Changes in Utilization of Bariatric Surgery in the United States From 1993 to 2016

Guilherme M. Campos, MD, PhD; Jad Khoraki, MD; Matthew G. Browning, PhD; Bernardo M. Pessoa, MD; Guilherme S. Mazzini, MD, PhD; Luke Wolfe, MS


Annals of Surgery. 2020;271(2):201-209. 

In This Article


Over 1.9 million inpatient primary bariatric surgeries were performed in the United States between 1993 and 2016. We estimated an 18.9-fold increase in the annual numbers of bariatric surgeries performed across this period that paralleled significant changes in surgical techniques, hospital settings, patient characteristics, and admission outcomes. Importantly, perioperative complication and mortality rates are now at an all-time low. The present study's results reinforce the relatively low risks of serious perioperative complications obtained with current techniques of bariatric surgery in the United States and corroborate recent findings that complication rates associated with bariatric surgery are low when performed in the appropriate setting.[2,28,29] These improvements in safety outcomes are a reflection of multiple factors including improved patient selection, anesthetic and surgical technique refinements, and dissemination of training for operative and perioperative care.[6] In addition, access and types of bariatric surgeries changed significantly, most noteworthy has been the almost complete uptake of laparoscopic techniques by the specialty. Other significant changes were related to techniques: the decrease in utilization of the VBG in the late 1990s, and then laparoscopic AGB around 2012, the persistent minimal utilization of BPD-DS thorough the years, a resilient long-term relatively large proportion of RYGB, both in the open and laparoscopic era; and the recent fast surge in utilization of laparoscopic SG. Although most studies comparing perioperative complications of the current 2 most commonly used operative techniques in the United States (laparoscopic SG and RYGB) have shown a slightly higher rate of complication with RYGB,[30] other large systematic reviews suggested that severe complications rates, including leak rates, may be similar.[31] Nonetheless, the practicing surgeon can and should chose in between the available procedures based on individual patient risk-assessment and by tailoring to specific patient characteristics and associated diseases. For example, although the laparoscopic SG does not involve a gastrojejunal anastomosis and thus carries no to little long-term risk of a marginal ulceration or hyperinsulinemic hypoglycemia; laparoscopic SG is associated with significantly inferior long-term control and higher rates of de novo gastroesophageal reflux disease than RYGB.[32,33] In addition, de novo Barrett's esophagus has now been reported to occur in 17% of patients 5 years after LSG,[34] whereas RYGB has been shown to provide superior control of GERD and, in 2 case series, prevent progression of Barrett's esophagus.[35,36]

Nevertheless and despite the advances in perioerative safety with bariatric surgery and well-documented benefits of bariatric surgery in most patients (eg, sustained and meaningful weight loss, significant improvements in quality of life, partial or complete remission of multiple obesity-related chronic diseases, and reduced risks of premature mortality), surgery utilization in the eligible adult population has only marginally increased over the last quarter-century and was estimated at 0.5% in 2016. To our knowledge, this study is the first to report trends in utilization of bariatric surgery in the estimated total eligible population.

The present findings highlight a low utilization of bariatric surgery that has persisted for a quarter-century. The reasons behind the relatively low utilization are not fully understood, but removing barriers to availability and access to bariatric surgery has the potential to improve quality of life, cardiometabolic and mental health, and life expectancy in many thousands (if not millions) of individuals. It is important to acknowledge that bariatric surgery should only be considered after a careful selection process and cannot be used in all possible candidates based on BMI and associated disease criteria alone. Current guidelines from US national societies[37] recommend that individuals should be considered candidates for surgery after counseling, testing, and confirmation of appropriate performance status for surgery, willingness to change eating habits and lifestyle, and stable psychosocial conditions and support. These criteria in and of themselves may disqualify many individuals from being considered for surgery.

Nevertheless, a vast and long list of other suspected barriers to access may prevent possible patients from even being evaluated for bariatric surgery and possibly other therapies offered by obesity specialists. Many potential referring practitioners overestimate surgical risks and have a poor understanding of the potential for surgery to ameliorate existing comorbidities.[38–40] This then contributes to poor communication of bariatric surgery options to patients by primary care, which some evidence suggests is a central barrier to utilization.[41] Given the important role of a physician referral in the patient's decision-making process to pursue bariatric surgery,[42] improved awareness of current safety profile of bariatric surgery is needed and may help allay referring practitioners' concerns.

