Public bariatric surgery in Tasmania : patient pathways and health service use

Background: Many patients eligible for bariatric surgery in Australia do not have private health insurance, which creates significant pressure on the public system, with prolonged wait-list times. Public hospital service use by this patient group is under-investigated. Australia-wide, public hospitals perform a higher proportion of revisional procedures than private hospitals (36% vs 25%), possibly limiting access to primary procedures. The aim of the study was to investigate the impact of bariatric surgery provision in the Tasmanian public sector on public hospital service use, particularly hospital admissions and emergency department (ED) presentation rates and to identify and describe revisional surgery pathways, including subsequent re-revisions. Methods: A statewide retrospective cohort study was conducted of public hospital service use by all Tasmanian patients on the wait-list for publicly funded bariatric surgery from 2008 to 2013. Multiple administrative databases and data linked with the Tasmanian Death Registry were used. Rates of hospital admissions in 2006‚-2014 and ED presentations in 2000‚-2014 were compared for operated-on patients and those who dropped-out of the wait-list. Public hospital service use was analysed in different periods: prior to wait-list placement, while on the wait-list, and after removal from the wait-list either after having a bariatric procedure or dropping out without surgery. Incidence rate ratios (IRR) with 95% confidence intervals (CI) for groups and periods comparisons were derived using a negative binomial regression mixed-effects model adjusted for sex, age and non-independent observation periods. Hospital service use was analysed for primary vs revisional bariatric surgery recipients using similar methods. A systematic review of revisional surgery outcomes, such as subsequent revisions and complication rates, was performed, including papers following at least 75% of patients for 12 months or more. Results: The cohort study identified 652 patients wait-listed for primary bariatric surgery, of whom 178 (27.3%) had bariatric surgery and 236 (36.2%) dropped-out from the wait-list. Together, these patients had 3,120 public hospital admissions and 5,149 ED presentations. Number of days in hospital per year was higher for the dropped-out patients than for surgery recipients while on the wait-list (IRR 2.22, 95% CI 1.36‚-3.61). Hospital admission rates did not increase post-surgery (IRR 1.08, 95% CI 0.83‚-1.41) but days admitted per year did increase (IRR 1.53, 95% CI 1.01‚-2.34). ED presentation rates did not change significantly post-surgery compared with the waiting period (IRR 1.19, 95% CI 0.90‚-1.56). Presentation rates significantly increased for digestive system (IRR 2.02, 95% CI 1.19‚-3.45) and psychiatric diseases (IRR 4.85, 95% CI 1.06‚-22.26) after surgery. The likelihood of being admitted from the ED significantly increased after surgery (31.7% to 38.9%, p<0.05). A total of 95 patients wait-listed for revisional surgery were identified; 91 (95.2%) of the patients were operated-on as planned, and two more had emergency surgeries after removal from the wait-list. Including subsequent planned and emergency revisions, of the 320 bariatric procedures performed with public funding for patients while wait-listed for bariatric surgery in 2008‚-2013, 142 (44.8%) were performed for revisional surgery. The mean primary surgery wait-list time was significantly (p<0.05) longer than for revisional surgery: 4.1¬±2.8 vs 0.8¬±1.0 years, respectively. Compared with primary-only surgery recipients, revisional surgery recipients had higher public hospital admission rates (IRR 2.60, 95% CI 1.63‚-4.13 while on wait-list and IRR 1.98, 95% CI 1.31‚-2.98 post-surgery); more days in hospital per year (IRR 2.68, 95% CI 1.44‚-4.99 while on wait-list and IRR 2.10, 95% CI 1.18‚-3.76 post-surgery); and higher ED presentation rates after the surgery (IRR 1.76, 95% CI 1.15‚-2.70). For the systematic review, the search identified 28 papers (1317 patients with surgical revisions) following at least 75% of patients for 12 months or more. For adjustable gastric banding (AGB), rebanding had higher re-revisional rates than conversions into other procedures. Conversion of AGB to Roux-en-Y gastric by-pass had the highest number of short-term (10.7%) and long-term (22.0%) complications. We estimated 8.8% of patients required tertiary surgery, with 194 reoperations (tertiary, subsequent and for complications) per 1000 patients undergoing a secondary procedure. Conclusion: Bariatric surgery performed in the public hospital setting in Tasmania was followed by an increase in hospital service use. Revisions represented nearly half of the public bariatric surgery procedures in Tasmania and were higher priorities. Revisional surgery in the public system, including revisions for primary surgery performed in the private system, limits access to public primary surgery. Future planning for bariatric surgery in the public sector should account for the increase in public hospital service use and demand for revisions, including tertiary and subsequent revisions and their long-term complications.