

Percentage of live births which were preterm in European Union member states participating in the Europeristat project.

(Data from Europeristat, 2008.)

Although the overall incidence of low birthweight in a population tends to be a reflection of the health of that population in general, and of women of childbearing age in particular, at the bottom end of the birthweight range it is affected by the country’s criteria for birth registration. In theory this should not affect live births, as in most countries a live birth is registrable regardless of gestational age or birthweight. There are, however, considerable variations in the criteria for the registration of late fetal deaths as stillbirths. The differences within Europe can be found in Table 3.1 of the Europeristat report ( ). Many countries, including those in the UK, do not follow the World Health Organization’s recommendation to use a 22-week cut-off.

Inevitably, these wide differences in the gestational age at which fetal deaths are registrable as stillbirths can affect decisions about whether a very preterm birth should be regarded as a registrable live birth or as a miscarriage, although the extent to which they do so appears to vary from country to country. As a result, in the Europeristat study, the percentage of live births weighing under 500 g ranged from 0.00% in many countries to 0.05% in Germany and 0.06% in England and Wales. Although these differences have a very limited impact on the comparability of statistics about low birthweight, they have a much larger impact on the comparability of mortality statistics ( ; Mohangoo et al. 2011). To deal with this problem, the World Health Organization recommends that babies weighing under 500 g or born before 22 completed weeks of gestation are excluded from comparative statistics ( ). In the Euronatal study, with data for the mid-1990s and common cut-offs, the differences between countries changed when common cut-offs were applied ( ).

A further factor that has to be taken into account is the extent to which data about gestational age and birthweight are missing, either because the information was not recorded initially or because it was not passed on to population-based data collection systems. This is likely to have affected the trends shown in Figures 1.3 and 1.4 , as well as comparisons between countries. Furthermore, where data are almost complete, birthweight is most likely to go unrecorded for babies who die very soon after birth.

Real differences in the low-birthweight rates within countries with the same or similar data collection systems are shown in Table 1.7 . In 2009, the incidence of low-weight births in the Government Office regions of England ranged from 6.4% in the South-west region to 8.5% in the West Midlands region. Each of these regions includes a variety of different populations. Differences between local authorities are great. Even though these small babies make up a tiny proportion of all births, they make a considerable contribution to mortality rates. Comparing the countries of the UK, the incidences of low-weight and very-low-weight births in Wales were lower than those for England in 2009, while those for Scotland and Northern Ireland were lower still and the lowest rate of low birthweight shown in Table 1.7 was in the Republic of Ireland.

Differences between smaller areas, such as electoral wards, are even more marked. In each of the countries of the UK, area deprivation scores have been constructed to classify small areas. These use data from the census and other data, such as unemployment and crime rates, that can be disaggregated to a local level. They can be used to group together areas with similar characteristics within each country, but not for the whole of the UK, given that the composition of the scores is different in England, Wales, Scotland and Northern Ireland ( ).

Social class differences

Ethnic origin and country of birth

Birthweight and gestational age distributions are known to differ between ethnic and racial groups ( ; ; ; ).

At the time of writing, ethnic origin is not recorded at birth registration, but service users’ self-reported ethnicity should be recorded in national NHS data collection systems. Although mothers’ ethnic origins had been recorded on most hospital notes and on some districts’ birth notification forms for some years, the way it was recorded and classified varied widely. In the mid-1990s, it was decided that the definitions used in the 1991 census should be used universally ( ). These were superseded by the revised classifications used in the 2001 census and will subsequently be replaced by the categories used in 2011. In some cases, the data are incomplete and of questionable quality. Although data on birthweight and method of delivery, collected through the Maternity HES, have been published, black and Asian groups were aggregated in an attempt to overcome problems with data quality, but this obscures known differences within them ( ).

The 1991 population census was the first in which people were asked to indicate how they described their ethnic origin. The categories used in the question were: white, black–Caribbean, black–African, black–other, Indian, Pakistani, Bangladeshi, Chinese, and any other ethnic group. People descended from more than one ethnic or racial group were asked to tick the one to which they considered they belonged, or to tick the ‘any other ethnic group’ box and describe their ancestry ( ). This classification has been criticised on the grounds that it is more an indicator of skin colour than of cultural and social identity ( ).

