Comment
11 Tang H, Hammack C, Ogden SC, et al. Zika virus infects human cortical neural
progenitors and attenuates their growth. Cell Stem Cell 2016; 18: 587–90.
12 Gilmore EC, Walsh CA. Genetic causes of microcephaly and lessons for
neuronal development. Wiley Interdiscip Rev Dev Biol 2013; 2: 461–78.
13 Cohen MM Jr. Holoprosencephaly: clinical, anatomic, and molecular
dimensions. Birth Defects Res A Clin Mol Teratol 2006; 76: 658–73.
Weight loss diet studies: we need help not hype
Over the past several decades, dozens of randomised
controlled trials have compared various diets for the
treatment of obesity. Ideally, such studies should
have provided strong evidence for clear clinical
recommendations and also put a stop to society’s endless
parade of fad diets. Unfortunately, the evidence base
remains contested and the “diet wars” continue unabated.
One insight that can be gleaned from the existing
weight loss literature is that even the most divergent
of diets seem capable of affecting a degree of short-
term success, with some diets perhaps leading to
marginally greater losses than others over periods of
several months.1 But since obesity is a chronic condition,
it is the long term that matters. An effective diet for
clinical weight management needs to be established
over timescales of years to decades. Studies that have
lasted 1 year or more typically do not show significant
differences between prescribed diets, much less any
clinically meaningful differences in maintenance of lost
weight.1,2 One example is in the Dietary Intervention
Randomized Controlled Trial (DIRECT), which has been
hailed as proof of the superiority of low-carbohydrate
diets over low-fat diets.3,4 The DIRECT investigators used
a 2-year workplace intervention and found that a low-
carbohydrate diet prescription led to a significant 1·8 kg
greater mean bodyweight loss than the prescription of
a low-fat diet.3,4 These bodyweight differences between
the diets are among the largest differences that have
been observed over a 2-year period. But from the clinical
perspective, such small bodyweight differences do not
instil confidence for prescribing one diet over another to
a patient with obesity.
What is especially striking is the similarity of the long-
term pattern of mean bodyweight change, irrespective
of diet prescription.5 For example, figure 1A shows data
from the DIRECT study in which both the low-fat and low-
carbohydrate diets resulted in rapid early weight loss that
plateaued after about 6 months at a likely disappointing
level6 and was then followed by slow bodyweight regain.
www.thelancet.com Vol 388 August 27, 2016
14 Raam MS, Solomon BD, Shalev SA, Muenke M. Holoprosencephaly and
craniosynostosis: a report of two siblings and review of the literature.
Am J Med Genet C Semin Med Genet 2010; 154C: 176–82.
15 Flaherty K, Singh N, Richtsmeier JT. Understanding craniosynostosis as a
growth disorder. Wiley Interdiscip Rev Dev Biol 2016; 5: 429–59.
What can we learn from the physiology underlying such a
bodyweight trajectory?
Complex physiological feedback mechanisms regulate
bodyweight and resist weight loss. Slowing of metabolism
can be substantial and persistent7 and plays a part in
halting weight loss and putting subsequent weight regain
into motion. However, the typical bodyweight trajectory is
primarily driven by patients experiencing an exponential
decay of diet adherence due to an increase in appetite
in proportion to the loss of bodyweight,8 along with
difficulties in sustaining changes to dietary choices and
behaviours that affect patients’ ability to enjoy, celebrate,
and socialise with food.
Figure 1B shows the energy intake changes underlying
the DIRECT trial’s observed bodyweight trajectories, which
we have calculated using a validated mathematical model
of human energy metabolism and bodyweight dynamics.9
At the plateau point of maximum weight loss, energy
intake is balanced by expenditure and has decreased from
baseline by about 200 kcal per day. By contrast, mean
energy intake at the bodyweight plateau has increased
by about 700–1000 kcal per day from its early reduction
at the start of the intervention. After 1 year, mean
bodyweights, although still reduced by several kilograms,
climb back up in response to the average energy intakes
returning almost to baseline levels.
Diet adherence is so challenging that it is poor even
in short-term studies where all food is provided.10 When
diets are prescribed, adherence is likely to diminish
over the long term despite self-reports to the contrary.
Figure 1B illustrates that the common self-report
methods for measuring food intake (24 h recall and
food frequency questionnaire3,4) mistakenly indicate
that the reduction in energy intake remained unchanged
throughout the intervention. Such erroneous
measurements have led to speculation that a reduction
in energy expenditure, rather than loss of diet adherence,
is the main driver of the bodyweight plateau. However,
these self-reported measurements are known to be
849
 Comment

