High SSI Rate? Consider the Procedure
High SSI Rate? Consider the Procedure
Saeed MJ, Dubberke ER, Fraser VJ, Olsen MA
Am J Infect Control. 2015;43:617-623
A very large, multistate analysis of surgical-site infection (SSI) rates using administrative data confirms the fear that the considerable variation in SSI risk by operative procedure and hospital type is not currently accounted for in the National Healthcare Safety Network (NHSN) reporting.
Saeed and colleagues recently reported on their retrospective cohort study using the State Inpatient Database from New York, California, and Florida from January 2009 to September 2011, with which they compared 90-day SSI rates within five relatively heterogeneous NHSN-defined operative procedure categories: amputation surgery; bile duct, liver, or pancreas surgery; breast surgery; colon surgery; and hernia surgery. Within each NHSN operative procedure grouping of ICD-9-CM codes, more specific categories were created using ICD-9-CM procedure codes. Specific categories were also used to represent combination procedures (eg, colon surgery with small bowel surgery), although NHSN dictates a specific hierarchy for assigning infections to only one of those procedure types. Infections were classified in the administrative dataset using specific ICD-9-CM codes within 90 days of the index surgery.
Their analysis demonstrated that the overall incidence of SSI in an NHSN category differed from the SSI incidence of individual subgroups. They also found significant variation within each NHSN surgical category. For example, the overall SSI incidence after colon surgery was 19.2% but the range was 7.6% (for laparoscopic colectomy) to 52.5% (for open colectomy, with or without other bowel surgery plus colostomy). Colon surgeries that were combined with other surgeries (eg, open colectomy with small bowel surgery) had a much higher infection rate (29.3%) compared with the overall SSI rate (19.2%). Within some groupings, teaching hospitals also completed a disproportionate number of surgery types (80% of all pancreas surgeries).
Although this study relied on administrative data to classify infections rather than using Centers for Disease Control and Prevention-defined criteria, this large-scale, robust analysis highlights a critical limitation with the NHSN risk adjustment and reporting for SSI. Of importance, SSI after colon surgery and abdominal hysterectomy is publically reported, and starting in 2016, hospitals will be financially penalized by the Centers for Medicare & Medicaid Services if they rank in the highest quartile of infection incidence.
Currently, very few procedural factors are included in NHSN's risk adjustment—only duration and endoscope use for some procedure categories. The disparate SSI rates by specific procedure type reported here underscore the need for improved risk adjustment, as the surgical volume and procedure mix can lead to very different risks for each hospital, making fair interhospital comparisons unattainable.
Abstract
Procedure- Specific Surgical Site Infection Incidence Varies Widely Within Certain National Healthcare Safety Network Surgery Groups
Saeed MJ, Dubberke ER, Fraser VJ, Olsen MA
Am J Infect Control. 2015;43:617-623
Surgical-Site Infections and Operative Procedure Categories
A very large, multistate analysis of surgical-site infection (SSI) rates using administrative data confirms the fear that the considerable variation in SSI risk by operative procedure and hospital type is not currently accounted for in the National Healthcare Safety Network (NHSN) reporting.
Saeed and colleagues recently reported on their retrospective cohort study using the State Inpatient Database from New York, California, and Florida from January 2009 to September 2011, with which they compared 90-day SSI rates within five relatively heterogeneous NHSN-defined operative procedure categories: amputation surgery; bile duct, liver, or pancreas surgery; breast surgery; colon surgery; and hernia surgery. Within each NHSN operative procedure grouping of ICD-9-CM codes, more specific categories were created using ICD-9-CM procedure codes. Specific categories were also used to represent combination procedures (eg, colon surgery with small bowel surgery), although NHSN dictates a specific hierarchy for assigning infections to only one of those procedure types. Infections were classified in the administrative dataset using specific ICD-9-CM codes within 90 days of the index surgery.
Their analysis demonstrated that the overall incidence of SSI in an NHSN category differed from the SSI incidence of individual subgroups. They also found significant variation within each NHSN surgical category. For example, the overall SSI incidence after colon surgery was 19.2% but the range was 7.6% (for laparoscopic colectomy) to 52.5% (for open colectomy, with or without other bowel surgery plus colostomy). Colon surgeries that were combined with other surgeries (eg, open colectomy with small bowel surgery) had a much higher infection rate (29.3%) compared with the overall SSI rate (19.2%). Within some groupings, teaching hospitals also completed a disproportionate number of surgery types (80% of all pancreas surgeries).
Viewpoint
Although this study relied on administrative data to classify infections rather than using Centers for Disease Control and Prevention-defined criteria, this large-scale, robust analysis highlights a critical limitation with the NHSN risk adjustment and reporting for SSI. Of importance, SSI after colon surgery and abdominal hysterectomy is publically reported, and starting in 2016, hospitals will be financially penalized by the Centers for Medicare & Medicaid Services if they rank in the highest quartile of infection incidence.
Currently, very few procedural factors are included in NHSN's risk adjustment—only duration and endoscope use for some procedure categories. The disparate SSI rates by specific procedure type reported here underscore the need for improved risk adjustment, as the surgical volume and procedure mix can lead to very different risks for each hospital, making fair interhospital comparisons unattainable.
Abstract
Source...