Owusu-Edusei, Kwame Jr PhD; Introcaso, Camille E. MD; Chesson, Harrell W. PhD
From the Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, GA
The authors thank Dr. Thomas L. Gift for his critical review and insightful suggestions.
Conflicts of interest: None.
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control and Prevention. Mention of company names or products does not imply endorsement by the Centers for Disease Control and Prevention.
Correspondence: Kwame Owusu-Edusei Jr, PhD, Centers for Disease Control and Prevention, 1600 Clifton Road MS E-80, Atlanta, GA 30333. E-mail: Kowusuedusei@cdc.gov.
Received for publication April 26, 2012, and accepted November 2, 2012.
Untreated syphilis during pregnancy can lead to stillbirth, neonatal death, or to liveborn infants with congenital syphilis (CS).1,2 Congenital syphilis can cause severe and chronic adverse health effects such as deafness, blindness, neurologic impairment, and bone deformities.1–7 Reported cases of CS in the United States reached a historic low of 339 cases in 2005, increased from 2006 to 2008, and have since fallen to 377 cases in 2010.8
Congenital syphilis can be prevented by early detection of maternal infection and treatment at least 1 month before delivery.1 Thus, routine serologic screening of pregnant women and rapid, adequate treatment for those infected are recommended to prevent CS.9 The Centers for Disease Control and Prevention recommends routine serologic syphilis screening in the first trimester for all pregnant women and repeat screening during the third trimester and at delivery for women at high risk for syphilis.1 However, owing to lack of, or inadequate prenatal care, some women are not diagnosed or treated for syphilis until delivery or after.
Estimates of the cost of treating CS are needed for cost-effectiveness analyses of interventions to prevent CS. However, the only previously published studies of the hospitalization costs of CS were based on data obtained in 1989 and 1990.10,11 Owing to rapidly changing medical technology, treatment recommendations, and health service delivery, periodic updates of medical costs are essential. Thus, the purpose of this study was to estimate the hospitalization cost per case of CS in the United States, using more recent medical claims data (2005–2009).
We used inpatient admission claims data from the MarketScan Database (Thomson Reuters MarketScan Database, Thomson Reuters [Healthcare] Inc, Ann Arbor, MI) for 2005 through 2009. As of 2005, the MarketScan database contained claims data on more than 17 million people who have employer-sponsored health insurance from more than 100 payers, including large employers, health plans, and government and public organizations.12 The database includes data on fully adjudicated and paid claims for persons with employer-sponsored health plans, including their spouses and dependents.13
We used International Classification of Diseases, Ninth Revision codes for CS (090.0–090.914,15) to identify and extract hospital admissions with CS diagnoses. To enable us to estimate the excess inpatient admission cost attributable to CS (when compared with “healthy/uncomplicated normal newborns”), we extracted claims for live birth admissions using International Classification of Diseases, Ninth Revision codes V30.00–V39.20.15–17 We then used Diagnosis Related Group codes 391 or 795 (i.e., uncomplicated normal newborn12) to identify those with no disease nor delivery complications.
Total cost was the total gross payments to all providers who submitted claims for covered services rendered during the admission.18 We deleted claims records with total payment less than $1 because these records were possible data entry errors. We deleted claims records with missing or invalid entries for costs and enrollee characteristics.
We conducted simple summary statistics on the total payments for those with and without CS diagnosis to determine unadjusted estimates. Following previous studies,10,19 we used regression analyses to adjust the cost estimates (adjusted estimates) for enrollee characteristics (independent variables) provided in the medical claims database. Specifically, we used a regression model in which the dependent variable was the natural log of the hospitalization costs. The natural log transformation of the costs was to reduce the influence of outliers on the estimates.
The independent variables were dichotomous or polychotomous variables that included sex, year of service, region, and other factors listed in Table 1. The independent variable of most interest to this study was a dichotomous variable for CS diagnosis—which was set to 1 for infants with a CS diagnosis and set to 0 otherwise.
We used MEDSTAT DataProbe System version 3.2.7 (Thomson Reuters [Healthcare] Inc) for data extraction. STATA version 11.1 (StataCorp LP, College Station, TX) and SAS version 9.2 (SAS Institute, Cary, NC) were used for regression result validation and diagnostics. Microsoft Excel, version 2010 (Microsoft Corporation, Redmond, WA), was used for summary analyses and presentation of results. All costs were adjusted to 2009 US dollars using the medical care component of the Consumer Price Index for All Urban Consumers.20
From the available inpatient admissions data for 2005 to 2009 (Table 1), we identified 44 claims records from infants with a CS diagnosis and almost half a million usable claims records from uncomplicated healthy infants (n = 474,849).
