Review Obesity

CLINICAL TRIALS AND INVESTIGATIONS

The Long-Term Effect of Energy Restricted Diets

for Treating Obesity
Mirjam Langeveld1 and J. Hans DeVries2

Objective: To quantify the effectiveness of diet interventions by systematic review of the long-term
effects of energy restricted diets in individuals with overweight and obesity.
Methods: A systematic literature search was conducted for all clinical trials studying the effect of energy
restricted diets on body weight in individuals with a BMI>25 kg/m2 with a follow-up of at least 3 years
and a minimum of 50 participants. Weight change from baseline and rates of loss to follow-up at the lon-
gest follow-up were extracted and analyzed using random-effects models.
Results: Weighted mean weight loss after 3 years follow-up was 3.5% (95% CI 0.2-6.8%) (n 5 6,163)
and after 4 years follow-up 4.5% (95% CI 4.3-4.8%) (n 5 5,696). Energy restricted diets (n 5 1,433)
resulted in an average 2.9% (95% CI 23.8 to 22.1%) greater weight loss compared to untreated con-
trol groups (n 5 1,361). Weight regain was observed in the majority of individuals in all studies. Interpre-
tation of the data is limited by high rates of loss to follow-up and lack of truly untreated control groups.
Conclusions: On average, the long-term effect of diets on body weight in individuals with obesity is
modest, and the response is highly heterogeneous.
Obesity (2015) 23, 1529–1538. doi:10.1002/oby.21146

Introduction US, an additional three weight loss drugs, phentermine–topiramate

extended-release combination, bupropion–naltrexone combination,
A surplus of body fat can have negative cardiovascular and meta- and lorcaserin, have recently been approved by the FDA. These
bolic consequences. The higher the body mass index (BMI), the drugs seem effective in inducing weight loss, though data up to only
greater the risk of type 2 diabetes, hypertension, dyslipidemia, and 2 years are available and long-term cardiovascular safety as well as
obstructive sleep apnea syndrome (1). Changes in carbohydrate and improvement in cardiovascular outcome still need to be established
fat metabolism and blood pressure regulation that underlie this (4,5).
increased risk are, at least partly, reversible through weight loss,
resulting in positive changes in blood pressure, fasting blood glu-
cose, and lipid profile (2). First-line treatment of patients with overweight and obesity who are
thought to benefit from weight loss is an energy restricted diet.
Though numerous publications show positive short-term results of
Reduction in the excess of stored energy, mainly fat mass, can be energy reduced diets on body weight, in practice both doctors and
achieved through bariatric surgery, weight loss drugs, or energy patients are often disappointed by the results in the long run. This
restricted diets. Bariatric surgery is highly effective in lowering raises the question whether energy restricted diets are effective in
body weight and reducing the negative metabolic and cardiovascular establishing long-term weight loss as well as a reduction in obesity-
consequences of obesity (3). The invasive nature of this treatment, associated morbidity and mortality. There is no recent full-scope
the associated potential surgical complications, and the perioperative review focusing on the long-term effects of energy restricted diets.
morbidity currently do not make this the first option in treatment of In published systematic reviews of ‘‘long-term’’ effects of diets on
obesity. Though several weight loss drugs have been shown to be weight loss, the median follow-up duration ranges from 12 to 36
effective in facilitating weight loss, many of them are associated months (see Supporting Information Table 1) (6-12). The most
with (severe) side effects and have not been shown to reduce cardio- recent of these reviews was published in 2011 (9) and does not
vascular morbidity and mortality (4). At this moment, the only include the now available data of the two largest dietary weight loss
weight loss drug used in Europe is orlistat, which reduces fat uptake studies: the Diabetes Prevention Program Outcomes Study (13) and
from food but has only a modest effect on body weight (4). In the the Look AHEAD study (14). The latter study, being the largest

1
Division of Pharmacology, Vascular and Metabolic Diseases, Department of Internal Medicine, Erasmus Medical Center, Rotterdam, The Netherlands.
Correspondence: M. Langeveld ([email protected]) 2 Department of Endocrinology, Academic Medical Center, Amsterdam, The Netherlands.

Disclosure: The authors declared no conflict of interest.

