Gender Differences in Hospitalized Patients with Sickle Cell Disease (SCD): 1990–2002.

We show in a companion abstract that there has been no decrease in overall admissions for adults with sickle cell anemia (SCA) in the US for the period 1990–2002. The objective of this study was to further describe the characteristics of all adult and pediatric inpatients with sickle cell disease (SCD) in the US for this period.

Methods: We examined data from the National Hospital Discharge Survey (NHDS) for the period 1990–2002. The NHDS is a nationally representative survey of non-federal short-stay hospitals in the U.S. which allows for the estimation of inpatient utilization data. Inclusion criteria for our analysis consisted of the following ICD-9 codes: 282.61 or 282.62 (classified as SS), and 282.60, 282.63, 282.64, and 282.69 (classified as Other/Not Specified) found in any one of seven possible patient diagnostic records. Gender and SCD genotype differences in mean length of stay (LOS) were assessed using univariate analyses, and multivariate analyses adjusting for insurance status, age, hospital bedsize, time, and region of the country. Analyses were conducted for the population in the aggregate, and stratified by adult/pediatric status (18+ years for adults, 0–17 for pediatric).

Results: There were 13,221 records, which met our inclusion criteria resulting in an estimate of 1,079,523 SCD discharges over the 13 year period. (83,040 discharges/year) More than half of the adult discharges were for females, while less than half of the pediatric discharges were for females (55% vs. 44%, p < 0.0001). A greater percentage of adult than pediatric discharges were coded as SS (81% vs. 74%, p < 0.0001). Among adults, patients coded as SS were younger than patients with an Other/Not Specified SCD code (30.6 vs. 35, p < 0.001). The reverse was true for pediatric cases, where patients coded as SS were slightly older than patients coded as Other/Not Specified SCD (9.6 vs. 7.7, p < 0.001). In univariate analyses of LOS data, adult females with SS had longer mean LOS than adult males with SS (7.27 days vs. 6.29 days, p = 0.001). There were no differences in mean LOS for adult females and adult males with Other/Not Specified SCD codes (p = 0.160). Adult males with SS and adult males with Other/Not Specified SCD codes did not differ in their mean LOS (p = 0.850), while adult females with SS experienced significantly longer mean LOS than adult females with Other/Not Specified SCD codes (7.27 days vs. 5.45 days, p < 0.001). Among the pediatric cases, females with SS experienced longer LOS than females with Other/Not Specified SCD codes (4.57 days vs. 3.22 days, p < 0.001). Adult females continued to experience nearly a full day longer mean LOS than adult males in multivariate analyses (β = 0.83, p = 0.006). Among adults, 27% of men had medicare compared to 19% of the women and 17% of men compared to 25% of women had private insurance. (p < 0.0001)

Conclusions: Our analysis of the characteristics of inpatients with SCD demonstrates that adult females with SCD account for a higher percentage of inpatient visits than adult males with SCD. Females with SS experience longer mean inpatient LOS than females without SS and males with or without SS, regardless of age. The reasons for these differences are unknown, but warrant further study. Of the hosptilized adults more women had private insurance than men. Whether this impacts on likelihood of admissions needs closer examination. A better understanding of patient characteristics will help guide interventions that will limit barriers to care for this patient population.