The estimated unadjusted average hospitalization cost was $1689 for uncomplicated healthy newborns and $25,659 for admissions with CS diagnosis (Table 1). Thus, the excess unadjusted cost was $23,970 for infants with a CS diagnosis. The average length of stay was 2.0 days for uncomplicated healthy newborn admissions and 10.1 days for admissions with CS diagnoses (Table 1).
A summary of the results from our regression analyses for the hospitalization costs is presented in Table 2. Based on these regression results, the adjusted mean cost per hospitalization for healthy infants was $1408 (see Table 3). The average excess cost per infant diagnosed as having CS (above and beyond the hospitalization costs of healthy infants) was $9969 (95% confidence interval [CI], $5702–$16,769). Our estimate of $9969 is consistent with previous estimates of the hospitalization costs of CS (approximately $7000 to $8000 when updated to 2009 US dollars) based on data obtained in 1989 and 1990.10,11 Still, after inflation adjustments, our estimate of the cost of hospitalization for CS was at least 24% higher when compared with the previous estimates.
Our estimate for the average length of stay for uncomplicated newborn admissions (2.0 days) was virtually the same as was found in a recent study in the United States (1.9 days).16 For those with a CS diagnosis, we found an average stay of about 10 days, which is consistent with current treatment guidelines 21 and with the 7.5 additional days for infants with CS reported by Bateman et al.10
The coefficients from our regression analyses for the control variables were consistent with the existing literature. For example, we found that hospitalization costs were 13% lower for females than for males, which can be attributed to the high rate of neonatal male circumcision performed within the first 2 days of birth in the United States.22 Our finding that the cost of inpatient admissions was lowest for the South is consistent with a recent study of the cost of diabetes among privately insured persons in the United States.19 Similar to surveillance reports of CS cases, we found that most claims with a CS diagnosis were from the Southern region.8 However, this was partially caused by the fact that the South accounted for a larger proportion of the claims data than any other region.
Our study has several limitations. First and most important are the usual limitations associated with medical claims data. For example, diagnosis codes in claims data can be inaccurate owing to data entry errors or miscoding,23,24 and the presence of a code does not necessarily imply that it is a confirmed case. More detail analyses of health records (such as chart reviews) on both the mother and the child before, during, and after hospitalization would be needed to understand the complete basis of the diagnosis. Second, there were only 44 claims with a CS diagnosis in our analysis. Third, the incidence of CS varies significantly based on race/ethnicity,8 yet the claims database did not include information on race.18 It is difficult to determine how our estimates would change had we been able to control for race in the regression analyses. Fourth, our results are from a convenience sample and are not generalizable to the entire US population because only claims data from privately insured individuals were analyzed in this study. This limitation is particularly noteworthy because CS might occur more frequently and be more severe among the uninsured population or among people insured through Medicaid, resulting in higher costs. Fifth, our estimates only include the direct medical costs associated with the initial treatment of CS. Our cost estimates do not include costs incurred before admission, such as costs related to syphilitic stillbirths or after discharge, such as costs associated with children presenting with late CS. Furthermore, our estimates do not include intangible costs (such as pain and suffering), nonmedical costs (such as travel-related costs to the hospital), or productivity costs associated with CS.
The 3 key strengths of our study are that (1) we used a large insurance database with claims from all states (whereas previous estimates were based on CS costs in a single city or state10,11), (2) we used more recent data (2005–2009), and (3) our estimates represent actual dollar amounts paid for the medical services rendered.13,18 In addition, our estimate indicated that the hospitalization cost associated with CS diagnoses was substantially higher (>24%) when compared with previous estimates adjusted to 2009 dollars. Thus, although evaluation and treatment of CS itself have not changed significantly in the last few decades, numerous advances have been made in supportive care for the sick neonate that may have increased cost. It was therefore worthwhile to update these estimates from real and more recent hospitalization payments rather than just adjusting previous estimates to 2009 dollars. The updated estimate of the excess hospitalization costs per case of CS ($9969) can help to quantify the overall burden of syphilis in the United States and can inform the economic evaluation of interventions to prevent syphilis and the associated adverse health outcomes of syphilis.