A
Nevertheless, and hearteningly, anecdotal long-term
0 diet success stories abound for most dietary approaches,
–1 and focusing on mean bodyweight trajectories masks the
high individual weight loss variability within each diet
–2
Bodyweight change (kg)

group. The question is: why are some individuals more

–3
successful than others? When it comes to clinical weight
–4 management, success is predicated on long-term dietary
–5 adherence. Therefore, we need to increase our efforts to
–6 understand the individual differences between patients
–7
that have an effect on diet maintenance and prevent
Low-fat diet
Low-carbohydrate diet its erosion. Studies should determine how to target
–8
effective diets to individual patients,12 as well as improve
B our understanding of the real world considerations
0
that impinge on patients’ abilities to sustain healthy
–200 dietary changes,13 such as those wrought by the food
environment, socioeconomic factors, cooking skills,
Intake change (kcal/day)

–400 job requirements, medical comorbidities, caregiving

–600
–800
Low-fat diet
Low-carbohydrate diet
–1000
Food frequency questionnaire
24 h recall
–1200
0 3 6 9 12 15 18 21
Time (months)
Figure: Mean bodyweight changes (A) and energy intake changes (B) in the DIRECT trial
Error bars are ±1 SE. Mathematical model simulations are represented by curves. Figure B calculated using validated
mathematical model8 in both diet groups along with self-reported measurements obtained by 24 h recall 3 or food
frequency questionnaire.4 We used data from the DIRECT trial for this figure.3,4
inaccurate for estimating energy intake11 and provide
unreliable data on energy intake changes that are not
quantitatively reconcilable with objectively measured
weight regain and the known physiology of energy
metabolism adaptations.
The similarity of mean bodyweight trajectories
between long-term diet interventions, whether
targeting macronutrients, calories, or food patterns, is
explained by the fact that no diet has yet been shown
to be uniformly easier to stick with than another in the
long run. If there existed a diet that led to substantially
improved long-term adherence in most patients—
because it better addressed appetite changes, provided
a sustained metabolic advantage, or was simply easier
for patients to maintain—such effects would result
in substantial and sustained differences in mean
bodyweight. This result has not been observed despite
repeated efforts using widely different diets.
responsibilities, and many more. After all, as with every
chronic disease, successful obesity management requires
lifelong treatment and there is a pressing need to help
patients navigate day-to-day realities in the face of
maintaining a permanent and intentional behaviour
change. We also need to better understand how family,
24
community, and society as a whole can help support and
sustain healthy lifestyles.
Fewer resources should be invested in studying whether
or not a low-carbohydrate diet is marginally better than
a low-fat diet, or whether intermittent fasting provides
marginally better short-term outcomes than a so-called
Paleo diet. Crowning a diet king because it delivers a
clinically meaningless difference in bodyweight fuels diet
hype, not diet help. It’s high time we started helping.
Yoni Freedhoff, *Kevin D Hall
Bariatric Medical Institute, Ottawa, ON, Canada (YF); and Integrative
Physiology Section, Laboratory of Biological Modeling, National
Institute of Diabetes and Digestive and Kidney Diseases, Bethesda,
MD 20892, USA (KDH)
kevinh@niddk.nih.gov
YF has received honoraria and travel expenses from Boston Children’s Hospital,
Canadian Obesity Network, Centre for Effect Practice, Academy of Medicine
Ottawa, Physical and Health Education Canada, North York General Hospital, IDEA
Health and Fitness Association, and the Royal Society of Medicine, London, for
speaking engagements and for his role as clinical lead in the development of a
Canadian Ministry of Health funded tool for primary care providers working with
families of children with obesity; and has received fees for developing and
delivering educational seminars to medical students and residents from the
University of Ottawa. YF writes a blog, Weighty Matters, that is non-monetised
with no advertisements or requests for donations. YF is the co-author of Best
Weight: A Practical Guide to Office-Based Obesity Management. All royalties from the
book go to the Canadian Obesity Network. He is the author of The Diet Fix (Random
House) and receives royalties from this book. KDH reports a patent pending on a

