User login
SCOTTSDALE, AZ. – Level II trauma centers appeared to benefit most from attaining American College of Surgeons trauma verification, a retrospective analysis of a large national sample suggests.
Attaining ACS trauma consultation/verification was an independent predictor of patient survival in level II trauma centers (adjusted odds ratio, 1.26; P less than .01), but not in level I trauma centers (OR, 1.0; P = .84), which already performed well as a group.
Level II centers with state designation only were 3.5 times more likely to be poor performers based on survival outcomes than were level II centers that had attained ACS verification, Dr. Joshua Brown reported at the annual meeting of the Eastern Association for the Surgery of Trauma.
"Promoting ACS verification among level II trauma centers may represent an opportunity for improved outcomes in select trauma centers," he said.
A rigorous process that demands significant financial and personnel resources, ACS verification has been shown in single-center reports to improve outcomes. Yet less than 25% of level I and II trauma centers pursue ACS verification if already designated by their state as a trauma center, said Dr. Brown, a general surgery resident at the University of Pittsburgh Medical Center.
To compare mortality at ACS verified and state designated centers in a national sample, the investigators identified 900,274 patients, at least 16 years old, in the National Trauma Data Bank 2007-2008. Their mean age was 42 years, mean Injury Severity Score was 9, and 76% were transported from the trauma scene. The patients were admitted to 189 level I (65% ACS verified) and 185 level II trauma centers (67% ACS verified).
A mortality prediction model was constructed based on published Trauma Quality Improvement Program (TQIP) methodology, and multiple imputation used for missing data. The model had an area under the curve of 0.92, indicating excellent discrimination to predict death, Dr. Brown said.
Level I ACS centers had a significantly lower median observed-to-expected (O/E) mortality ratio than did level I state centers, (0.95 vs. 1.02; P less than .01).
No difference was observed in median O/E mortality ratio between level II ACS and state centers (0.94 vs. 0.87; P = .30), he said.
As noted earlier, however, level II state centers had more than three times the poor performers – high O/E ratio outliers, defined by an O/E and 90% confidence interval greater than 1.0 – than level II ACS centers (13% vs. 4%; OR 3.59; P = .03).
The proportion of good performers – low O/E outliers, defined by an O/E and C.I. less than 1.0 – was similar at state vs. ACS level II centers (21% vs. 23%;P = .57).
At level I trauma centers, no significant differences were observed between ACS-verified and state-verified centers in the proportion of high outliers (15% vs. 24%; P = .16) or low outliers (28% vs. 16%; P = .1), Dr. Brown reported.
Dr. Keith Clancy of York (Penn.) Hospital, who was invited to discuss the study, said, "It was a well-written, well-designed paper on a very important topic that will provide some pause for states that may inappropriately believe that simply by incorporating ACS standards into their state trauma standards they can automatically achieve the same outcomes as ACS verification."
Dr. Brown and Dr. Clancy reported no financial disclosures.
SCOTTSDALE, AZ. – Level II trauma centers appeared to benefit most from attaining American College of Surgeons trauma verification, a retrospective analysis of a large national sample suggests.
Attaining ACS trauma consultation/verification was an independent predictor of patient survival in level II trauma centers (adjusted odds ratio, 1.26; P less than .01), but not in level I trauma centers (OR, 1.0; P = .84), which already performed well as a group.
Level II centers with state designation only were 3.5 times more likely to be poor performers based on survival outcomes than were level II centers that had attained ACS verification, Dr. Joshua Brown reported at the annual meeting of the Eastern Association for the Surgery of Trauma.
"Promoting ACS verification among level II trauma centers may represent an opportunity for improved outcomes in select trauma centers," he said.
A rigorous process that demands significant financial and personnel resources, ACS verification has been shown in single-center reports to improve outcomes. Yet less than 25% of level I and II trauma centers pursue ACS verification if already designated by their state as a trauma center, said Dr. Brown, a general surgery resident at the University of Pittsburgh Medical Center.
To compare mortality at ACS verified and state designated centers in a national sample, the investigators identified 900,274 patients, at least 16 years old, in the National Trauma Data Bank 2007-2008. Their mean age was 42 years, mean Injury Severity Score was 9, and 76% were transported from the trauma scene. The patients were admitted to 189 level I (65% ACS verified) and 185 level II trauma centers (67% ACS verified).
A mortality prediction model was constructed based on published Trauma Quality Improvement Program (TQIP) methodology, and multiple imputation used for missing data. The model had an area under the curve of 0.92, indicating excellent discrimination to predict death, Dr. Brown said.
