Hospital Admissions Greater Than 30 Days Following Bariatric Surgery: Patient and Procedure Matter

Dana A Telem, MD, Wendy L Patterson, MPH, Brittany D Peoples, MS, Maria Altieri, MD, Gerald Gracia, MD, Jie Yang, PhD, Qiao Zhang, MS, Aurora D Pryor, MD. Stony Brook Medicine, New York State Department of Health Office of Quality and Patient Safety.

Introduction: Sparse data is available regarding hospital admission rates greater than 30-days following a bariatric procedure. This has important implications in terms of patient care, resource utilization assessment and healthcare expenditure. The purpose of this study is to assess hospital admissions for two-years following bariatric surgery in order to identify potential differences by procedure and at risk patient populations.

Methods: Following IRB approval, New York Statewide Planning and Research Cooperative System (SPARCS) administrative data was used to identify 22,139 adult patients who underwent bariatric surgery from 2006-2008. SPARCS is a comprehensive data reporting system which collects patient-level risk characteristics, treatments, and outcomes for all New York State (NYS) hospital discharges. Bariatric surgery was identified by discharges with a primary diagnosis of overweight or obesity and a primary procedure code for laparoscopic gastric band (LGB), laparoscopic Roux-en-y gastric bypass (RYGB) and laparoscopic sleeve gastrectomy (LSG). Identified bariatric patients were followed for two years after surgery to identify all-cause hospital admissions. Statistical analysis was performed via the chi square test and calculation of odds ratios.

Results: Of the 22,139 patients, 12,439 underwent RYGB, 9,099 LGB and 601 LSG. In total, 5,718(26%) patients were admitted within two years of surgery for a total of 9,502 readmissions. Thirty-day admission rate was 5%. Stratifying the 9,502 admissions demonstrated 14%  occurred within 30-days, 23% within 90-days, 32% within 6 months, 54% within one-year, and 76% within 18-months of surgery. The number of admissions per patient ranged from 1-22. Assessing the number of admissions per patient demonstrated that 3,741(17%) patients had one, 1,575(7%) had 2-3 and 402(2%) patients had greater than 4 admissions. By bariatric procedure, LSG had the highest admission rate(32%), followed by RYGB(29%) and then LGB(22%),p<0.001. The number of admissions by procedure follow: 1 (18% RYGB vs. 15% LGB vs. 20% LSG,p<0.001), 2-3 (8% RYGB vs. 6% LGB vs. 8% LSG,p<0.001), and 4+(2% RYGB vs. 1% LGB vs. 3% LSG,p<0.001). Table 1 lists significant admission predictors.

Conclusion: One out of four bariatric patients will be admitted to the hospital within 2 years of surgery. As such 30-day assessment underestimates hospital utilization. While the majority of patients are admitted only once, 7% of patients required more than 1 admission. Operative procedure significantly correlates with hospital admission. LSG is associated with both the highest rate as well as highest frequency of hospital admissions, followed by RYGB, then LGB. In addition, several factors were identified that significantly identify at risk patients. Consideration and attention to these factors are necessary for operative planning and postoperative monitoring.

Significant patient specific risk factors for hospital readmission within 2 years of bariatric surgery
Patient Factor Odds Ratio  95% Confidence Interval P-value
Black race1.21.1-1.3<0.001
Female gender1.21.1-1.3<0.001
Operative Complication1.71.5-1.9<0.001
Congestive heart failure1.81.1-2.90.025
Neurological disorders1.51.2-1.90.002
Chronic pulmonary disease1.31.2-1.4<0.001
Rheumatoid arthritis1.81.3-2.4<0.001
Substance abuse history2.01.1-3.50.022


« Return to SAGES 2014 abstract archive