This retrospective cohort study conducted during the calendar year 1981 at Johns Hopkins Hospital, a large teaching hospital located in Baltimore, Maryland, was implemented to assess the pattern of use of routine prenatal ultrasound examinations; in particular, whether usage varied by socioeconomic status and race. The study contrasted the sociodemographic (age, race, highest education level attained, and clinic vs. private patient status) and reproductive characteristics (pre-pregnancy weight, weight gain during pregnancy, time period of initial prenatal visit, number of prenatal visits, history of preterm delivery, receipt of amniocentesis, and sex of offspring) of women who received prenatal ultrasound examinations with women who did not receive such examinations. All women who delivered live singletons during 1981 at John Hopkins were initially eligible. Multiple births were excluded since obstetric ultrasound is diagnostically indicated for multiple pregnancy as well as assessment of growth retardation and placenta previa. Furthermore, ineligible women were those who had no record of prenatal care at John Hopkins, who delivered infants who were designated as other than Black or white, and women who received ultrasound during pregnancy for non-related pregnancy related indications. Thus, 2542 women were retained in the study. Since the use of ultrasound examinations is included in individual obstetric record and also recorded in the patient billing records, the validity of ultrasound usage was verified by determining the extent of agreement between the two sources. Random selection of over 250 obstetric records was compared to the billing information and agreement between the two sources was approximately 99%. Furthermore, only 2% of the randomly selected obstetric records cited outside use of prenatal ultrasound, which clearly is not recorded in the billing records. However, all these women also received in-house ultrasound examinations. The ultrasound exposure status was ultimately derived from patient billing records. A standard form was used to abstract from the medical records the information on all other covariates, the health status and medical status of the women.

Black women compared with white women, women with lower formal education compared with women with higher formal education, and clinical patients compared with private patients were more likely to undergo prenatal ultrasound testing (all p-values<0.001). The number and timing of prenatal visits were also correlated with ultrasound usage. The later during her pregnancy a woman presented for her initial prenatal visit (a clear dose-response was observed) and the fewer the number of prenatal visits, the greater the likelihood of undergoing a prenatal ultrasound. Furthermore, the association between pregnancy risk factors and use of ultrasound was correlated. Low pre-pregnancy birth weight and high pre-pregnancy birth weight, weight gain of less than 20 lbs. during pregnancy, previous preterm pregnancy, and women greater than 35 years were all more likely to undergo ultrasound examination. Also, women who underwent amniocentesis were more likely to receive ultrasound. In contrast, sex of the offspring was not associated with the receipt of prenatal ultrasound.

The strengths of the study lie in its selection of its subjects, 2542 women who delivered during a set calendar year, and the verification of the ultrasound exposure. Furthermore the use of education and income, measured by registration status (private vs. clinic patient) are traditional and acceptable indicators of socioeconomic status. However, a major limitation of the study rests with its analyses. It only assessed the association between ultrasound and one variable at a time. Thus, why black women underwent ultrasound more than white women as well as why clinic patients were more likely to receive ultrasound than private patients cannot be determined. The authors conjecture that women of lower socioeconomic status are more likely than their higher socioeconomic counterparts to have adverse pregnancy risk factors and outcomes. They never examine the association between adverse pregnancy risk factors and maternal race, education status and registration status. However, the authors contend that low socioeconomic status is not a barrier to prenatal ultrasound since the test is not prohibitively expensive, availability of both ultrasound equipment and personnel at Johns Hopkins, and potential fear of medical litigation. In the introduction the authors state that the use of routine obstetric ultrasound has not been shown to improve fetal outcomes significantly, although some obstetricians have advocated its use. Interestingly, the authors do not evaluate their findings in view of the undemonstrated efficacy of routine fetal ultrasound. Nevertheless, the authors state that results are “encouraging” since they challenge strongly the preconceived inequality of access to medical care. They conclude that “the performance of ultrasound examination on the mother in this study appear to be far more related to the existence of risk factors in the mother for a complicated pregnancy than to socioeconomic status as measured by educational attainment or the mother’s ability to pay."