Author contributions: ML and JhDV designed the study, ML collected and analysed the data, and ML and JhDV interpreted the results and wrote the manuscript.
Additional Supporting Information may be found in the online version of this article.
Received: 1 October 2014; Accepted: 8 April 2015; Published online 14 July 2015. doi:10.1002/oby.21146

www.obesityjournal.org Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 1529

 1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Obesity Effect of Diets for Treating Obesity Langeveld and DeVries

TABLE 1 MEDLINE search strategy

(((((((‘‘Overweight’’[Mesh] OR overweight[tiab] OR over weight[tiab] OR obes*[tiab] OR antiobes*[tiab])) AND (‘‘Diet Therapy’’[Mesh] OR ‘‘Weight
Loss’’[Mesh] OR ‘‘Weight Reduction Programs’’[Mesh] OR diet[tiab] OR diets[tiab] OR dietary[tiab] OR weight intervention*[tiab] OR weight
loss*[tiab] OR weight maintenance[tiab] OR weight management[tiab] OR weight program*[tiab] OR weight reducing[tiab] OR weight
reduction*[tiab])) AND (longterm[tiab] OR long-term[tiab] OR lasting[tiab] OR longlasting[tiab])) AND (‘‘Adult’’[Mesh] OR adult*[tiab])) AND
(‘‘Clinical Trial’’[Publication Type] OR ‘‘Clinical Trials as Topic’’[Mesh] OR random*[tiab] OR trial[tiab] OR groups[tiab] OR ‘‘Cohort Studies’’[Mesh]
OR longitudinal[tiab] OR prospective[tiab] OR retrospective[tiab] OR follow up[tiab])) NOT (animals[Mesh] NOT humans[Mesh]))

weight loss study executed so far, shows that rate of weight regain roid disease, and hypopituitarism. If data on more than one time
is very high between 2 and 3 years after start of the weight loss point after 3 years were available we report the last observation of
intervention. After 3 years, though weight regain still takes place, the study. All data on study duration are reported from start of the
the rate levels off, and an estimate of the long-term effect on body weight loss intervention. Reviews identified through this search
weight can be made. Since we were primarily interested in the long- were checked for relevant references.
term outcome of weight loss diets, hypothesizing that this is the
major determinant of the health benefit of these interventions, we Main reasons for exclusion during abstract screening were: different
decided to assess the outcome after a minimum follow-up of 3 intervention (40%, mainly surgical treatment of obesity), follow-up
years. too short (19%), and different population (10%). Main reasons for
exclusion during selection based on full-text articles were different
The aim of the analysis was to enable clinicians to give advice on population (29%), follow-up too short (16%), too few participants
the chance of long-term success of an energy restricted diet to an (13%), and different intervention (9%) (Figure 1). Follow-up of only
individual seeking treatment for obesity. We therefore wanted to a part of the initial study population was also a frequent reason for
identify studies that report the percentage of the participants suc- exclusion during the latter selection.
cessful at reaching a specific weight loss target. Since only four
studies reported the percentage of individuals reaching weight loss
targets (14-17), we broadened our search to studies reporting the
mean weight loss after a minimum of 3 years follow-up.

In this meta-analysis, we examine the evidence for establishing

long-term weight loss by energy restricted diets. In addition, the bio-
logical mechanisms behind the frequently observed weight regain
after initial successful weight loss are discussed.

Methods
We performed a review of the literature searching PubMed (for
search terms see Table 1) and the Cochrane Library. Articles written
in English published online up to February 2014 were included.

Studies included in the review should have a minimum follow-up

duration of 3 or more years after start of the diet and participants
who were 18 years or older and either overweight (BMI>25 kg/m2)
or obese (BMI>30 kg/m2), with or without concurrent metabolic ill-
nesses such as diabetes or hypertension. Studies with a minimum of
50 participants were included. Included studies often prescribed diet
in combination with exercise advice and/or a form of behavioral
therapy. We included all dietary interventions aimed to restrict
energy intake, irrespective of the composition of the diet. Studies
that reassessed only a subset of participants at any moment after
conclusion of the original study were excluded. Studies performed
in populations with conditions or diseases that influence body
weight regulation, either in itself or due to medication used, were
excluded from the review. As such we excluded studies conducted
during pregnancy or directly postpartum as well as studies in indi-
viduals with (cured) cancer, psychiatric illnesses, monogenetic obe- Figure 1 Flowchart of studies included in the meta-analysis.
sity, organ transplantation, heart failure, chronic kidney failure, thy-

1530 Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 www.obesityjournal.org

 1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Review Obesity
CLINICAL TRIALS AND INVESTIGATIONS

