Very little epidemiologic research is available about pregnancy complications that lead to hospitalization in the United States. Data from the National Hospital Discharge Survey for 1986 and 1987, which uses a representative sample of non-federal, short-stay hospitals in the US, was analyzed to obtain a national overview of the public health burden associated with inpatient care for pregnancy complications. Hospitalizations that involved an induced abortion, occurred postpartum or included delivery were excluded since these hospital stays could have lead to a secondary diagnosis of pregnancy complications, which alone would not have required inpatient care. Pregnancy complications were divided into two major categories for the analysis. Antenatal hospitalization referred to a woman who was still pregnant at time of discharge. Pregnancy loss hospitalization referred to a woman who was hospitalized for pregnancy complication, but during her stay incurred a pregnancy loss without the delivery of a live or stillborn fetus.

In 1986 and 1987, there were approximately 849,200 pregnancy-related hospitalizations per year that did not involve an induced abortion, postpartum diagnosis and delivery. There were 14.6 antenatal hospitalizations and 7.6 pregnancy loss hospitalizations per 100 deliveries. Black women had a 40% higher rate of hospitalizations per 100 deliveries for both categories than the white women did. Thus, all further analyses were stratified by race. With respect to antenatal hospitalizations, the ratio decreased with age among white women and increased with age for black women. At time of hospitalization, 20.5% of the white women were unmarried, while 64.3% of the black women were unmarried. The antenatal hospitalization ratio was 60% higher among unmarried than among married white women. In contrast, the antenatal hospitalization rate among black women did not vary by marital status. Also, 36.5% of the white women and 66.1% of the black women had no health insurance. The antenatal hospitalization ratio was 40% higher among uninsured than among insured white women, while it was only 14% higher among uninsured than among insured black womem. For all antenatal diagnoses, the mean length of stay for black women was 3.33 days, approximately one third greater than the 2.5 days for the white women. With respect to pregnancy loss hospitalization, the ratio significantly increased from age group 20-34 to age group 35-44 across the white and black population. At age group 15-19, the ratio for blacks compared with whites was 4% lower, but by age 20-34 the ratio for blacks was 55% greater than for whites, and, for oldest age category 35-44, the ratio for blacks was 76% greater. Marital status did not affect the ratio of hospitalization for blacks, but the ratio for unmarried white women was almost double that of married white women. Also, the pregnancy loss hospitalization ratio was 24% lower among uninsured than among insured black women, while among white women, the ratio did not vary by insurance status.

A major finding was that blacks had a 40% greater ratio of hospitalization for antenatal conditions and pregnancy loss management per 100 deliveries than whites had. The authors cite several possible explanations for the racial difference. They include actual greater incidence of pregnancy complications among blacks, greater likelihood of blacks receiving inpatient than outpatient care for pregnancy complication, greater likelihood of multiple hospitalizations during pregnancy among blacks, and greater severity of clinical presentation of complications among the black women. The last two explanations are compatible with findings that show black women generally receive less adequate prenatal care than their white counterparts. With the lack of both appropriate prenatal care and early intervention, pregnancy complications may progress significantly in black women and require inpatient care.

The lack of differentiation between multiple admissions of the same person from hospitalizations for different persons and the paucity of information in hospital discharge data on sociodemographic factors may limit the precision and utility of the findings to some extent. Nevertheless, the finding of this study based upon hospital discharge data demonstrated the magnitude of the public heath burden of pregnancy morbidity. However, the study did not account for the indirect burden or cost of pregnancy complication hospitalization, which includes family disruption, loss of productive days on part of the hospitalized woman or other individual and emotional distress. Furthermore, the study suggests that the quality of prenatal care and medical intervention for obstetric complications vary by race. Inadequate routine outpatient prenatal care among black women appears to result in greater likelihood of obstetric complication hospitalization, which is significantly more costly than outpatient prenatal care.