BACKGROUND: Public reporting and reduced Medicare payments because of health care-associated infections have resulted in the consideration of administrative discharge codes as markers of health care-associated infections. This study aims to determine whether specific secondary ICD-9-CM infection codes linked to cases from a large data set of surgical procedures are predictors of surgical site infections (SSIs). METHODS: All patients undergoing 1 of 9 surgical procedures from January 1, 2005, through December 31, 2005, at a large academic medical center and who were assigned a secondary ICD-9-CM infection code at discharge were eligible for study inclusion. All cases were reviewed to determine the presence of SSIs. Logistic regression was used to determine which secondary codes were predictors of SSIs. RESULTS: Among 75 secondary infection codes applied at discharge to 454 patients, only 1 code (998.59) appeared to be reliably associated with SSIs. Two other general infection codes (996.63 and 996.67) and 1 specific infection code (320.3) may also have utility. CONCLUSION: Administrative coding data do not perform well to identify SSIs. Some general secondary infection codes, however, may have the potential to be utilized in screening algorithms of electronic health data to assist in SSI surveillance.
BACKGROUND: Public reporting and reduced Medicare payments because of health care-associated infections have resulted in the consideration of administrative discharge codes as markers of health care-associated infections. This study aims to determine whether specific secondary ICD-9-CM infection codes linked to cases from a large data set of surgical procedures are predictors of surgical site infections (SSIs). METHODS: All patients undergoing 1 of 9 surgical procedures from January 1, 2005, through December 31, 2005, at a large academic medical center and who were assigned a secondary ICD-9-CM infection code at discharge were eligible for study inclusion. All cases were reviewed to determine the presence of SSIs. Logistic regression was used to determine which secondary codes were predictors of SSIs. RESULTS: Among 75 secondary infection codes applied at discharge to 454 patients, only 1 code (998.59) appeared to be reliably associated with SSIs. Two other general infection codes (996.63 and 996.67) and 1 specific infection code (320.3) may also have utility. CONCLUSION: Administrative coding data do not perform well to identify SSIs. Some general secondary infection codes, however, may have the potential to be utilized in screening algorithms of electronic health data to assist in SSI surveillance.
Authors: Mohammad R Rasouli; Mitchell Gil Maltenfort; James J Purtill; William J Hozack; Javad Parvizi Journal: Clin Orthop Relat Res Date: 2013-10 Impact factor: 4.176
Authors: Matthew T Eisenberg; Andrew M Block; Matthew L Vopat; Margaret A Olsen; Jeffrey J Nepple Journal: J Pediatr Orthop Date: 2022-04-01 Impact factor: 2.324
Authors: David K Warren; Katelin B Nickel; Anna E Wallace; Daniel Mines; Victoria J Fraser; Margaret A Olsen Journal: Infect Control Hosp Epidemiol Date: 2014-10 Impact factor: 3.254
Authors: Margaret A Olsen; Kelly E Ball; Katelin B Nickel; Anna E Wallace; Victoria J Fraser Journal: Infect Control Hosp Epidemiol Date: 2016-12-15 Impact factor: 3.254