A random-effects meta-analysis for the weight loss difference was Three out of these four studies also report mean weight loss and are
done for those studies in which diet intervention was compared to a thus also included in the next paragraph and Table 2 (14-16).
control group. We calculated pooled estimates of the mean weight
loss for studies with a follow-up duration of 3 and 4 years by a
random-effects model (DerSimonian–Laird method). Only one study Mean weight loss
reported weight loss at both 3 and 4 years and is thus included at Fourteen studies reporting mean weight loss that met the selection
both time points (18). If the study had more than one intervention criteria were identified (13-16,19-28). These studies express the
arm, results from all intervention arms were included. Weight loss results in different units: average weight loss was reported in kilo-
corrected for loss to follow-up was calculated for those studies for grams, kg/m2 or percentage of initial weight, limiting comparability.
which percentage weight loss and percentage loss to follow-up was We therefore recalculated weight loss to mean percentage weight
known, and correction was not applied in the original article. Cor- loss if possible. Looking at studies that report both kilograms and
rection was made by the ‘‘baseline value carried forward approach,’’ percentage weight loss our calculated percentage weight loss closely
assuming weight regain to baseline level with an effective weight mirrors mean percentage weight loss calculated from original patient
change of 0% in those individuals who were lost to follow-up. To data (for comparison see Supporting Information Table 1).
determine the correlation between study size and percentage weight
loss, a Spearman’s rank correlation coefficient was calculated. Cor- Weighted mean weight loss after 3 years was 3.5% (95% CI 0.2-
relations between intervention duration, study duration, or percent- 6.8%) (average of five studies, total number of participants 6,163)
age loss to follow-up and percentage weight loss were assessed by (18,22-24,27). Weighted mean weight loss after 4 years was 4.5%
weighted regression analysis. (95% CI 4.3-4.8%) (an average of five studies, total number of par-
ticipants 5,696) (15,16,18,25,28).

Loss to follow-up
Results Loss to follow-up was high in most studies reported, ranging from 4
First we discuss the studies that report the percentage of participants to 39% and was over 10% in 10 out 14 studies (Table 2). In the
successful in losing a predefined percentage of body weight. Next, majority of studies, no correction for loss to follow-up was made
the results from studies reporting mean weight loss are presented. (Table 2). As recently discussed by Ware, the most rational
The following paragraphs deal with rates of loss to follow-up, approach to these missing data might be the baseline value carried
weight regain, and the use of control groups in the included studies. forward approach, which assumes that all participants lost to follow-
Finally we summarize the current understanding of the biological up have fully regained the weight they lost initially (29). We applied
mechanisms mediating weight regain. this correction to studies in which no correction for loss to follow-
up was made (Table 2).

Studies reporting a specified percentage of Weight regain

weight loss Weight regain after initial weight loss is observed in the majority of
Four publications report the percentage of participants reaching a individuals in all 14 studies included in this review. Table 3 shows
specific weight loss target. The first one is the 8-year follow-up the mean percentage of initial weight lost that is regained during
report on the Look AHEAD study, in which 5,145 participants with follow-up, ranging from 0 to 120%. Only for two small subgroups
diabetes and overweight or obesity were included (14). The study (n 5 159 and n 5 26) in two different studies, 0% weight regain was
had two arms, one arm comprising an intensive program consisting reported (21,24). Mean percentage weight regained in the subgroups
of frequent meetings, personal diet advice, and a thorough exercise of the other 12 studies was highly variable, ranging from 9 to 120%
program. The second arm was a less intensive program of initially (Table 3). In 7 out of 14 studies, weight was reported at more than
3, and later on 1, diet and exercise advice meetings per year. After two time points. Weight was the lowest in the first year in six stud-
8 years follow-up, 5% weight loss was achieved in 36% of partici- ies and in the second year in one study. The lowest weight time
pants in the less intensive and 50% in the intensive arm. Weight point coincided with the end of the weight loss intervention in two
gain above baseline weight occurred in 39% of the less intensively studies; in five studies the weight loss intervention was ongoing at
treated and 26% of the intensively treated individuals. Overall, that time point. Weight loss course in all study groups for which a
weight loss was 2.6% greater in the intensively treated group, and percentage weight loss could be calculated at more than two time
loss to follow-up was 6%. The second study is a 4-year study in 481 points is depicted in Figure 2.
postmenopausal women with overweight and obesity; 42% of the
participants were successful in losing 3-11% of their initial weight,
and the percentage successful in losing at least 12% was 15% (15). Control group
Weight gain of at least 3% was observed in 19% of the participants; The issue of weight regain and the fact that a substantial part of par-
loss to follow-up was 18%. In the third study, including 100 individ- ticipants end up significantly heavier than before the start of the
uals with overweight or obesity, which was funded by the manufac- intervention emphasizes the need for control groups in weight loss
turer of the meal replacement it studied, 56% of the participants studies. By comparing the weight of individuals at the end of long-
were reported to have lost at least 5% of their initial weight after 4 term weight loss studies to untreated controls would show to which
years follow-up. Weight gain was observed in 8%; and loss to extent the intervention resulted in lower weight and whether that
follow-up was 25% (16). The fourth study reports 50 out of 122 par- gain weight during these interventions do so at a higher rate com-
ticipants (41%) being successful at losing at least 2 kg (2.6%) of pared to background weight gain. Control groups were included in 6
their body weight after a 3-year-long long diet intervention (17). out 14 studies discussed in this review (15,19,21-23,27). However,

www.obesityjournal.org Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 1531

 1532
TABLE 2 Studies reporting weight loss in response to an energy restricted diet with a follow-up of  3 years
Obesity

Duration Mean % or kg
Study Intervention Population intervention/follow-up weight change Loss to follow-up

