Sickle-cell disease is not a reportable condition, making it difficult to ascertain the number of affected individuals. We estimated the number of people with sickle-cell disease for the United States and each individual state, adjusting for increased mortality. US Census population data for each of the 50 states plus the District of Columbia were obtained. The published prevalence of sickle-cell disease for blacks and Hispanics of either Mexican or non-Mexican ancestry was applied. Analysis revealed 89,079 (95% confidence interval: 88,494-89,664) people with sickle-cell disease in the United States, 80,151 black and 8928 Hispanic. The state with the highest sickle-cell population was New York with 8308, followed by Florida with 7539, and Texas with 6765 people with sickle-cell disease. This study provides important information for researchers and policymakers attempting to better plan for the care of the sickle-cell population.
Sickle-cell disease is the most common inherited blood disorder in the United States. It is responsible for approximately 113,000 hospitalizations and $488 million dollars in hospitalization costs annually in the United States [1]. Although sickle-cell disease is a genetic disease diagnosed in this era by newborn screening, it is not a reportable condition. Therefore, it is difficult to ascertain the number of affected individuals in the United States. Newborn screening is an effective tool for diagnosis, but its use was not widespread until the late 1980s and 1990s [2], and it does not account for immigration. Further complicating population estimates is the increased mortality associated with the disease, both compared to the mortality for the underlying racial/ethnic group and when comparing more severe forms of the disease (hemoglobin SS (HgbSS) and hemoglobin S beta thalessemia (HgbSB 0 )) to less severe forms of the disease (hemoglobin SC (HgbSC) and HgbSB 1 ) [3][4][5][6][7]. Recent advances such as prophylactic penicillin and vaccines have reduced mortality, but only for those young enough to have received the treatments [8,9]. While no published studies have estimated the prevalence of sickle-cell disease in the United States, one NIH estimate puts the number ''between 50,000 and 75,000;'' another at 80,000, and the Sickle Cell Disease Association of America estimates the number to be ' 'over 70,000'' [10-12]
Methods
Census level-data and prevalence. Census data by age and race/ethnicity were obtained for each individual state and the District of Columbia and summed to equal the United States as a whole [13]. Year 2005 was used as the baseline. The applied prevalence rate of sickle-cell disease for blacks was 289 per 100,000 live births [14]. Prevalence rates for Hispanics were 89.8 Hispanic children of non-Mexican ancestry per 100,000 live births and 3.14 Hispanic children of Mexican ancestry per 100,000 live births [14].
These prevalence rates were applied to the Hispanic population numbers in each state based on the state's ratio of the number of Mexican births to non-Mexican births, assuming birthrate was reflective of the current ethnic population proportions. People identified as black-Hispanic (0.56% of the population) were classified as black. We did not include sickle-cell disease for whites (including Mediterraneans), Asian Indians, or other Asians in our estimate. While the disease occurs in these populations, the prevalence is very low, it is not possible to identify the country of origin from census data, and the mortality data specific to these populations is not known.
Consistent with previous literature, 60% of children at birth were classified as having HgbSS/HgbSB 0 and 40% were classified as having HgbSC/SB 1 [8,15,16]. Other sickle-cell disease genotypes, because the prevalence of these is very low, were excluded. Mortality adjustment. We adjusted for mortality based on age and sicklecell type. Children (<18 years old): For children with HgbSC/HgbSB 1 , survival is similar to the underlying population [8]. For children with HgbSS/HgbSB 0 , Quinn et al. [8] reported a 6.4% sickle-cell disease-related mortality in children by 18 years of age. We used linear interpolation to obtain an adjustment by the year. Although the majority of deaths occur early in childhood, this is countered by the use of the conjugate pneumococcal vaccine, which was started in 2001 and would only affect those 4 years old in our study. Young adults (age 18-45 years): For young adults with HgbSC/HgbSB 1 , there is no increase in sickle-cell-related mortality through 30 years of age [7]. From age 31 years through 45 years, there is a linear decrease in survival to 85% at 45 years of age; for similar ages, the black mortality rate is 5%, leaving an excess mortality of 10% between the ages of 31 and 45 years [7]. For young adults with HgbSS/HgbSB 0 , published reports show improved survival for those born during or after 1975 [4]. For those born during or after 1975 (18-30 years old), 89% survive to adulthood, and 71% survive to age 30 years [4]. With 75% of the deaths being sickle-cell related, the survival in our population of 18-30 years old was 85.4% of predicted based on population estimates. For those born before 1975 (31-45 years old in our population), 79% survived to age 20 years [4]. Ten-year survival from age 20 years shows that 67% would live to 30 years old, 52% would live to 40 years old, and 43% to 45 years old [4]. Middle-aged adults (age 46-65 years): For those with HgbSC/HgbSB 1 , approximately 50% will be alive at age 65 years compared to 65% of blacks, but this excess mortality occurred at younger ages [7]. For those with HgbSS/HgbSB 0 , 43% survived to age 45 years and only 10% will survive to age 65 years [4]. Older adults (age > 65 years): For those with HgbSS/HgbSB 0 , about 1% fewer each year were alive from age 55 to 65 years and we continued with this rate. For those with HgbSC/HgbSB 1 , we used linear extrapolation to continue the adjustment to the mortality rate from 45 to 65 years. States were grouped by geographic census regions. A 95% confidence interval (CI) was included for the national estimate based on a Poisson assumption for the variability of the count data. In addition to the CI, two sensitivity analyses were performed. The first analysis assumed that mortality was 10% lower for each age group than predicted from the literature; the second, based on a small study showing a higher prevalence of severe disease in Hispanics, assumed that 80% of Hispanics had severe disease at birth instead of 60% [16]. Population estimates, with mortality adjusted by age and sickle-cell type, yielded a total year 2005 estimate of 89,079 (95% CI: 88,494-89,664) people with sickle-cell disease in the United States, of which 80,151 were black
and 8,928 Hispanic (Fig. 1). The distribution across age groups revealed 35,726 (40% of the population) children with sickle-cell disease. Regional totals are shown in Fig. 1.
The South, with a sickle-cell population of 47,354 people, comprised 53% of the sickle-cell population. The estimated sickle-cell population of New York (8308 people), was more than three-fourths of that of the entire Western region.
The estimated number of people with sickle-cell disease in each state is shown in Supporting Information Table I. The states with the highest estimated number of people with sickle-cell disease were New York with 8308;
Florida with 7539; Texas with 6765; California with 6474; and Georgia with 5890. These five states comprised more than 43% of the total sickle-cell population. Supporting Information Table II shows both the absolute and relative percentages of sickle-cell genotypes by age. The increased mortality for HgbSS/HgbSB 0 leads to an alteration in the relative percentages of sickle-