INTRODUCTION

How should primary care physicians and other clinicians counsel patients with obesity about the potential benefits and risks of weight loss? Because over half of US adults are classified as either overweight or obese according to body mass index (BMI),1,2 conversations about obesity are a routine part of primary care practice, yet much controversy exists regarding the evidence for strategies to guide these discussions. The 2018 United States Preventative Task Force (USPSTF) recommendations and the Health at Every Size (HAES) model of care present conflicting advice about the management of obesity. The USPSTF gives a grade B (moderate certainty of moderate net benefit) to their recommendation to offer lifestyle interventions that target at least 5% weight loss to adults with a BMI ≥ 30 kg/m2, or a BMI of 25–30 kg/m2 with hypertension, dyslipidemia, pre-diabetes, or diabetes.3 The primary clinical benefits cited to support these recommendations are a reduced incidence of type 2 diabetes in individuals with pre-diabetes, and the potential for clinically significant improvements in weight status with behavioral interventions.

In contrast, the HAES approach promotes weight inclusivity, which rejects the pathologizing of specific weights as inherently unhealthy.4 HAES treatment principles promote health improvement through enjoyable physical activity and eating for well-being without a focus on weight loss. Proponents of HAES cite improvements in quality of life, psychological well-being, eating behaviors, and aerobic capacity with weight-neutral interventions.5,6 The conflict between the weight-centric USPSTF recommendations and weight-neutral HAES principles should give pause to primary care clinicians who wish to practice patient-centered, evidence-based care. In this perspective, we describe what we term a “weight-skeptical” approach to the care of patients with obesity that offers an alternative to these guidelines by (1) acknowledging the potential harms of a weight-centric paradigm, (2) focusing on weight-neutral lifestyle and/or pharmacologic interventions when indicated for most patients with obesity, but also (3) accepting a role for weight loss interventions such as metabolic surgery and anti-obesity medications (AOM) in individuals at high risk for complications from obesity.

WHAT IS A “HEALTHY” WEIGHT?

In 1995, the World Health Organization (WHO) codified BMI as a screening index for obesity by defining a BMI 25–29.9 kg/m2 as overweight, and a BMI ≥ 30 kg/m2 as obesity.7 WHO further categorized obesity severity by classes I (30–34.9 kg/m2), II (35–39.9 kg/m2), and III (≥ 40 kg/m2).8 WHO BMI classes were created in part to emphasize the known association of increased mortality with a BMI both below and above the “normal” range (18.5–24.9 kg/m2).9,10 However, in the past half-century, control of cardiovascular risk factors such as hyperlipidemia and hypertension, primary drivers of mortality in individuals with obesity, has significantly improved across all BMI cohorts.11 Perhaps in part as a result, many modern cohort studies suggest that the lowest mortality-risk BMI, at least in studies primarily of European and North American populations, is in the overweight WHO classification, and that class I obesity does not confer significantly increased mortality risk.12,13 Class II and III obesity, however, continues to be associated with increased mortality risk in these studies.

Further, subsequent research has identified a wide diversity in health outcomes for different ethnic groups at similar BMIs.14,15 In a striking example from the UK, BMI significantly underestimated the risk for diabetes for many minority ethnic populations, especially those of South Asian descent, compared to White populations.16 In view of these differences, use of BMI cutoffs to identify individuals eligible for AOM or metabolic surgery might lead to substantial racial disparities in care for individuals who may benefit from weight loss interventions.