Eriksson and Lindgarde, Energy restricted diet and Patients with newly detected 1 year/6 years Group A: 23.7 6 4.8% (SD) Loss to follow-up 11%
1991/Diabetologia (19) exercise type 2 diabetes (group A, Group B: 22.3 6 5.4% (SD) No correction for loss to
n 5 41), individuals with follow-up
impaired glucose tolerance Corrected % weight lossa:
(intervention group B Group A: 23.2%
n 5 181), control (group Group B: 22.1%
C, n 5 79), control group Group C: 10.5 6 4.1% (SD) Group C: 10.5%
of normal-weight subjects Group D: 11.7 6 4.5% (SD) Group D: 11.5
(group D, n 5 114)
Hakala et al., 1993/In J Energy restricted diet and Individuals with obesity and 2 years/5 years Group A: 22.4 kgb Loss to follow-up 22%

Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015

Obes Relat Metab Disord group counseling (e.g., without concurrent Group B: 27.1 kgb No correction for loss to
(20) physical activity) (group A, illnesses follow-up
n 5 30), energy restricted Corrected weight lossa:
diet and individual Group A: 21.9 kg
counseling (group B, Group B: 25.5 kg
n 5 30)
Manning et al., 1998/Diab Energy restricted diet (group Patients with diabetes and 1 years/4 years Group A: 21.2 6 1.7 kg (95% Loss to follow-up 10%
Med (21) A, n 5 159), energy BMI 28-45 kg/m2 CI)b Intention to treat analysis
restricted diet and Group B: 22.4 6 0.9 kg (95%
behavioral therapy (group CI)b
B, n 5 159), energy Group D: 22.8 6 0.8 kg (95%
restricted diet and 3 CI)b
months dexfenfluramine Group E: 20.8 6 1.5 kg (95%
(group C, n 5 159), CI)b
energy restricted diet and
home visits from dietician
(group D, n 5 159),
standard care control
group (group E, n 5 58)

www.obesityjournal.org
Effect of Diets for Treating Obesity Langeveld and DeVries

1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 TABLE 2 (continued).

Duration Mean % or kg
Review

Study Intervention Population intervention/follow-up weight change Loss to follow-up

Fletchner-Mors et al., Weight loss phase: Individuals with overweight 3 months weight loss/4 Group A: 23.2 6 4.9% (SEM) Loss to follow-up 25% (no
2000/Obes Res (16) Energy restricted diet (group or obesity, dissatisfied years weight maintenance Group B: 28.4 6 5% (SEM) correction)
A, n 5 50), same energy with standard energy- Transient higher dropout rate

www.obesityjournal.org
restriction with two or restricted diet plan (individuals who reentered
three meal replacements the study)
(group B, n 5 50) No control group
Corrected weight lossa:
CLINICAL TRIALS AND INVESTIGATIONS

Group A: 22.4%
Weight maintenance phase: Group B: 26.3%
same energy restriction
with one meal and one
snack replacement (group
A and B)
Stevens et al., 2001/Ann Energy restricted diet and Individuals aged 30-54 with 18 months followed by Group A: 20.2 kg (95% CI Usual care not described
Inter Med (22) exercise (group A, BMI 24.4-37.4 kg/m2 and facultative extension program 20.7 to 10.3) [0.2%] Loss to follow-up 7-8%
n 5 595), control high-normal blood until the end of the study/3 Corrected weight lossa:
(standard care, group B, pressure years Group A: 20.2 kg
n 5 596) Group B: 11.8 kg (95% CI Group B: 11.7 kg
12.2 to 11.3) [1.9%]
Lindstrom et al., 2003/Diab Energy restricted diet, Individuals aged 40-64 with 3 years/3 years Group A: 24.0 6 5.8% (SD) Loss to follow-up 13-21%
Care (23) individual counseling, and BMI>25 kg/m2 and Group B: 21.1% 6 6.2 (SD) No correction for loss to
exercise (group A, impaired glucose tolerance follow-up
n 5 265), control group Corrected weight lossa:
receiving advice on weight Group A: 23.5%
loss and exercise during a Group B: 20.9%
single session (group B,
n 5 257)
Cardillo et al., 2006/Eur Energy restricted diet, low- Individuals with 1 year/3 years Group A: 24.2 6 12.1% (SD/ Only data of completers
Rev Med Pharmacol fat (group A, n 5 26), BMI35 kg/m2 SEM?) included in analysis
Scie (24) low-carbo-hydrate diet, no Group B: 24.0 6 12.7% (SD/ Loss to follow-up unknown
fixed energy restriction SEM?)
(group B, n 5 27)
Esposito et al., 2009/Ann Energy restricted, low- Patients with type 2 diabetes 4 years/4 years Group A: 23.8 6 2.0 kg (SD) Loss to follow-up 9%
Int Med (25) carbohydrate, with BMI>25 kg/m2 aged [24.4%] No correction for loss to
Mediterranean-style diet 30 to 75 Group B: 23.2 6 1.9 kg (SD) follow-up

Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015

and exercise (group A, [23.7%] Corrected weight lossa:
n 5 108), energy Group A: 23.5 kg
restricted, low-fat diet and Group B: 22.9 kg
exercise (group B,
Obesity

1533
n 5 107)

1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 TABLE 2 (continued).

