Evolution of a controversy
The notion of metabolically healthy obesity – that people can be obese without an increased risk for poor health outcomes – can be traced back several decades.
“There was a chap called Steve Blair,” says epidemiologist Mark Hamer (Loughborough University, UK). “In the 80s he turned this phrase ‘the fit fat’ and he showed that, quite consistently, if you were obese but had a certain level of cardiorespiratory fitness, you would be at lower risk of cardiovascular disease and mortality than the obese people who were not fit.”
He notes that the definition has evolved since then, to encompass people’s metabolic profiles. “But it’s not that different really from the fit–fat concept, to be honest, so I think the basic concept of ‘can you be healthy and obese’ has been knocking around for quite a few years.”
The theory gained high-profile exposure in 2013, with the publication in JAMA of a meta-analysis that attempted to shed light on the relationship between bodyweight and all-cause mortality, using standard BMI categories and data from nearly 3 million people across 97 studies . It showed the expected increased risk associated with grade 2 and 3 obesity (BMI ≥35 kg/m2) but no increased risk with grade 1 obesity and a mildly protective effect with overweight, and was met by a storm of criticism. Yet this paper was by no means the first to suggest the existence of a subgroup of metabolically healthy obese people.
As early as 1997, researchers were categorizing obese people into those with and without insulin resistance and hypothesizing that the latter subgroup could have a lower risk for diabetes and cardiovascular disease than the former . A number of studies over the following years refined the concept, showing for instance that the metabolically healthy obese phenotype was associated with a reduced accumulation of visceral fat , and ascribing the increased risk associated with elevated BMI to co-occurring metabolic abnormalities, as opposed to obesity per se [4, 5, 6].
These earlier papers actively proposed and supported the concept of metabolically healthy obesity, whereas the JAMA paper did not even allude to it – yet the latter received, in the words of Hamer, “an absolute beasting.”
Much of the problem, Hamer feels, was that the paper was published in an extremely high-profile journal by someone from within the US national healthcare system – lead author Katherine Flegal is based at the Centers for Disease Control and Prevention, “but she didn’t show what people wanted to see.” In other words, the bodyweight advice healthcare professionals wished to deliver to their patients was gainsaid by the paper’s implication that mild obesity could be harmless or even protective.
“In terms of a public health message, that’s really atrocious,” says Hamer.
At around the same time, other publications appeared to contradict the existence of metabolically healthy obesity. For example, findings from the Uppsala Longitudinal Study of Adult Men showed a significantly increased cardiovascular disease and death risk for metabolically healthy overweight and obese people, and two large cohort studies showed they also had an elevated type 2 diabetes risk [7, 8, 9]. This was replicated on a larger scale in a meta-analysis published in the Annals of Internal Medicine, involving more than 61,000 people, whose authors concluded that “there is no healthy pattern of increased weight.” 
The reaction to this last study, published 1 year after Flegal and colleagues’ JAMA paper, demonstrates the frequently divergent opinions of epidemiologists and other researchers attempting to pick apart the metabolically healthy obesity effect. Both publications attracted a clutch of letters challenging the methodology; those disagreeing with the Annals paper, which included one from Flegal, took issue with the definition of metabolically healthy obesity, the supposed exclusion of high-quality studies, the use of unadjusted data, and the fact that BMI does not adequately account for fat mass. The subjectivity of these views is aptly illustrated by the fact that one of the letters disputing the findings of the JAMA meta-analysis did so also on the basis that BMI does not adequately account for fat mass . The letters challenging the findings of this paper also disputed the handling of heterogeneity between the included studies and the unaccounted effects of a heterogeneous control group.
Conversely, the authors of two letters in response to the JAMA publication felt the analysis had not gone far enough and that accounting for the effects of age and vascular risk factors might have produced an even stronger metabolically healthy obesity effect.
Since the JAMA paper’s publication, Hamer has seen two “quite distinct sides” in operation: people who were concerned by its findings, and those who hoped it was a true effect, that doctors could be more relaxed about their patients becoming overweight or only mildly obese. So research since then has been driven by the two sides “trying to find out reasons to either disprove [the research] or to say well there is actually a plausible mechanism.”
“I think everyone was surprised by how controversial [the paper] really was,” says Hamer.
Several factors have emerged from this later research that help to explain the conflicting results of earlier papers, mostly involving the definition of healthy and unhealthy people and the fact that people can change between definitions over time. Hamer himself was first author of one of the papers showing no increase in cardiovascular risk in the metabolically healthy obese, but he says that since its publication his team has looked more closely at the stability of definitions over time.
“One argument for the so-called healthy obesity people is just that these are the people who have not been obese long enough to actually develop the metabolic risk factors,” he says. Indeed, this group has an increased likelihood of developing metabolic abnormalities over time , with around half doing so in one 20-year follow-up study . Consequently, a meta-analysis that factored in study duration found the metabolically healthy phenotype to be associated with poor health outcomes . Likewise, a large recent analysis based on people’s maximum BMI over a 16-year period, rather than on a single baseline measurement, found excess risk for all weight categories above the normal range . By contrast, when the authors of this study used a single baseline weight measurement, they found a paradoxical protective effect of overweight on mortality.
This latter point was one failing of Flegal’s JAMA paper, says Hamer, being based on BMI measured at one point in time.
