Toward the Rational and Equitable Use of Bariatric Surgery

Bariatric surgery is the only health care intervention that facilitates significant and sustained weight loss.¹ Surgery results in remission of diabetes in 80% to 90% of obese patients with diabetes, and reduces the risk of death associated with obesity by nearly 30%.² More than 5% (at least 15 million)³ of the US adult population meets the criteria for obesity surgery (body mass index [BMI] ≥ 40 or BMI ≥ 35 with other conditions such as diabetes). In actuality, only a small fraction of this group is considered for and undergoes surgery (180 000 estimated bariatric surgery procedures in 2006).⁴ This mismatch between eligibility and receipt of surgical care is related to multiple factors, including the ways in which the health care system considers patients for surgery and delivers bariatric surgical care. Growing evidence also suggests that the cohort that does undergo surgery is not drawn evenly from the pool of candidates with extreme obesity. Understanding the factors involved in these mismatches is essential to developing a rational and equitable approach to bariatric surgery.

The demographic characteristics of patients who have bariatric surgery are not reflective of individuals with severe obesity in the United States. For example, nearly 84% of patients who undergo the surgery are women,⁵ more than 90% are white,⁶ and most have higher income levels.⁵

Conversely, individuals with low income levels and who are black or Hispanic⁷ are more likely to be obese and morbidly obese than the general population. Obesity among these groups may have even greater devastating social and clinical consequences. Despite this, Medicaid beneficiaries were one of the few groups receiving bariatric surgery less often⁵ during the last 10 years, and in some states few (or even no) bariatric centers perform surgeries on Medicaid beneficiaries. In certain regions of California, the wait times for bariatric surgery among public insurance enrollees were suspected to be more than 10 years.⁸ Similarly, blacks and Hispanics represent some of the most overweight of all racial/ethnic groups but are much less likely to undergo surgery, accounting for less than 10% of most operated cohorts.⁹ Men have a slightly lower rate of morbid obesity when compared with women (2.8% of men and 6.9% of women had morbid obesity in a recent assessment⁷ ) but 40% of individuals with morbid obesity are men¹⁰ while only about 15% of those having surgery are men.⁵ Surgery among patients older than 64 years is also less common (<1% of bariatric procedures)⁵ than would be expected from the distribution of obesity among older patients. Similarly, 16% of patients with obesity or morbid obesity are older than 60 years,¹⁰ but they comprise less than 4% of patients undergoing the surgery.¹⁰ In the past 10 years, the prevalence of more extreme levels of obesity (BMI ≥ 50) has increased 5-fold while lower levels of obesity have increased at a much less rapid rate.¹¹ It is therefore surprising that patients with BMI of 50 or greater account for a relatively small fraction of those having surgery and that the mean BMI of patients undergoing the surgery is 41 to 45.¹²

Clinicians are uncertain of the proper role of surgery for obesity and related conditions. In part, this reflects a debate about the cause and, by extension, the appropriate treatment of obesity. This clinical confusion is also related to the absence of a triage strategy for patients with obesity, varying patient and clinician perceptions of obesity, and the effectiveness and safety of bariatric procedures.

To varying degrees, clinicians adhere to the belief that obesity is a manifestation of overeating and underexercising and that behavioral modification should be the mainstay of treatment. Others believe that to some extent, genetic factors favor weight gain during periods of high-caloric exposure—the thrifty gene theory¹³ —and the concept of a “defended weight setpoint”¹⁴ and consider these barriers to behavioral modification approaches. For these clinicians, surgery is often viewed as an approach to resetting the defended weight, allowing behavioral changes to work more effectively in spite of genetic factors. The latter group may be more inclined to consider bariatric surgery for their patients.

For either group of clinicians, the BMI-based criteria may be part of the barrier for considering patients for bariatric surgery. Since there does appear to be a population of patients who are obese without associated health conditions (so-called healthy fat),¹⁵ some clinicians are skeptical that patients should be considered eligible for surgery based solely on their BMI. The finding that 80% to 90% of individuals with diabetes no longer need diabetes medications after bariatric surgery¹⁶ and that bariatric surgery reverses diabetes even among patients with a BMI¹⁷ of less than 35 suggest that the presence of certain clinical conditions should direct patients to surgery rather than BMI alone. In fact, the eligibility criteria based on BMI alone may have created a subset of patients who “meet the numbers” but not the clinical threshold for referral. Perhaps comorbid illness–based eligibility criteria will better help guide clinical referrals. Moreover, the confusion and inconsistency among clinicians about who should be directed to surgery may be tied to the fact that, like their patients (approximately 45% of men), a substantial group of US physicians are also overweight or obese and are likely dealing with obesity and weight control approaches themselves.¹⁸ For most individuals struggling with obesity, beliefs about treatment are linked to beliefs about cause.

