Potentialities and Limitations of Medical Data on Official Birth Certificates

Abstract

TO KEEP ABREAST of changes in the characteristics of a population with regard to pregnancies and newborns and to keep public health programs in this area geared to these changes, it is desirable and helpful to collect and study comprehensive statistics with which to measure and evaluate these changes and the program needs which they entail. A simple and inexpensive method of collecting source data for such statistics is through the addition of a medical supplement to the birth certificate. Yet, this method has not gained intensive application because it has been found that the data so collected are often incomplete and inaccurate and because it has not been demonstrated convincingly that the data are of practical value for program purposes. This paper describes our more recent experience in the District of Columbia with the collection of comprehensive data on a medical supplement and with the utilization of statistics derived therefrom. The experience to be reported is that of an urban community having some 33,000 live births a year. A medical supplement to the birth certificate has been used in the District of Columbia since 1940. Little was done with this material until special resources became available for a comprehensive study of the 1952 birth certificate data. The methodology and findings of this study have been reported by Oppenheimer, et al.,' and the 1952 data used in this paper have been taken from that report. * A standing advisory committee of representatives of the obstetric and pediatric groups in the community participated in the revisions of the medical supplement and in promoting reporting thereon.

Use of Congenital Malformation Data Reported on Live Birth Certificates

“…However, with the use of live birth certificate information many difficulties arise which should be resolved in order for the resultant data to be meaningful. Foremost is the problem of incompleteness of reporting, a matter commented on by a number of our respondents and discussed by many experts (5)(6)(7)(8)(9)(10)). This incompleteness is due to a number of factors, among which are the following.…”

“…5. The questions asked should be periodically reviewed, so that obsolete items may be deleted and new ones added as medical progress is achieved. 6. The questions should be carefully worded to avoid ambiguity.…”

Variations in the reporting of gestational age at induced termination of pregnancy.

“…This study was reported in part at the 116th annual meeting of the American Public Health Association, Boston, MA, November [13][14][15][16][17] 1988.…”

Comparative accuracy of clinical estimate versus menstrual gestational age in computerized birth certificates

“…Anomalies of the distribution of gestational age obtained by clinical estimate included obvious digit preference (higher frequencies with even weeks) and as many as 87% of values concentrated in the single week's category for 40 weeks, a finding considered unrealistic. [9][10][11] However, the limitations of LMP-based gestational age determinations have also been reported widely in the scientific literature. Several factors may lead to an error in gestational age derived from the LMP, including irregularities of menstrual cycle, individual variation in length of the cycle, preconception amenorrhea following oral contraceptives, implantation bleeding or other bleeding early in pregnancy, and recall errors by mothers.…”