1534
Duration Mean % or kg
Obesity

Study Intervention Population intervention/follow-up weight change Loss to follow-up

Diabetes Prevention Intensive lifestyle intervention Individuals with BMI Mean total follow-up 10 Group A: 22 kgb Study has been conducted in
Program Outcomes (ILS, consisting of energy >24 kg/m2, impaired years Group B: 21 kgb (no SD/SEM two parts
Study, 2009/Lancet (13) restricted diet, exercise, glucose tolerance, age reported) Loss to follow-up 30-32%
and behavioral >25 years No correction for loss to
intervention) for 2.8 and follow-up
5.7 years (group A, Corrected weight lossa:
n 5 1,079), placebo for Group A: 21.7 kg
2.8 and ILS for 5.7 years Group B: 20.9 kg
(group B, n 5 1,082),
metformin for 2.8 years
and metformin plus ILS

Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015

for 5.7 years (group C,
n 5 1073)
Wolfson et al., 2010/Ann Energy intake reducing diet Subjects with obesity 6 months/3 years 22 kg/m2 [25.6%] (no SD/ Loss to follow-up 30%
Nutr Metab (26) and exercise (n 5 67) SEM reported) No correction for loss to
follow-up
Corrected weight lossa:
21.4 kg/m2
Silva et al., 2010/Med Sci Energy intake reducing diet Females with BMI>25 kg/m2 1 year/3 years Group A: 23.9 6 7.6% (SD) Loss to follow-up 29%
Sports Exerc (27) and exercise (group A), aged 25-50, premeno- Group B: 21.9 6 7.4% (SD) Correction for loss to follow-
control group, multiple pausal, no major illnesses up by baseline imputation
sessions of health (n 5 211) for all dropouts
education only (group B)
(exact group sizes not
mentioned)
Legenbauer et al., 2011/ Very-low-calorie diet, exer- Individuals with obesity, 1 year (3 months very-low- 23.4 6 0.9% (SE) (completers Loss to follow-up 39%
Compr Psychiatry (28) cise, and behavioral modi- aged 18-65 calorie diet and 9 months only) Intention to treat analysis
fication (n 5 250) refeeding)/4 years 24.5 6 1.4% (SE) (intention to
treat analysis)

Look AHEAD Research Energy intake reducing diet Patients with type 2 8 years/8 years Group A: 24.7 6 0.2% (SE) Loss to follow-up 6%
Group, 2014/Obesity (14) and exercise with frequent diabetes, aged 45-75, Group B: 22.1 6 0.2%(SE) Intention to treat analysis
counseling (group A, BMI >25 kg/m2 (or >27
n 5 2570), infrequent diet when taking insulin)
and lifestyle education
(group B, n 5 2575)

www.obesityjournal.org
Effect of Diets for Treating Obesity Langeveld and DeVries

1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Review Obesity
CLINICAL TRIALS AND INVESTIGATIONS

since these control groups received either standard care (for hyper-
tension and diabetes) or health education (including advice on

Loss to follow-up

No correction for loss to

Corrected weight lossa:

Loss to follow-up 18%
weight loss) on average they showed a small amount of weight loss,
in contrast to the background weight gain which is known to occur

Group A: 23.0 kg
Group B: 20.2 kg
in the general population (30,31). Mean weighted difference
between treated and control group could be calculated based on data

follow-up
from five studies, which showed an average 2.9% (95% CI 23.8 to
22.1%) greater weight loss in the intervention groups (Figure 3).
The only study with a truly untreated control group of 79 over-
weight individuals showed weight gain of 10.5 6 4.1% during 6
years follow-up, this being significantly different from the
Group B: 20.2 6 5.6 kg (SD)

22.3 6 5.4% weight loss in the treated group (19).

