171 Estimating the Inpatient Costs of Group A Streptococcus—An Application Utilizing a National Dataset

Monday, June 10, 2013
Exhibit Hall A (Pasadena Convention Center)
Danielle S Abraham , New York State Department of Health, Albany, NY
Millicent Eidson , University of Albany School of Public Health, Rensselaer, NY
Greg P Giambrone , New York State Department of Health, Albany, NY
Ashley E Giambrone , University of Albany School of Public Health, Rensselaer, NY
Shelley M Zansky , University of Albany School of Public Health, Rensselaer, NY

BACKGROUND:   While most illnesses associated with Group A Streptococcus (GAS) infections such as pharyngitis are treated in the outpatient setting, some illnesses can require hospitalization.  Unlike the outpatient setting, little is known about the national inpatient cost of GAS or the population requiring hospitalization for GAS.

METHODS:   Data was examined from the 2002-2009 Nationwide Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality.  NIS is the largest all-payer, publically available hospital discharge dataset. Individuals with a GAS diagnostic code were assigned to severity levels.  Superficial and upper respiratory tract infections were classified as level one (L1) severity. Level three (L3) consisted of necrotizing fasciitis or Streptococcal toxic shock syndrome diagnoses.  Other presentations were assigned to level two (L2).  Costs adjusted to 2011 US dollars and length of stay (LOS) were examined based on severity and procedures performed (e.g., amputation, ventilation). Demographics, risk factors, and outcomes were assessed, stratified by severity.

RESULTS:   From 2002-09, GAS hospitalization costs exceeded $1.6 billion.  L1 and L2 hospitalizations accounted for the majority of overall costs.  Per discharge, average cost (L1-$9,296, L2-$15,138, L3-$35,923), LOS (L1-5.6 days, L2-8.3 days, L3-13.8 days), and adverse outcomes (e.g., death, discharge to another facility) significantly increased with severity. All procedures significantly increased cost, with ventilation procedures causing the largest increase.  No significant time trend (p=0.09) was seen in the annual average cost per discharge.  L1 discharges were significantly more likely to be younger (mean=38.6 years), male (59.2%), and covered by Medicaid (22.1%) or self pay (12.6%).  L2 discharges were significantly more likely to be older (mean=49.8 years), and Medicare recipients (38.2%).  L3 discharges had an average age of 45.8 years.  Comorbidities were significantly higher in groups L2 and L3.

CONCLUSIONS:   L3 discharges incurred the highest cost per discharge; however, the large overall costs associated with L1, a group with presentations similar to those seen in outpatient settings, is concerning.  Demographics and LOS suggest that a segment of this group may be using the hospital as their primary source of care and/or delaying care until complications arise.  The study supports the findings in the literature that older age and comorbidities are risk factors for severe GAS.  One limitation to this analysis is NIS does not include prescription data.  If prescription, post-discharge care (e.g., rehab facility care), and nonmedical costs were considered, inpatient GAS would be substantially more costly.