Level I ACS centers had a significantly lower median observed-to-expected (O/E) mortality ratio than did level I state centers, (0.95 vs. 1.02; P less than .01).
No difference was observed in median O/E mortality ratio between level II ACS and state centers (0.94 vs. 0.87; P = .30), he said.
As noted earlier, however, level II state centers had more than three times the poor performers – high O/E ratio outliers, defined by an O/E and 90% confidence interval greater than 1.0 – than level II ACS centers (13% vs. 4%; OR 3.59; P = .03).
The proportion of good performers – low O/E outliers, defined by an O/E and C.I. less than 1.0 – was similar at state vs. ACS level II centers (21% vs. 23%;P = .57).
At level I trauma centers, no significant differences were observed between ACS-verified and state-verified centers in the proportion of high outliers (15% vs. 24%; P = .16) or low outliers (28% vs. 16%; P = .1), Dr. Brown reported.
Dr. Keith Clancy of York (Penn.) Hospital, who was invited to discuss the study, said, "It was a well-written, well-designed paper on a very important topic that will provide some pause for states that may inappropriately believe that simply by incorporating ACS standards into their state trauma standards they can automatically achieve the same outcomes as ACS verification."
Dr. Brown and Dr. Clancy reported no financial disclosures.
SCOTTSDALE, AZ. – Level II trauma centers appeared to benefit most from attaining American College of Surgeons trauma verification, a retrospective analysis of a large national sample suggests.
Attaining ACS trauma consultation/verification was an independent predictor of patient survival in level II trauma centers (adjusted odds ratio, 1.26; P less than .01), but not in level I trauma centers (OR, 1.0; P = .84), which already performed well as a group.
Level II centers with state designation only were 3.5 times more likely to be poor performers based on survival outcomes than were level II centers that had attained ACS verification, Dr. Joshua Brown reported at the annual meeting of the Eastern Association for the Surgery of Trauma.
"Promoting ACS verification among level II trauma centers may represent an opportunity for improved outcomes in select trauma centers," he said.
A rigorous process that demands significant financial and personnel resources, ACS verification has been shown in single-center reports to improve outcomes. Yet less than 25% of level I and II trauma centers pursue ACS verification if already designated by their state as a trauma center, said Dr. Brown, a general surgery resident at the University of Pittsburgh Medical Center.
To compare mortality at ACS verified and state designated centers in a national sample, the investigators identified 900,274 patients, at least 16 years old, in the National Trauma Data Bank 2007-2008. Their mean age was 42 years, mean Injury Severity Score was 9, and 76% were transported from the trauma scene. The patients were admitted to 189 level I (65% ACS verified) and 185 level II trauma centers (67% ACS verified).
A mortality prediction model was constructed based on published Trauma Quality Improvement Program (TQIP) methodology, and multiple imputation used for missing data. The model had an area under the curve of 0.92, indicating excellent discrimination to predict death, Dr. Brown said.
Level I ACS centers had a significantly lower median observed-to-expected (O/E) mortality ratio than did level I state centers, (0.95 vs. 1.02; P less than .01).
No difference was observed in median O/E mortality ratio between level II ACS and state centers (0.94 vs. 0.87; P = .30), he said.
As noted earlier, however, level II state centers had more than three times the poor performers – high O/E ratio outliers, defined by an O/E and 90% confidence interval greater than 1.0 – than level II ACS centers (13% vs. 4%; OR 3.59; P = .03).
The proportion of good performers – low O/E outliers, defined by an O/E and C.I. less than 1.0 – was similar at state vs. ACS level II centers (21% vs. 23%;P = .57).
At level I trauma centers, no significant differences were observed between ACS-verified and state-verified centers in the proportion of high outliers (15% vs. 24%; P = .16) or low outliers (28% vs. 16%; P = .1), Dr. Brown reported.
Dr. Keith Clancy of York (Penn.) Hospital, who was invited to discuss the study, said, "It was a well-written, well-designed paper on a very important topic that will provide some pause for states that may inappropriately believe that simply by incorporating ACS standards into their state trauma standards they can automatically achieve the same outcomes as ACS verification."
Dr. Brown and Dr. Clancy reported no financial disclosures.
AT THE ANNUAL MEETING OF THE EASTERN ASSOCIATION FOR THE SURGERY OF TRAUMA
Major Finding: ACS trauma verification was a significant independent predictor of survival at level II trauma centers (adjusted OR, 1.26; P less than .01), but not at level I centers (OR, 1.0; P = .84).
Data Source: Retrospective analysis of 900,274 patients in the National Trauma Data Bank.
Disclosures: Dr. Brown and Dr. Clancy reported no disclosures.