These questions were revised for use in the 2001 census, when slightly different questions were asked in each of the countries of the UK. In England and Wales, specific categories for people of mixed ethnic background were added, along with the terms black British and Asian British, while Scotland added black Scottish and Asian Scottish categories. The Office for National Statistics has subsequently revised its categories further for use in the 2011 census, in the light of further criticisms, and added a question about national identity within the UK ( ). In addition, in 2001 and 2011, questions on religion, previously asked only in Northern Ireland, were added in England, Wales and Scotland, where they focused particularly on religions practised by minority ethnic groups.

The ethnic origins of parents are not recorded at birth and infant death registration in England and Wales but their countries of birth are recorded. Although not a measure of ethnic origin, as many women in some minority ethnic groups having babies in the UK today were themselves born in the UK and some were born in other countries to which their parents migrated, it is an approximate measure of migration status. Table 1.11 gives some insight into the differences in the incidence of low birthweight in 2009. As in other years and in particular in an analysis of data for the years 1983–2000, it was highest among babies with mothers born in Pakistan, India and Bangladesh, and nearly as high among babies whose mothers were born in the Caribbean Commonwealth, East Africa and the ‘rest of Africa’, which is predominantly West Africa. In contrast, the incidence of very low birthweight was markedly higher among babies whose mothers were born in the Caribbean Commonwealth and the ‘rest of Africa’ ( ). For data about births from 2010 onwards, the Office for National Statistics has adopted a new country of birth classification, which is not tied to membership of the British Commonwealth.

Table 1.11

Low birthweight by mother’s country of birth, England and Wales, 2009

MOTHER’S COUNTRY OF BIRTH	ALL	ALL STATED	UNDER 1500 G	1500–2499 G	UNDER 2500 G	UNDER 1500 G	1500–2499 G	UNDER 2500 G
	Numbers of live births					Percentage of live births with stated birthweight
All	706 248	701 011	8365	41 983	50 348	1.19	5.99	7.18
UK	532 046	528 274	6155	31 614	37 769	1.17	5.98	7.15
England and Wales *	522 808	519 106	6058	31 092	37 150	1.17	5.99	7.16
Scotland	7024	6967	71	416	487	1.02	5.97	6.99
Northern Ireland	2214	2201	26	106	132	1.18	4.82	6.00
Outside the UK	174 202	172 737	2210	10 369	12 579	1.28	6.00	7.28
Irish Republic	2971	2940	32	141	173	1.09	4.80	5.88
Other European Union	44 128	43 752	429	1982	2411	0.98	4.53	5.51
Rest of Europe	7861	7801	96	310	406	1.23	3.97	5.20
Commonwealth
Australia, Canada
and New Zealand	4567	4529	41	200	241	0.91	4.42	5.32
New Commonwealth	69 633	69 047	1070	5482	6552	1.55	7.94	9.49
Asia
Bangladesh	8452	8373	109	773	882	1.30	9.23	10.53
India	12 499	12 400	182	1208	1390	1.47	9.74	11.21
Pakistan	18 394	18 261	226	1567	1793	1.24	8.58	9.82
East Africa	4040	4004	67	292	359	1.67	7.29	8.97
Southern Africa	4654	4625	45	215	260	0.97	4.65	5.62
Rest of Africa	13 026	12 910	292	831	1123	2.26	6.44	8.70
Far East	1361	1349	9	86	95	0.67	6.38	7.04
Caribbean	3457	3414	96	235	331	2.81	6.88	9.70
Rest of the New Commonwealth	3750	3711	44	275	319	1.19	7.41	8.60
USA	3102	3057	32	138	170	1.05	4.51	5.56
Rest of the world and not stated	41 940	41 611	510	2116	2626	1.23	5.09	6.31

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Antenatal diagnosis and fetal medicine
2. Pharmacology
3. Ophthalmology
4. Coagulation values in term and preterm infants

Stay updated, free articles. Join our Telegram channel

Join