Australian Bureau of Statistics
Celebrating the International Year of Statistics 2013

APPENDIX 2 DIFFERENCES BETWEEN COLLECTIONS


BIRTH REGISTRATIONS COMPARED TO THE PERINATAL DATA COLLECTION

Birth registrations data in this publication are not the only births data available in Australia. The National Perinatal Statistics Unit of the Australian Institute of Health and Welfare (AIHW) also collects birth data from midwives and other health professionals who attend births. This data is published annually in Australia's Mothers and Babies.

As information from these two collections are from different sources, the statistics obtained vary. For example, the Perinatal Data Collection reported the occurrence of 292,000 live births in Australia in 2007 (the latest available data), 2.4% more than the 285,200 births registered in the same year.

Graph A2.1 shows the total number of live births in Australia by type of collection and recording basis from 1994 to 2009. Births from the Perinatal Data Collection are shown on a year of occurrence basis. Registered births from the ABS Birth Registrations collection are shown on both a year of registration and year of occurrence basis, which includes births that occurred in a particular year but may have been registered up to and including the year 2009. Prior to 1994, the Perinatal Data Collection showed fewer births than births registered. This position then reversed, with the gap between births reported in the two collections widening until 1999 (when there were 2.7% more births recorded in the Perinatal Data Collection than births registered). The size of the difference between the two collections decreased to 0.4% by 2004, and increased to 2.4% in 2007.

While difficult to explain the differences, the greater number of births in the Perinatal Data Collection may be due to improvements in quality and coverage, particularly with the introduction of a perinatal National Minimum Dataset (NMDS) in 1997 which developed national standards for the collection of perinatal statistics. The trend may also reflect the increasing likelihood over time of parent(s) to delay or fail to register the birth of a child.

The following analysis of differences between the collections is made on live births between the Perinatal Data Collection on a year of occurrence basis and the Birth Registrations collection on a year of registration basis.


Age of mother

Age of mother is a factor which contributes to differences between the two collections. In 2007, higher numbers of confinements were recorded in the Perinatal Data Collection than in the Birth Registrations Collection for mothers in all age groups. For mothers aged 19 years and under, there were 7.0% more confinements recorded in the Perinatal Data Collection than in the Birth Registrations Collection, while for mothers aged 20-24 years the difference was 4.2%.

A2.2 Confinements, Difference between collections by age of mother - 2007
	Perinatal Data Collection	Birth Registrations	Difference(a)
	no.	no.	%
19 years and under	11 912	11 128	7.0
20-24 years	42 233	40 529	4.2
25-29 years	77 481	74 998	3.3
30-34 years	93 414	91 461	2.1
35-39 years	54 138	52 400	3.3
40 years and over	10 292	9 945	3.5
Total(b)	289 496	280 781	3.1
(a) Positive figures denote more confinements recorded in the Perinatal Data Collection than the Birth Registrations collection.
(b) Includes age of mother not stated.

Births to Indigenous mothers

There are differences between Indigenous data from the Perinatal Data Collection and birth registrations data. In all years from 1999 to 2007 the number of births to Indigenous mothers as recorded in the Perinatal Data Collection exceeded registered births to Indigenous mothers. In 2007 there were 10,900 live births to Indigenous mothers recorded in the Perinatal Data Collection, 700 more than the number recorded in the Birth Registrations collection (10,200).

It is important to recognise that data concerning Indigenous status is affected by identification issues. Differences between the Perinatal Data Collection and birth registrations data may in part be due to undercoverage of registrations of Indigenous births or delays in the registration of Indigenous births. However, failure to determine Indigenous status may also affect the Perinatal Data Collection. As a result, caution should be used when interpreting Indigenous births data from either source. Paragraphs 34 to 42 of the Explanatory Notes provide further information about Indigenous births.

State and territory comparisons

While birth registration data provides information on state/territory of registration or state/territory of usual residence of mother, the Perinatal Data Collection only provides data on the state/territory in which the birth took place (that is, the state/territory of occurrence). The following table compares state/territory of usual residence from birth registrations to state/territory of occurrence from the Perinatal Data Collection. As a small number of births occur in a different state or territory to that of the mother's usual residence, there are some minor differences. For example, some women living in rural New South Wales close to the Australian Capital Territory have their babies in Canberra. As a consequence, the Australian Capital Territory has been excluded from the comparisons below.

In 2007, the Perinatal Data Collection recorded more births in New South Wales, Victoria and Western Australia, while the ABS birth registrations collection recorded more births in Queensland, South Australia, Tasmania and the Northern Territory.

A2.4 Live births, Type of collection by state/territory - 2007
	NSW	Vic.	Qld	SA	WA	Tas.	NT	Aust.(a)
Perinatal Data Collection (no.)	95 387	71 778	59 827	19 620	29 885	6 268	3 767	292 027
Birth Registrations (no.)	89 495	70 313	61 249	19 662	29 164	6 662	3 894	285 213
Difference (%)(b)	6.6	2.1	-2.3	-0.2	2.5	-5.9	-3.3	2.4
(a) Includes Australian Capital Territory and Other Territories.
(b) Positive figures denote more births recorded in the Perinatal Data Collection than the Birth Registrations Collection.