Group A: 23.6 6 7.3 kg (SD)
weight change
Mean % or kg

Determinants of the weight loss response

Study size was not correlated with weight loss (r2 5 20.18,
P 5 0.50). Using regression analysis corrected for study size we
[24.4%]

[20.2%]

found no significant correlation between intervention duration

(r2 5 20.038, P 5 0.514), total study duration (r2 5 20.05,
P 5 0.58) or loss to follow-up (r2 5 0.15, P 5 0.10), and percentage
weight loss.
intervention/follow-up

4 years/4 years

Mechanisms mediating weight regain

Duration

In the past decades, human and animal research has started to

unravel the mechanisms that drive weight regain after substantial
weight loss. Most importantly, total energy expenditure drops after
Percentage weight loss cannot be calculated. [] calculated percentage from average weight loss given in kg or kg/m2.

weight loss and this reduction persists during long-term mainte-

nance of lower body weight. Loss of body weight itself and thus
loss of metabolically active tissue is the first explanation for this
reduction. Total energy expenditure can be broken down into resting
energy expenditure, activity-related energy expenditure, thermal
energy of feeding and adaptive energy expenditure in response to
aged 52-62, BMI 25-
Postmenopausal women
Population

e.g. cold exposure. All these components of total energy expendi-

ture are affected by weight loss. Resting energy expenditure is
reported to be reduced after weight loss (32-34). The explanation
39.9 kg/m2

for this drop in resting energy expenditure is subject of a long-

Weight loss corrected for loss to follow-up by baseline carried forward approach.

standing scientific debate. One theory is that, after correction for

body composition, individual tissues have a lower metabolic rate
after weight loss. Others claim that if very sensitive body composi-
tion analysis is performed, the reduction can be fully accounted for
and exercise with frequent

by loss of metabolically highly active tissue such as liver and heart

Energy intake reducing diet

tissue (35).
counseling (group A,
Intervention

education (group B,
n 5 241), health

Activity-related energy expenditure may be altered by changes in

both the amount of activity and the energy spent during activity. For
example, a recent study in 66 individuals with overweight and obe-
n 5 240)

sity who had lost 13 6 4% of their body weight showed an increase

in physical activity, but a paradoxical decrease in activity energy
expenditure (36). This is again related to the weight loss itself, as
less mass needs to be moved, but beyond that metabolic efficiency
appears to be increased after weight loss. This may be the case for
Barone Gibbs et al., 2012/

processes such as utilization and storage of food as well. For

J Acad Nutr Diet (15)
TABLE 2 (continued).

reviews that deal with this topic in more detail see MacLean 2011
and Muller 2012 (37,38). Regardless of whether there are changes in
the metabolic rate of tissues or if the lower energy expenditure is
caused solely by changes in body weight and composition or meta-
bolic efficiency increases after weight loss, what is important for
Study

clinicians and patients is that total energy expenditure is signifi-

cantly reduced after weight loss. This explains why weight regain
b
a

www.obesityjournal.org Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 1535

 1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Obesity Effect of Diets for Treating Obesity Langeveld and DeVries

TABLE 3 Weight regain during follow-up

Duration intervention/
Study follow-up Time lowest weight Weight regain

Eriksson and Lindgarde, 1 year/6 years Only two time points Yes, no percentage can
1991/Diabetologia (19) reported be calculated
Hakala et al., 1993/In J Obes 2 years/4 years Between 3 months to 1 year Females group A: 86%
Relat Metab Disord (20) in the different groups Males group A: 80%
Females group B: 71%
Males group B: 51%
Manning et al., 1998/Diab Med (21) 1 years/4 years Only two time points Group A: 48%
reported Group B: 29%
Group D: 0%
Fletchner-Mors et al., 2000/Obes Res (16) 3 months weight loss/4 Between 22 and 28 months Group A: 42%
years weight maintenance Group B: 9%
Stevens et al., 2001/Ann Inter Med (22) 18 months followed by 6 months 95%
facultative extension
program until the end of
the study/3 years
Lindstrom et al., 2003/Diabetes Care (23) 3 years/3 years Only two time points reported 22%
Cardillo et al., 2006/Eur Rev 1 year/3 years Only two time points reported Group A: 0%
Med Pharmacol Sci (24) Group B: 120%
Esposito et al., 2009/Ann Int Med (25) 4 years/4 years 1 year Group A: 39%
Group B: 24%
Diabetes Prevention Program Outcomes Mean total follow-up 10 1 year Group A: 71%
Study, 2009/ Lancet (13) years
Wolfson et al., 2010/Ann Nutr Metab (26) 6 months/3 years 6 months 29%
Silva et al., 2010/Med Sci Sports Exerc (27) 1 year/3 years 1 year 47%
Legenbauer et al., 2011/Compr Psychiatry (28) 1 year (3 months very-low- Only two time points 68%
calorie diet and 9 months reported
refeeding)/4 years
Look AHEAD Research Group, 2014/Obesity (14) 8 years/8 years 1 year Group A: 45%
Barone Gibbs et al., 2012/J Acad Nutr Diet (15) 4 years/4 years Only two time points reported 54%

occurs in a weight reduced individual at a level of energy intake Studies conducted in individuals with obesity who have lost weight
that would not change body weight in someone who has always through a hypocaloric diet show an increased in hunger and reduced
been weight stable. satiety in the weight reduced state (40-43).

