SSAT - EFFECT OF ACADEMIC STATUS ON OUTCOMES OF COLORECTAL SURGERY

Back to 2017 Posters

EFFECT OF ACADEMIC STATUS ON OUTCOMES OF COLORECTAL SURGERY
Kristen Cagino^*2, Maria Altieri³, Jie Yang¹, Lizhou Nie¹, Paula Denoya³, Mark A. Talamini³, Aurora D. Pryor³
¹Department of Applied Mathematics and Statistics, Stony Brook Medicine, Stony Brook, NY; ²School of Medicine, Stony Brook University, Stony Brook, NY; ³Department of Surgery, Stony Brook University Medical Center, Stony Brook, NY

There is a growing debate regarding whether academic hospital status affects outcomes of surgical procedures including colorectal surgery for rectal and anal cancers. The purpose of our study is to investigate surgical outcomes at academic versus community institutions in terms of laparoscopic versus open procedure, surgery type, hospital volume, and stoma formation.
The Statewide Planning And Research Cooperative System administrative database was used to identify patients undergoing Abdominoperineal resection (APR) and Low Anterior Resection (LAR) between 2009-2014 through the use of ICD-9 codes. Linear mixed models were used to compare length of stay (LOS) and generalized linear mixed models were used to compare all other outcomes after controlling for confounding factors. Laparoscopic versus open procedures, surgery type and stoma formation between academic and community facilities were compared based on multivariable logistic regression models after controlling for age, gender and any comorbidity.
There were 15,879 procedures performed during this time period. Univariate analysis demonstrated that academic centers performed more laparoscopic procedures, APR, less stoma formation, and had more high volume facilities (P-value<0.01). After adjusting for confounding factors, academic facilities were more likely to perform APR surgeries (OR 1.35, 95% CI:1.04-1.74, P-value=0.0235). No significant difference was observed between academic and community facilities in terms of laparoscopic versus open procedures and stoma formation (p-value>0.05). Clinical outcomes for a specific surgery type, laparoscopic versus open procedure, or having stoma formation were similar between academic and community facilities, except for longer LOS after stoma formation (ratio in LOS+1=1.06 with 95% CI:1.00-1.13) or LAR at academic facilities (ratio in LOS+1=1.05 with 95% CI:1.00-1.11). Multivariable regression models suggest that at academic facilities stoma formation and open surgery were associated with longer LOS, greater chance of in-hospital death, 30-day readmission, ED visit, and complications. APR was associated with longer LOS, 30-day readmission and ED visit, but lower in-hospital death and complications at academic facilities (Table 1). Similar association between surgical outcomes and stoma formation, laparoscopic versus open procedure or surgery types were found at community facilities with the exceptions that open surgery was not significantly associated 30-day readmission and stoma formation was not significantly associated with bigger chance of 30-day ED visits (Table 1).
Univariate analysis shows that more laparoscopic procedures, APR, less stoma formations are performed at academic centers. Patient selection may play a role as following multivariable regression there were no differences between procedures performed at academic and community hospitals.

Effect of Stoma Formation, Laparoscopic versus Open Procedure, and Surgery Type on Outcome measures at Academic and Community facilities
Variable Specific level 30 day ED visits 30 day readmission Overall Complication In hospital death LOS
odds ratio (95% CI) p-value odds ratio (95% CI) p-value odds ratio (95% CI) p-value odds ratio (95% CI) p-value Ratio in LOS+1 p-value
Stoma formation (Yes vs No) Academic Facility 1.21 (1.06, 1.37) 0.2543 1.17 (1.03, 1.34) 0.8183 2.88 (2.55, 3.26) 0.1003 2.49 (1.97, 3.16) 0.2097 1.33 (1.29, 1.37) 0.0433
Community Facility 1.08 (0.95, 1.24) 1.15 (1.00, 1.32) 3.34 (2.95, 3.78) 2.03 (1.65, 2.51) 1.27 (1.24, 1.31)
Procedure (Laparoscopic vs Open) Academic Facility 0.83 (0.73, 0.95) 0.8318 0.73 (0.64, 0.84) 0.0467 0.42 (0.37, 0.46) 0.361 0.31 (0.19, 0.51) 0.6475 0.81 (0.79, 0.84) 0.1556
Community Facility 0.85 (0.73, 0.99) 0.91 (0.77, 1.07) 0.45 (0.40, 0.51) 0.36 (0.23, 0.58 0.79 (0.76, 0.82)
Surgery Type (APR vs LAR) Academic Facility 1.31 (1.16, 1.49) 0.2278 1.41 (1.24, 1.60) 0.1622 0.77 (0.69, 0.86) 0.0178 0.32 (0.21, 0.50) 0.3297 1.05 (1.02, 1.08) 0.2501
Community Facility 1.48 (1.27, 1.73) 1.63 (1.39, 1.92) 0.63 (0.55, 0.72) 0.44 (0.28, 0.70) 1.08 (1.04, 1.11)

*The covariates included in the multivariable regression models were surgery year, age, gender, race, hospital volume, insurance, any comorbidity, any complication and geographical health regions. P-values suggest if the relationships between variable and each clinical outcome were the same between academic facility and community facility.

Back to 2017 Posters