Those who counsel and refer patients for bariatric surgery may have genuine uncertainty about who is the best candidate for an operation. In part, this reflects a knowledge gap; it is unclear who among the eligible 15 million obese patients has the most to gain from bariatric surgery. For example, while some patients are interested in having bariatric surgery because they hope it will help them live longer, a typical 40-year-old white woman with a BMI of 35 is predicted to lose only about 2 years of expected life related to obesity.¹⁹ The impact of extreme obesity on the life expectancy of a 60-year-old is more difficult to calculate and may be even less significant. While obesity may be among the leading sources of preventable deaths in the United States, it is unclear who will be among the healthy fat who live well into their 70s vs those who are at highest risk of preventable death, for whom weight loss of any kind is a true health imperative. Predicting who will lose some of their life expectancy due to obesity and if extension of life will be accomplished by surgery is a central part of the clinical decision-making process but not well supported by evidence.

In contradistinction, for most patients with end-stage liver failure, transplantation provides the only chance for survival in the proximate years and a rational prioritization scheme has been developed. In liver failure, this prioritization scheme (the model for end-stage liver disease [MELD] score) predicts death without transplantation. Those who have higher MELD scores are at higher risk for the consequences of the disease and therefore have higher priority for surgery. The health care community has no “MELD” for obesity and therefore cannot determine which patients are at highest risk for death (or functional, social, quality-of-life, and clinical decline) without significant weight loss. As such, there is widespread irregularity in the identification of at-risk patients and variation in who is considered and undergoes surgery. By helping identify individuals who have the most to gain from surgery, an obesity score based on the survival risk of nonsurgically treated obesity would be a helpful contribution. Similarly, a risk prediction strategy for bariatric surgery that helps determine who has the most to lose from surgery would help inform the rational prioritization of bariatric surgery.

The importance of both aspects of risk/success prediction becomes clear in evaluating the apparent inequity of surgical care among individuals with extreme obesity and the functionally disabled. Patients with the most extreme obesity are often socially and financially dependent on society, and returning them to a functional status would be of tremendous societal and individual benefit. Conversely, patients with extreme, disabling obesity also may have end-organ damage that may increase perioperative risk,²⁰ and it is unclear if this damage is reversible enough to warrant surgical intervention.

With better predictive scoring of both obesity risk and surgical risk, clinicians could be expected to rationally balance these competing harms. This more rational approach may confound the current care climate that expects “zero” adverse outcomes from bariatric surgical procedures. For example, individuals with extreme disability may have the most to gain in functional recovery and the most to lose (highest surgical risk) from a bariatric procedure. An obesity score and a surgical risk score would help objectively inform what now is addressed through the varying opinions of the clinical community. While such a risk stratification system for surgery is currently in development through the National Institutes of Health–sponsored Longitudinal Assessment of Bariatric Surgery,²¹ an obesity risk score has yet to be created.

In addition to reimbursement issues, bariatric surgeons may have attitudinal barriers that restrict the use of surgery among the poor. When surveyed about hypothetical scenarios, surgeons reported they were less likely to recommend surgery to patients receiving public insurance or who had less social support relative to those with private insurance.²² Maintaining the appropriate payer mix is the focus of seminars, professional development exercises, and part of the daily experience of surgeons in many care delivery systems. While this appropriate payer mix may be essential to financial success in surgical practices, it undoubtedly restricts access to care for individuals with less favorable insurance coverage. In other clinical arenas, differences in treatment based on race have been identified,²³ yet in this survey,²² patient race was not a predictor of recommended obesity surgery.

Surgeons at centers that perform bariatric surgery may also be targeting lower-risk patients and patients who are less well insured for other reasons. Given the climate of increased scrutiny of outcomes and increasing malpractice claims against bariatric surgeons, some appear to be less inclined to solicit and recruit higher-risk patients (extreme obesity, older patients, those with severe comorbidities). In several regions, higher-risk patients are referred to medical centers with a practice mission that focuses on community service or to clinicians who are more insulated from malpractice claims.

In part, different demographic groups may undergo bariatric surgery less often because they have less interest in the surgery. This may include more favorable attitudes about obesity among some racial or ethnic groups²⁴ and more resistance to weight loss interventions among men.²⁵ Other behavioral and social factors may make patients less inclined to pursue surgery. Some patients who are dependent on others (eg, homeless, disabled, alcohol dependent, and drug dependent) may not have the appropriate social structures and stability to pursue or effect the lifelong behavioral, dietary, and supplement-related changes required by the surgical procedures. Other groups of eligible patients may be disinclined to consider surgery because of their perception of risk and perhaps media portrayal of some adverse outcomes from the surgery. The decision to consider bariatric surgery after a lifetime of struggle with obesity is uniquely personal and the barriers to that consideration are complex.

Fewer men, black and Hispanic individuals, and individuals with low incomes undergo bariatric surgery than would be expected given the similar or more common prevalence of obesity. Those who appear to be most affected by extreme obesity do not receive surgery before healthier, better-insured, and lower-risk patients. These disparities appear to be the result of gaps in clinician knowledge and structural and attitudinal barriers. In trying to move toward a more enlightened system, the health care community needs to address the lack of a prioritization strategy for patients whose lives are most affected by extreme obesity and to counter a set of incentives and disincentives that favor operating on individuals who are at lower risk and better insured. The cultural differences and gaps in knowledge and belief that make certain groups less inclined to consider surgery should be better characterized so that effective educational and outreach programs can be created. Given the large number of US residents eligible for bariatric surgery and the significant health care costs and safety issues of both untreated obesity and obesity surgery, the lack of an equitable and rational system for delivering care represents an ethical and public health dilemma.