In addition to the decrease in energy expenditure, the drive for

energy intake increases compared to the situation before diet initia-
tion (39). Appetite and satiety are regulated in the hypothalamus
Discussion
where signals related to changes in energy balance are sensed and Limitations of the reviewed studies include:
integrated. After weight loss, one of the key signals to the hypothal-
amus is a change in adipose tissue derived cytokines and hormones 1. Few studies report a highly relevant outcome parameter—the
(adipokines), most importantly leptin. The drop in leptin levels in number of participants successful in maintaining a significantly
response to weight loss is the most important signal of depletion of lower body weight.
fat cell energy stores. Administration of leptin in weight reduced
2. Most studies lack a control group.
individuals reverses the weight loss induced changes in appetite,
3. The rate of loss to follow-up is high, falsely inflating the reported
satiety, and energy expenditure (39). In addition to adipose tissue
mean weight loss.
signals, altered secretion of gut hormones such as ghrelin and pep-
tide YY play a role in the post weight loss increase in appetite and When correcting for loss to follow-up, by the baseline value car-
reduction in satiety. These changes in hormone levels persist as the ried forward approach, the weight loss effect is significantly reduced
lower body weight is maintained (40). Overall, these signals result in all but one study (Table 2). This substantiates the observation that
in the hypothalamic perception of energy deficiency, which stimu- weight loss diet interventions in clinical practice are generally less
lates energy intake. successful compared to reports in literature.

1536 Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 www.obesityjournal.org

 1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Review Obesity
CLINICAL TRIALS AND INVESTIGATIONS

The long-term result of treatment of obesity with energy restricted

diets is heterogeneous, resulting in maintained weight loss in a
minority and gradual weight regain in the majority of individuals.
Powerful biological compensatory mechanisms hamper maintenance
of lower body weight after a diet intervention. Both reduced energy
expenditure and an increased drive for energy intake have to be
overcome constantly to prevent weight regain. Thus only those that
manage to permanently adhere to stringent lifestyle changes will be
successful in the long run. For the majority of individuals such a
stringent lifestyle change will not be feasible and they will regain
weight.

One of the key questions that remains to be answered is whether

weight loss followed by weight regain (weight cycling) in itself has
negative consequences. Though there are studies suggesting that
weight cycling is related to increased mortality (44), these data
come from observational cohort studies and it is therefore not
entirely certain whether this weight cycling was related to inten-
tional weight loss through dieting.

Since so many individuals are initially successful in losing weight

through an energy intake reducing diet, future research should aim to
identify ways to prevent weight regain. An obvious strategy would be
to increase energy expenditure, correcting the post weight loss drop.
In the past, drugs that increased energy expenditure, e.g. ephedrin, thy-
roid hormone, had unacceptable cardiovascular side effects. Current
research into stimulating brown adipose tissue in humans, thus
increasing thermogenesis, will hopefully identify safe alternatives.
One of the potential problems however could be a concomitant
increase in appetite with increased energy expenditure, as is the case
for exercise (45). A second focus of future research should be predict-
Figure 2 Weight loss over time in studies reporting body weight at at least three
time points. Thickness of the lines approximates the weight of the study based on ing the chance of permanent weight loss before the start of a diet. Re-
number of individuals included. analysis of published studies or new studies should examine whether
sex, age, diet history, or duration of the obese state are predictors of
weight loss success, so that tailored advice can be given.

Taking these limitations into account, two conclusions can be drawn. When prescribing a diet for the treatment of obesity, health practi-
First, on average, the long-term effect of diets on body weight in indi- tioners should discuss with their patients that this intervention is
viduals with obesity is modest. Second, the response is heterogeneous, only useful if permanent dietary changes are made. The high chance
a small minority of individuals will be able to maintain a significantly of weight regain and the mechanisms behind the phenomenon
lower body weight, the majority of individuals will first lose weight should be explained and a plan should be made to prevent weight
and then regain most of initial weight loss and a second minority will regain as much as possible. This plan should include a feasible
(re)gain weight up to a level that is higher than the baseline weight. weight maintenance diet, emphasizing that energy intake will have
Based on the current analysis, it is not possible to predict into which to be permanently lower compared to the situation prior to weight
category an individual will fall at start of the diet. loss. Data from individuals successful in maintaining significantly

Figure 3 Percentage weight loss in subjects treated with an energy restricted diet versus control groups. [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com.]