850 www.thelancet.com Vol 388 August 27, 2016


method of personalised dynamic feedback control of bodyweight (US Patent
Application No 13/754,058; assigned to the National Institutes of Health) and has
received funding from the Nutrition Science Initiative to investigate the effects of
ketogenic diets on human energy expenditure.
1 Johnston BC, Kanters S, Bandayrel K, et al. Comparison of weight loss among
named diet programs in overweight and obese adults: a meta-analysis. JAMA
2014; 312: 923–33.
2 Tobias DK, Chen M, Manson JE, Ludwig DS, Willett W, Hu FB. Effect of low-fat
vs. other diet interventions on long-term weight change in adults: a systematic
review and meta-analysis. Lancet Diabetes Endocrinol 2015; 3: 968–79
3 Greenberg I, Stampfer MJ, Schwarzfuchs D, Shai I. Adherence and success in
long-term weight loss diets: the dietary intervention randomized controlled
trial (DIRECT). J Am Coll Nutr 2009; 28: 159–68.
4 Shai I, Schwarzfuchs D, Henkin Y, et al. Weight loss with a low-carbohydrate,
Mediterranean, or low-fat diet. N Engl J Med 2008; 359: 229–41.
5 Franz MJ, VanWormer JJ, Crain AL, et al. Weight-loss outcomes: a systematic
review and meta-analysis of weight-loss clinical trials with a minimum 1-year
follow-up. J Am Diet Assoc 2007; 107: 1755–67.
6 Foster GD, Wadden TA, Vogt RA, Brewer G. What is a reasonable weight loss?
Patients’ expectations and evaluations of obesity treatment outcomes.
J Consult Clin Psychol 1997; 65: 79–85.
Dietary guidelines on trial: the charges are not evidence based
Current nutrient guidelines recommend a wide
acceptable range of total fat and carbohydrate intakes,
emphasising quality and source rather than quantity
of macronutrients, substantial restriction of free
sugars, and usually restriction of saturated fat.1 Recent
food-based dietary guidelines are based on similar
recommendations.2 Wide dietary variation, including
typical healthy dietary patterns, can be accommodated
within such nutrient-based advice. And so criticisms of
nutrition guidelines confuse both health professionals
and the public, and provide justification for inaction by
policy makers.3 In particular, claims that carbohydrate
rather than saturated fat be restricted are based on
incomplete assessment of evidence and undermine
measures to reduce non-communicable diseases (NCDs).
The 2015 updated Cochrane review on dietary fat
reported a meta-analysis of 15 trials of at least 2 years
duration involving 59 000 participants.4 The trials that
involved replacing saturated with polyunsaturated fats
suggested a 27% reduction in cardiovascular disease,
but no clear health benefits were seen for replacement
of saturated fat with carbohydrates or proteins.
A dose-response analysis showed that this protection
was proportional to reduction of total cholesterol, in
turn influenced by extent of reduction in saturated fat
consumption. No effect of replacing saturated with
polyunsaturated fat on all-cause or cardiovascular
disease mortality was observed or expected, given
the short duration of trials (about 4 years), that most
www.thelancet.com Vol 388 August 27, 2016
Comment
7 Fothergill E, Guo J, Howard L, et al. Persistent metabolic adaptation 6 years
after “The Biggest Loser” competition. Obesity (Silver Spring) 2016; published
online May 2. DOI:10.1002/oby.21538.
8 Polidori D, Sanghvi A, Seeley RJ, Hall KD. How strongly does appetite counter
weight loss? Quantification of the feedback control of human energy intake.
Obesity (Silver Spring) (in press).
9 Sanghvi A, Redman LA, Martin CK, Ravussin E, Hall KD. Validation of an
inexpensive and accurate mathematical method to measure long-term
changes in free-living energy intake. Am J Clin Nutr 2015; 102: 353–58.
10 Das SK, Gilhooly CH, Golden JK, et al. Long-term effects of 2 energy-restricted
diets differing in glycemic load on dietary adherence, body composition, and
metabolism in CALERIE: a 1-y randomized controlled trial. Am J Clin Nutr
2007; 85: 1023–30.
11 Dhurandhar NV, Schoeller DA, Brown AW, et al. Energy balance
measurement: when something is not better than nothing. Int J Obes (Lond)
2015; 39: 1109–13.
12 Bray MS, Loos RJ, McCaffery JM, et al. NIH working group report-using
genomic information to guide weight management: from universal to
precision treatment. Obesity (Silver Spring) 2016; 24: 14–22.
13 MacLean PS, Wing RR, Davidson T, et al. NIH working group report:
innovative research to improve maintenance of weight loss.
Obesity (Silver Spring) 2015; 23: 7–15.
cardiovascular disease events are not fatal, and only about
a third of total deaths were attributed to cardiovascular
disease. Prospective analysis of PREDIMED trial data
showed an inverse association in high-risk individuals
TEK Image/Science Photo Library
between monosaturated and polyunsaturated fats and
cardiovascular disease and death, whereas saturated
fat and trans fat were associated with increased risk of
cardiovascular disease.5 No effect of saturated fat is seen
in cohort studies which do not take into account nature
of replacement energy source.6 Some cohort studies
show that saturated fat from meat but not from dairy
food is associated with increased cardiovascular risk.7
The importance of considering the nature of
replacement energy is confirmed by short-term studies
that examine the effects of various dietary fatty acids
on lipids, lipoproteins, and the ratio of total:HDL
cholesterol.8 Replacement of trans and saturated fat with
carbohydrate or cis-unsaturated fat reduces total and LDL
cholesterol.8 However the total:HDL cholesterol ratio and
apolipoprotein B are reduced only when cis-unsaturated
fatty acids replace saturated fatty acids.8 The four major
saturated fatty acids (lauric, myristic, palmitic, and
stearic acids) have different effects. Lauric acid (found
in coconut oil) has a smaller effect on LDL cholesterol
than myristic and palmitic acids, whereas stearic acid has
no such effect.8 Ecological data also contribute to this
evidence base. In Finland, 40% of the decline in coronary
mortality between 1982 and 1997 was attributed to
the decline in cholesterol, 60% of which was attributed
851