TL;DR:

  • Evidence for negative health effects of obesity seems robust. Evidence for any health effects (whether positive or negative) of overweight seems surprisingly sparse, heterogeneous, and not particularly clinically meaningful.
  • Even in obese people, whether “lose weight” actually causes positive health outcomes compared to “not losing weight” has not yet been demonstrated to my satisfaction. I didn’t look very hard! But I just wanted to flag that “fat = bad”, even if proven beyond the shadow of a doubt, does not necessarily lead to the conclusion that “losing fat = good”; that latter conclusion requires an extra assumption (losing fat reverses the negative health effect) which itself requires empirical testing. And, of course, there is the question of the negative health effects of messages about body weight, including eating disorders, mental health challenges and stigma faced when trying to exercise, and so on. But that’s a topic for another day…!
  • BMI, loosely speaking, is rubbish. No surprise there. In particular, heterogeneity across different populations seems like a blindingly reasonable conclusion (hahaha) given the context. In particular, being transgender and on hormone replacement therapy, I do not see myself represented in these study cohorts, which leads me to question whether the empirical evidence that exists is even representative of people with my particular biological and physiological characteristics.
  • I can’t shake the feeling when reading studies on this topic (especially some that I didn’t note) that the rigour used to support claims like “fat = bad” is quite poor compared to other areas of science, and in particular there seems to be a double standard that assumes that “normal” and “underweight” BMI values are innocent until proven guilty, while “overweight” and “obese” BMI values are guilty until proven innocent.
  • So basically I’m actually leaning towards the conclusion, promoted in the critical fat studies literature, that medical discourse on body fat is reductionist and harmful. The Mathur studies in particular do a good job of demonstrating that health messaging systematically overstates the risk of overweight (though perhaps not obesity), which to me is a clear sign of a reductionist and harmful message. That is to say, there seems to be a value judgment being made somewhere along the way, meaning that it’s hard to disentangle the empirical evidence from doctors’ and researchers’ pre-existing views on a topic (which in turn should shed serious doubt on the general public’s views on this topic, which are clearly and obviously highly politicised).
  • I think this evidence, taken together, leads me to believe that a person-centered approach that is aligned with a particular individual’s experiences, context, and goals is much more reasonable than “BMI > some number = bad”. Probably the increased risk of obesity should be included as one part of this approach, but this really is only one part among many even for people in the (still mostly arbitrary) “obese” category.

Jayedi et al 2020, Central fatness and risk of all cause mortality: systematic review and dose-response meta-analysis of 72 prospective cohort studies

  • Of 98745 studies screened, 1950 full texts were fully reviewed for eligibility. The final analyses consisted of 72 prospective cohort studies with 2528297 participants.
  • Indices of central fatness including waist circumference, waist-to-hip ratio, waist-to-height ratio, waistto-thigh ratio, body adiposity index, and A body shape index, independent of overall adiposity, were positively and significantly associated with a higher all cause mortality risk. Larger hip circumference and thigh circumference were associated with a lower risk.
  • We found that the associations remained significant after body mass index was accounted for, which indicated that abdominal deposition of fat, independent of overall obesity, is associated with a higher risk.
  • However, this study showed that two indices of central fatness—thigh circumference and hip circumference—were inversely associated with all cause mortality risk.
  • (discussion sorta makes the case that BMI is a rubbish proxy for adiposity and waist circumference has much more empirical value in predicting negative health outcomes)
  • Participants with pre-existing diseases are more likely to experience death in the early years of follow-up, and so the proportion of deaths caused by pre-existing diseases, and the confounding effects of pre-existing diseases, decline with the increase in follow-up duration. Therefore, studies with long term follow-up durations might present a more accurate estimation of the associations.
  • Follow-up durations were between 3 and 24 years.

Now, two perspective-type studies from Maya Mathur. I’ve read Mathur’s work in other contexts and have been consistently impressed by its rigour.

Mathur and Mathur 2024, Toward evidence-based communication on overweight body mass index and mortality

  • The current evidence base suggests that although both obese BMI and underweight BMI are consistently associated with increased all-cause mortality, overweight BMI (without obesity) is not meaningfully associated with increased mortality. In fact, a number of studies suggest modest protective, rather than detrimental, associations of overweight BMI with all-cause mortality. Given this current evidence base, clinical guidelines and physician perceptions substantially overstate all-cause mortality risks associated with the range of BMIs classified as “overweight” but not “obese.”
  • The question of whether BMIs currently classified as “overweight” are a risk factor for all-cause mortality is not fully resolved given methodological limitations. However, as noted above, the strongest available evidence typically does not suggest clinically meaningful associations of overweight with increased mortality and often instead suggests modest associations in the opposite direction.
  • Recommendation 1. Top-line recommendations to the public and to health practitioners who inform the public should carefully distinguish overweight BMI from obese BMIs. At this time, BMIs categorized as “overweight” should no longer be described as contributing to all-cause mortality.
  • Recommendation 2. Continuing education and clinical practice guidelines should address primary care physicians’ widespread misconception that overweight BMIs substantially increase mortality risk, because these conceptions far exceed empirical estimates and may distort physicians’ recommendations to patients.
  • Despite the null or weak associations of overweight BMI with all-cause mortality, it is possible that overweight BMI could affect cause-specific mortality, such as cardiovascular mortality, or could affect disease burden and quality of life. Additionally, for some individuals, becoming overweight could ultimately lead to becoming obese, which is a greater health concern.
  • Recommendation 3. The evidence basis for other potential risks or benefits of overweight BMI (e.g., cardiovascular complications or development of obesity) should be rigorously examined and incorporated appropriately into health communications from health authorities to health practitioners so that the latter can, in turn, counsel individual patients appropriately.
  • On the other hand, the current evidence overall suggests that overweight BMI (exclusive of obesity) is not associated, or is only modestly associated, with mortality when compared to normal BMI. Among studies that do indicate an association, the direction is inconsistent: in many studies, overweight BMI is associated with modestly reduced, rather than increased, mortality. Two prominent meta-analyses in general populations reported modest associations in opposite directions for the overweight category (hazard ratio HR = 0.94 versus HR = 1.11 [9, 10]). Another meta-analysis suggested that in older adults, overweight BMI may be associated with modestly reduced mortality [11].
  • Corroborating the epidemiologic evidence from these studies and the meta-analyses, randomized trials have suggested that for patients with overweight or obese BMI, weight loss does not clearly improve either morbidity or mortality, although there may be a benefit for patients with morbid obesity [16, 17].

Mathur and VanderWeele 2022, Effects of being overweight on mortality are unclear given multiple methodological problems

  • Overall, we consider the question of whether being overweight (but not obese) affects all-cause mortality to be largely unresolved, though we suspect that effects are typically small and heterogeneous across populations. It is rather puzzling that public health messaging has almost exclusively discussed detrimental effects when in fact, there is currently not particularly robust evidence for effects in either direction.