“If you look at lifetime weight history you actually get a different picture, because what is sometimes happening is that someone with obesity gets sick they then start to lose weight and they drop into the lower categories and that messes up the reference category,” he says.
Added to which, people originally placed in the normal-weight category can become overweight or obese, “so when you’re only using BMI data from one point in time you just don’t know what’s going on in terms of the whole weight history.”
On top of that there is no consensus regarding how these categories – healthy and unhealthy normal weight and obesity – should be defined. Thus far, the categories have been data-driven, determined by the available measurements. “Because large cohort studies are designed to answer multiple questions, not the one that you’re necessarily going back to answer, you tend to use what data are available and define healthy obesity based on what you have,” observes Hamer. “So that adds a bit of confusion.”
Consequently, the published prevalence of healthy obesity varies between around 10% and 50%. But if researchers used extremely conservative definitions, allowing no risk factors and including less frequently measured variables such as insulin resistance and fatty liver, Hamer suspects the prevalence would be closer to 2%.
“It’s a little bit of a myth that these people are very abundant in the population,” he says, “because they’re just not.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2017 Springer Healthcare part of the Springer Nature group
 Flegal KM, Kit BK, Orpana H, Graubard BI. Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA 2013; 309: 71–82 http://jamanetwork.com/journals/jama/fullarticle/1555137
 Ferrannini E, Natali A, Bell P, et al. Insulin resistance and hypersecretion in obesity. European Group for the Study of Insulin Resistance (EGIR). J Clin Invest 1997; 100: 1166–1173 https://www.jci.org/articles/view/119628
 Brochu M, Tchernof A, Dionne IJ, et al. What are the physical characteristics associated with a normal metabolic profile despite a high level of obesity in postmenopausal women? J Clin Endocrinol Metab 2001; 86: 1020–1025 https://doi.org/10.1210/jcem.86.3.7365
 Meigs JB, Wilson PW, Fox CS, et al. Body Mass Index, Metabolic Syndrome, and Risk of Type 2 Diabetes or Cardiovascular Disease. J Clin Endocrinol Metab 2006; 91: 2906–2912 https://doi.org/10.1210/jc.2006-0594
 Kip KE, Marroquin OC, Kelley DE, et al. Clinical importance of obesity versus the metabolic syndrome in cardiovascular risk in women: a report from the Women's Ischemia Syndrome Evaluation (WISE) study. Circulation 2004; 109: 706–713 http://circ.ahajournals.org/content/109/6/706
 Hamer M, Stamatakis E. Metabolically Healthy Obesity and Risk of All-Cause and Cardiovascular Disease Mortality. J Clin Endocrinol Metab 2012; 97: 2482–2488 https://doi.org/10.1210/jc.2011-3475
 Arnlöv J, Ingelsson E, Sundström J, Lind L. Impact of Body Mass Index and the Metabolic Syndrome on the Risk of Cardiovascular Disease and Death in Middle-Aged Men. Circulation 2010; 121: 230–236. http://circ.ahajournals.org/content/121/2/230
 Hinnouho GM, Czernichow S, Dugravot A, et al. Metabolically healthy obesity and the risk of cardiovascular disease and type 2 diabetes: the Whitehall II cohort study. Eur Heart J 2015; 36: 551–559 https://doi.org/10.1093/eurheartj/ehu123
 Twig G, Afek A, Derazne E, et al. Diabetes Risk Among Overweight and Obese Metabolically Healthy Young Adults. Diabetes Care 2014; 37: 2989–2995. https://doi.org/10.2337/dc14-0869
 Kramer CK, Zinman B, Retnakaran R. Are metabolically healthy overweight and obesity benign conditions?: A systematic review and meta-analysis. Ann Intern Med 2013; 159: 758–769 http://annals.org/aim/article/1784291/metabolically-healthy-overweight-obesity-benign-conditions-systematic-review-meta-analysis
 Willett WC,. Hu FB, Thun M. Overweight, Obesity, and All-Cause Mortality. JAMA 2013; 309: 1681–1682 https://jamanetwork.com/journals/jama/article-abstract/1681397
 Bradshaw PT, Reynolds KR, Wagenknecht LE, Ndumele CE, Stevens J. Incidence of components of metabolic syndrome in the metabolically healthy obese over 9 years follow-up: the Atherosclerosis Risk In Communities study. Int J Obes 2017; Advance online publication https://www.nature.com/ijo/journal/vaop/ncurrent/abs/ijo2017249a.html
 Bell JA, Hamer M, Sabia S, et al. The Natural Course of Healthy Obesity Over 20 Years. J Am Coll Cardiol 2015; 65: 101–102 http://www.onlinejacc.org/content/65/1/101
 Fan J, Song Y, Chen Y, Hui R, Zhang W. Combined effect of obesity and cardio-metabolic abnormality on the risk of cardiovascular disease: A meta-analysis of prospective cohort studies. Int J Cardiol 2013; 168: 4761–4768 https://doi.org/10.1016/j.ijcard.2013.07.230
 Yu E, Ley SH, Manson JE, et al. Weight History and All-Cause and Cause-Specific Mortality in Three Prospective Cohort Studies. Ann Intern Med 2017; 166: 613–620 http://annals.org/aim/article-abstract/2615810/weight-history-all-cause-cause-specific-mortality-three-prospective-cohort