Cardiorespiratory fitness (CRF) and obesity-related comorbidities may significantly impact the mortality risk of an elevated BMI. Activity levels also have known strong associations with lower mortality,17,18 and increased physical activity improves CRF.19,20 Studies examining the relationship between BMI and mortality, when adjusted for CRF, have shown that the excess mortality risk of increased adiposity is greatly reduced or eliminated by having high CRF.21,22,23 Metabolically healthy obesity (MHO), defined as having a BMI ≥ 30 kg/m2 with ≤ 1 cardiovascular risk factors of dyslipidemia, hypertension, or fasting hyperglycemia,24 is present in one- to two-thirds of adults with obesity.25,26 Adults with MHO have significantly decreased mortality compared to those with metabolically unhealthy obesity.27 Therefore, a person with obesity with high CRF and/or without metabolic comorbidities may have similar mortality to normal BMI, fit, metabolically healthy individuals.28

Finally, there are many effective interventions to manage obesity-related diseases without weight loss. In individuals with hypertension, weight-neutral dietary modifications such as the Dietary Approaches to Stop Hypertension diet lower blood pressure to a similar degree to weight-loss diets,29,30,31 and exercise programs can reduce blood pressure independent of weight loss.32 Anti-hypertensives, when indicated, likely have a more pronounced blood pressure–lowering effect than these interventions.33 Weight loss diets for elevated LDL cholesterol in one randomized trial failed to lower LDL cholesterol,34 while statins,35 when indicated, and weight-neutral diets such as the Portfolio Diet36 have been shown to improve lipid profiles in hyperlipidemia. The DA Qing IGT and Diabetes Study showed similar reductions in diabetes incidence in those with prediabetes when comparing exercise alone to a weight loss diet,37 suggesting that exercise may be as effective as weight loss diets for diabetes prevention. Similar positive effects of exercise alone have been shown for glycemic control in individuals with type 2 diabetes.38 Healthy diets, physical activity, and pharmacotherapy can protect the health of persons with metabolic comorbidities of obesity without weight loss.

HOW EFFECTIVE ARE WEIGHT LOSS INTERVENTIONS?

Despite the rising prevalence of dieting in US adults with obesity,39,40 clinical trials of diet programs across the spectrum of macronutrient compositions consistently fail to produce clinically significant weight loss at 12 months for most individuals.41 Patients who are highly motivated to adhere to a strict diet and exercise program face the challenge of any significant weight loss attempt: adiposity stores are centrally regulated, and the body defends against weight loss.42 The mechanisms of this complex, long-term defense include metabolic adaptation, in which energy expenditure decreases more than expected from weight change alone,43,44 and increased hunger hormone signaling leading to an increased drive to eat.45 These defenses promote positive energy balance, leading to weight regain in most circumstances.

While weight loss diets are mostly ineffective, metabolic surgery is highly effective for long-term weight loss, resulting in an average of 22% weight loss at 20 years follow-up.46 The safety of these procedures has also greatly improved in the past 30 years with the advent of the laparoscopic technique and better guidelines for perioperative care for metabolic surgery.47,48 These procedures, however, remain generally reserved for patients with a BMI of 40 kg/m2, or 35 kg/m2 with comorbidities,48 and utilization in eligible patients remains exceptionally low in the USA (0.5%).47

Finally, the weight loss efficacy of AOM has substantially increased in the past decade since U.S. Food and Drug Administration approval of bupropion-naltrexone, liraglutide, phentermine-topiramate, and semaglutide. These therapies average 6–15% weight loss at 12 months, depending on the medication.49,50,51,52 Promising additional AOM include tirzepatide53 and semaglutide-cagrilintide,54 for which early evidence suggests increased weight loss compared to semaglutide to a degree approaching metabolic surgery. However, until prescription drug coverage for novel AOM improves,55 these interventions are likely to be cost-prohibitive for most patients.

DOES WEIGHT LOSS REALLY IMPROVE HEALTH?

In preparation for the USPSTF 2018 recommendations, investigators prepared a systematic review of behavioral and pharmacotherapy weight loss interventions to prevent obesity-related morbidity and mortality. This review reported no improvement in mortality, cardiovascular outcomes, or health-related quality of life for intervention arms of behavioral and pharmacological weight loss and maintenance trials.56 For behavioral interventions, these findings are not surprising given that most subjects do not maintain 5% weight loss, the amount necessary to result in long-term clinical benefits for a variety of obesity-related comorbidities.57

