Table 12 presents live births by gestational age for as many years as possible between 1990 and 1998.
Description of variables: Gestational age of a live-born infant is defined as the interval, in completed weeks, between the first day of the last menstrual period of the mother and the day, month and year of delivery. Three new definitions associated with gestational age were adopted at the Twenty-Ninth World Health Assembly following this recommendations of the Ninth Revision Conference, 1975, as follows: pre-time is less than 37 completed weeks (less than 259 days); term is 37 through 41 completed weeks (259 to 293 days); post-term is 42 or more completed weeks (294 days or more).
The gestational age classification used in this table is as follows: under 27 weeks, 28-31 weeks, 32-35 weeks, 36-39 weeks, 40 weeks and over and an unknown category. Where age has been reported in months - either calendar or lunar - it has been converted to gestational weeks according to the following scale:
months= under 28 weeks
7 months = 28-31 weeks
8 months = 32-35 weeks
9 months = 36-39 weeks
10 months = 40 weeks and over
Reliability of data: Data from civil registers of live birth which are reported as incomplete (less than 90 per cent completeness) or of unknown completeness are considered unreliable and are set in italics rather than in roman type. Table 1 and the technical notes for that table provide more detailed information on the completeness of live-birth registration. For more information about the quality of vital statistics data in general, and the information available on the basis of the completeness estimates in particular, see section 4.2 of the Technical Notes.
Limitations: Statistics on live births by gestational age are subject to the same qualifications which have been set forth for vital statistics in general and birth statistics in particular as discussed in section 4 of the Technical Notes.
The reliability of the data, an indication of which is described in the above paragraph, is an important factor in considering the limitations. In addition, some live births are tabulated by date of registration and not by date of occurrence; these have been indicated by a (+). Whenever the lag between the date of occurrence and date of registration is prolonged and, therefore, a large proportion of the live-birth registrations are delayed, birth statistics for any given year may be seriously affected.
Another factor which limits international comparability is the practice of some countries or areas not to include in live-birth statistics infants who were born alive but died before registration or within the first 24 hours of life, thus under estimating the total number of live births. Statistics of this type are footnoted.
The use of the onset of the last menstrual period as the basis for computing gestational age provides, in effect, only an estimate of the length of pregnancy, which is actually the time from conception to birth. Since the exact time of conception cannot be determined, it is recommended that gestational age, computed from the onset of the last menstrual period, be used to approximate the true period of gestation, although it probably overestimates the true period of gestation by about two weeks.
The actual estimation and recording of gestational age itself are subject to considerable uncertainty. In many cases, gestational age is estimated from the woman’s recollection of the date of onset of the last menstrual period or medical records of this date, while in other cases gestational age is inferred from the birth weight or other characteristics of the infant or from other information about the probable time of conception.
Moreover, gestational age can only be determined by inference for those conceptions that occur during post-partum amenorrhoea, and there is considerable variation between countries in the proportion of conceptions occurring in this period. However, even where it is possible to use the date of onset of last menstrual period to estimate gestational age, memory errors and errors arising in the conversion of this date to the gestational age of the infant may introduce biases into tabulations by gestational age.
It is important to note that gestational age should not be estimated on the basis of birth weight because it is often desirable to study the correlation of these two variables, which is not possible if gestational age is estimated on the basis of birth weight. Data on both gestational age and on birth weight, the latter presented in table 11 are useful in the study of infant mortality because infant of low birth weight and those of curtailed gestation are subject to higher risks of mortality and morbidity.
The relatively limited coverage of this table is due to the fact that many countries or areas do not collect and tabulate data on gestational age.
Coverage: Live births by gestational age are shown for 42 countries or areas.
Earlier data: Live births by gestational age have been shown in previous issues of the Demographic Yearbook featuring natality. For information on the years covered, readers should consult the Index.
 Principles and Recommendations for a Vital Statistics System Revision 2, United Nations Publication, Sales No. E.01.XVII.10, Para.142., New York, 2001.
 Manual of the International Statistical Classification of Disease, Injuries, and Causes of Death Based on the Ninth Revision Conference, 1975, and Adopted by the Twenty-ninth World Health Assembly, 1975 Revision, vol. 1, p.764., World Health Organization, Geneva, 1977.