1. Su JR, Berman SM, Davis D, et al.. Congenital syphilis—United States, 2003–2008. MMWR Morb Mortal Wkly Rep 2010; 59: 413–417.
2. Shafii T, Radolf JD, Sanchez PJ, et al.. Congenital syphilis. In: Holmes KK, Sparling PF, Stamm WE, et al., eds. Sexually Transmitted Diseases. New York: McGraw Hill, 2008: 1577–1612.
3. McDermott J, Steketee R, Larsen S, et al.. Syphilis-associated perinatal and infant mortality in rural Malawi. Bull World Health Organ 1993; 71: 773–780.
4. Hira SK, Bhat GJ, Chikamata DM, et al.. Syphilis intervention in pregnancy—Zambian demonstration project. Genitourin Med 1990; 66: 159–164.
5. Watson-Jones D, Changalucha J, Gumodoka B, et al.. Syphilis in pregnancy in Tanzania. I. Impact of maternal syphilis on outcome of pregnancy. J Infect Dis 2002; 186: 940–947.
6. Watson-Jones D, Gumodoka B, Weiss H, et al.. Syphilis in pregnancy in Tanzania. II. The effectiveness of antenatal syphilis screening and single-dose benzathine penicillin treatment for the prevention of adverse pregnancy outcomes. J Infect Dis 2002; 186: 948–957.
7. Gomez GB, Kamb ML, Newman LM, Broutet N, Mark J, Hawkes SJ. The impact of syphilis on adverse pregnancy outcomes: A systematic literature review and meta-analysis. Under review.
8. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2010. Atlanta, GA: US Department of Health and Human Services, 2011.
9. Wolff T, Shelton E, Sessions C, et al.. Screening for syphilis infection in pregnant women: Evidence for the U.S. Preventive Services Task Force reaffirmation recommendation statement. Ann Intern Med 2009; 150: 710–716.
10. Bateman DA, Phibbs CS, Joyce T, et al.. The hospital cost of congenital syphilis. JPediatr 1997; 130: 752–758.
11. de LG, Zenilman J, Nelson KE, et al.. The cost of a preventable disease: Estimated U.S. national medical expenditures for congenital syphilis, 1990. Public Health Rep 1995; 110: 403–409.
12. Thomson Rueters (Healthcare) Inc. MarketScan Database Training at the Centers for Disease Control and Prevention. Atlanta, GA: Thomson Rueters (Healthcare) Inc. November, 2011.
13. Adamson DM, Chang S, Hansen LG. Health research data for the real world: The MarketScan databases. White Paper. Thomson Healthcare, 2006: 1–32.
15. Buck CJ. 2011 ICD-9-CM
for Hospitals. Volumes 1, 2 & 3. Professional ed. St Louis, MI: Elsevier Saunders, 2011.
16. Russell RB, Green NS, Steiner CA, et al.. Cost of hospitalization for preterm and low birth weight infants in the United States. Pediatrics 2007; 120: e1–e9.
18. Thomson Rueters (Healthcare) Inc. Marketscan Research Database: User Guide and Database Dictionary. Ann Arbor, MI: Thomson Rueters (Healthcare) Inc., 2009.
19. Shrestha SS, Zhang P, Albright A, et al.. Medical expenditures associated with diabetes among privately insured U.S. youth in 2007. Diabetes Care 2011; 34: 1097–1101.
20. Consumer price indexes—all urban consumers. Bureau of Labor Statistics, United States Department of Labor, 2011. Available at: http://www.bls.gov/cpi/home.htm
. Accessed December 15, 2011).
21. Centers for Disease Control and Prevention. Sexually Transmitted Disease Treatment Guidelines, 2010. Atlanta, GA: Centers for Disease Control and Prevention, 2011.
22. Xu F, Markowitz LE, Sternberg MR, et al.. Prevalence of circumcision and herpes simplex virus type 2 infection in men in the United States: The National Health and Nutrition Examination Survey (NHANES), 1999–2004. Sex Transm Dis 2007; 34: 479–484.
23. Henderson T, Shepheard J, Sundararajan V. Quality of diagnosis and procedure coding in ICD-10
administrative data. Medical Care 2006; 44: 1011–1019.
24. Peabody JW, Luck J, Jain S, et al.. Assessing the accuracy of administrative data in health information systems. Med Care 2004; 42: 1066–1072.