While behavioral and pharmacologic interventions have not yet demonstrated improvement in most clinical outcomes, compelling evidence suggests that metabolic surgery in high-risk patients confers a substantial mortality benefit.58 However, the average BMI of the individuals enrolled in these studies was above 40 kg/m2. Therefore, it is unclear if the same health benefits would be evident in those with class I or class II obesity. As increasingly effective AOM develop with indications for class I and II obesity, studies of AOM must begin to assess health outcomes beyond weight loss itself. A promising upcoming example is SELECT, a randomized controlled trial of semaglutide for secondary prevention of cardiovascular disease in individuals with obesity and without diabetes.59

WHY RECOMMENDING WEIGHT LOSS DIETING MAY BE HARMFUL

Patients with obesity receiving health care in a weight-centric paradigm endorsed by the USPSTF may be at increased risk for weight stigma.60,61 When recommending diet and exercise with the goal of weight loss to individuals with obesity, most of whom have had unsuccessful weight loss attempts in the past, clinicians may be implying to these individuals that they simply need to try harder to lose weight, reenforcing a stereotype that individuals with obesity lack discipline and self-control. In surveys both of providers and patients, weight bias perpetrated by health care practitioners is common,62,63,64,65 and can contribute to weight bias internalization (WBI), in which an individual is aware of their stigmatized identity and applies negative stereotypes (e.g., that individuals with obesity are lazy, lack self-control) to themselves. WBI is strongly associated with poor health outcomes such as worsened mental health, stress, health-related quality of life, and disordered eating,66 and over half of individuals with a BMI ≥ 30 kg/m2 in one survey had very high scores on an index of WBI.67 Identifying implicit and explicit weight bias in primary care settings and avoiding stigmatizing language is critical to improve the care of patients with obesity.63,68,69

Beyond the risk for promoting weight stigma, an additional harm of recommending dieting for weight loss is the risk for weight cycling, in which those who attempt diets repeatedly lose and regain weight. Weight cycling may lead to long-term weight gain, insulin resistance, dyslipidemia, and hypertension,39 and, in some secondary analyses of clinical trial data and prospective cohort studies, has been associated with increased mortality.70,71,72 Especially in patients with a history of repeated weight loss and regain, clinicians should understand this risk.

A WEIGHT-SKEPTICAL APPROACH TO THE CARE OF PATIENTS WITH OBESITY

Table 1 summarizes the key principles of a weight-skeptical approach to the care of patients with obesity. Based on current evidence, we propose maintaining skepticism toward the treatment benefits of weight loss interventions, as they may be attenuated by the potential harms of weight stigma and weight cycling imposed by weight-centric strategies. Further, the potential clinical benefits of weight loss may depend on an individual’s underlying cardiorespiratory fitness and comorbidities, as well as the modality of treatment. Table 2 reviews several common patient scenarios, with suggested language for clinicians using a weight-skeptical approach to care.

Table 1 A Weight-Skeptical Approach to the Care of Patients with Obesity
Table 2 Common Patient Interactions with a Weight-Skeptical Approach

This framework allows a clinician to incorporate new evidence for the possible benefits of AOM or weight-neutral strategies in a likely future when highly effective AOM is broadly available, and more weight-neutral lifestyle programs are tested for their clinical benefits. Obesity treatment guidelines of the future may recommend against or make optional weight loss interventions for adults with class I or II obesity who have controlled cardiovascular risk factors and good CRF, while recommending AOM and/or metabolic surgery in patients as treatment options in patients with higher-risk obesity.

CONCLUSION

A weight-skeptical approach to the care of patients with obesity provides an alternative to strictly weight-centric or weight-neutral paradigms. It encourages clinicians to question the role of BMI in predicting health outcomes, and to acknowledge the lack of evidence for the long-term clinical benefits of lifestyle-based weight loss interventions for many adults with obesity. Critically, this model of care recognizes the potential harms of weight-centric treatment, such as the risk for weight bias and weight cycling. The impact of adopting such an approach is that a primary care clinician may often find themself steering away from recommendations for weight loss and toward weight-neutral interventions, while still offering AOM and metabolic surgery, when available and desired by the patient, to improve the health of patients at high risk for complications from obesity.