Another important barrier for greater utilization seems to be variable coverage for bariatric surgery across commercial, state, and federal insurance payers and programs across geographical regions.[39,40] Some payers' resistance to cover bariatric procedures seems to partly stem from questions regarding the cost-effectiveness of bariatric surgery. However, several publications with various analytical approaches have consistently shown that bariatric surgery is cost-effective over the longer term for patients with severe obesity and for specific obesity-related comorbidities such as T2D,[43–46] nonalcoholic steatohepatitis (NASH) and NASH cirrhosis[47,48] degenerative joint disease,[49] among others. In the United States, estimated incremental cost-effectiveness ratios (ICERs) of bariatric surgery range between $6,000 and $25,000 across different models and patient populations, including those with T2D.[43,46,50–53] Our study showed significant reduction in the cost of admission for bariatric surgery likely due to the decrease of length of hospital stay, substantial reductions in the rates of complications occurring during the initial admission for surgery, among other reasons. Our estimate of median bariatric surgery hospitalization cost of around $12,900 is lower than the median cost of $17,800 reported for ventral hernia repair using the 2014 NIS data with similar methods to ours.[54] Kim et al[53] predicted that both commercial payers and government payers would accrue greater returns on their investment if they eliminated the cost-sharing for patient with T2D and a BMI of 40 kg/m2 or higher. Such an expansion in coverage at the population level was also associated with an additional 70.7 QALYs gained and an estimated $7 million in cost-savings.[53]

One other important finding of our study was a possible racial disparity in the utilization of bariatric surgery. Although we observed an increasing utilization of bariatric surgery in minorities and individuals covered by state or federal insurance programs, others suspected that minorities' racial groups (blacks and Hispanics) may not have the same access to bariatric surgery as whites.[55] The present study's results are in agreement with past reports that the patient population in general that receives bariatric surgery is composed of approximately 70% non-Hispanic whites, 15% non-Hispanic blacks, and 10% Hispanics. Although these proportions approximate the racial make-up of the general population as a whole (61% non-Hispanic white, 13% non-Hispanic black, and 18% Hispanic),[56] current best estimates indicate that the age-adjusted prevalence of severe obesity is higher in non-Hispanic blacks (11.8%) and Hispanics (8%) than non-Hispanic whites (7.5%).[12] Thus, as minority groups (non-Hispanic blacks and Hispanics) have a greater prevalence of class II and III obesity than non-Hispanic whites,[12,57] it is possible that the past and also current utilization rates of bariatric surgery by minorities is lower than what would be expected in view of greater obesity and metabolic complication rates in minority groups.[58]

Although the present study is the first to report on the evolution of bariatric surgery since adoption of the NIH guidelines, certain limitations are intrinsic to national databases. First, we were unable to account for possible miscoding of or missing data points in the NIS, particularly race groups. Also, the NIS only includes inpatient bariatric procedure and does not provide information regarding outpatient procedures. Although outpatient surgery was indeed more common in between 2008 to 2012 during the years when laparoscopic AGB was still frequently used, the American Society of Metabolic and Bariatric Surgery estimated that well over 90% of all procedures were performed in the inpatient setting in 2016.[15] Therefore, further inclusion of only primary outpatient procedures is expected to minimally impact our recent (under)utilization estimates, particularly when considering our conservative estimates of the eligible population with class II obesity as those with T2D, hypertension, and/or hypercholesterolemia. Of note, the estimates of number of surgery per year from the ASMBS differ from our study as the NIS data provide estimates for inpatient primary bariatric surgeries only, whereas the ASMBS estimates includes inpatient and outpatient surgeries, revisions, and endoscopic procedures such as intragastric balloons.[15] Although the ASMBS estimates are partially based on the NIS, it also draws information from the American College of Surgeons Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program and other quality improvements datasets. Lastly, and as pointed out above, it must be underscored that morbidity and mortality rate in any NIS analyses only represent complications and mortality that occurs during the index admission for surgery. In other words, the current NIS methodology and design do not capture or account for complications, mortality, and other admissions after initial discharge.

In conclusion, perioperative index admission safety of bariatric surgery has improved significantly from 1993 to 2016, alongside changes in types, number and cost of surgeries, characteristics of patients having surgery, and characteristics and location of hospitals where these surgeries are done. Although the number of surgeries has increased, utilization of surgery in the growing number of individuals that are considered eligible for surgery has remained low. Studying and addressing barriers to utilization may allow for greater access to surgical therapy.