www.obesityjournal.org Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 1537

 1930739x, 2015, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/oby.21146 by CAPES, Wiley Online Library on [17/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Obesity Effect of Diets for Treating Obesity Langeveld and DeVries

reduced body weight for over 5 years suggest that frequent monitor- 19. Eriksson KF, Lindg€arde F. Prevention of type 2 (non-insulin-dependent) diabetes
mellitus by diet and physical exercise. The 6-year malm€ o feasibility study.
ing of body weight and engagement in high levels of physical activ- Diabetologia 1991;34:891-898.
ity aid weight maintenance (46). Whether prescribing drugs to pre- 20. Hakala P, Karvetti RL, R€ onnemaa T. Group vs. individual weight reduction
vent weight regain result in long-term health benefits remains to be programmes in the treatment of severe obesity-a five year follow-up study. Int J
determined. Disapproval of those who do not succeed to lose Obes Relat Metab Disord 1993;17:97-102.
weight, or regain weight, is inappropriate. If weight loss is not 21. Manning RM, Jung RT, Leese GP, Newton RW. The comparison of four weight
reduction strategies aimed at overweight patients with diabetes mellitus: four-year
achieved, the advice to go on a diet should not be repeated auto- follow-up. Diabet Med 1998;15:497-502.
matically, unless a renewed willingness to change is noted. For 22. Stevens VJ, Obarzanek E, Cook NR, et al. Trials for the hypertension prevention
many, weight stabilization is probably a more realistic long-term research group. Long-term weight loss and changes in blood pressure: results of the
trials of hypertension prevention, phase II. Ann Intern Med 2001;134:1-11.
goal than losing weight. O
23. Lindstr€
om J, Louheranta A, Mannelin M, et al.; Finnish Diabetes Prevention Study
Group. The finnish diabetes prevention study (DPS): lifestyle intervention and 3-
year results on diet and physical activity. Diabetes Care 2003;26:3230-3236.

Acknowledgments 24. Cardillo S, Seshadri P, Iqbal N. The effects of a low-carbohydrate versus low-fat
diet on adipocytokines in severely obese adults: three-year follow-up of a
Authors would like to thank Ingeborg Nagel for her help with the randomized trial. Eur Rev Med Pharmacol Sci 2006;10:99-106.

literature search strategy and Rob Scholten for his helpful comments 25. Esposito K, Maiorino MI, Ciotola M, et al. Effects of a Mediterranean-style diet on
the need for antihyperglycemic drug therapy in patients with newly diagnosed type
on the statistical analysis. 2 diabetes: a randomized trial. Ann Intern Med 2009;151:306-314.
C 2015 The Obesity Society
V 26. Wolfson N, Garish D, Goldberg Y, Boaz M, Matas Z, Shargorodsky M. Effect of
weight loss maintenance on arterial compliance and metabolic and inflammatory
parameters: a three-year follow-up study. Ann Nutr Metab 2010;57:204-210.
References 27. Silva MN, Markland D, Carraça EV, et al. Exercise autonomous motivation predicts
1. Malnick SD, Knobler H. The medical complications of obesity. QJM 2006;99:565-
3-yr weight loss in women. Med Sci Sports Exerc 2011;43:728-737.
579.
2. Vidal J. Updated review on the benefits of weight loss. Int J Obes Relat Metab 28. Legenbauer T, Petrak F, de Zwaan M, Herpertz S. Influence of depressive and
Disord. 2002;26:S25-S28. eating disorders on short- and long-term course of weight after surgical and
nonsurgical weight loss treatment. Compr Psychiatry 2011;52:301-311.
3. Gloy VL, Briel M, Bhatt DL, et al. Bariatric surgery versus non-surgical treatment
for obesity: a systematic review and meta-analysis of randomised controlled trials. 29. Ware JH. Interpreting incomplete data in studies of diet and weight loss. N Engl J
BMJ 2013;347:f5934 Med 2003;348:2136-2137.
4. Yanovski SZ, Yanovski JA. Long-term drug treatment for obesity: a systematic and 30. Du H, van der A DL, Ginder V, et al. Dietary energy density in relation
clinical review. JAMA 2014;311:74-86. tosubsequent changes of weight and waist circumference in european men and
women. PLoS One 2009;4:e5339.
5. Rueda-Clausen CF, Padwal RS, Sharma AM. New pharmacological approaches for
obesity management. Nat Rev Endocrinol 2013;9:467-478. 31. Klenk J, Rapp K, Ulmer H, Concin H, Nagel G. Changes of body mass index in
6. Avenell A, Brown TJ, McGee MA, et al. What are the long-term benefits of weight relation to mortality: results of a cohort of 42,099 adults. PLoS One 2014;8;e84817.
reducing diets in adults? A systematic review of randomized controlled trials. 32. Camps SG, Verhoef SP, Westerterp KR. Weight loss, weight maintenance, and
J Hum Nutr Diet 2004;17:317-335. adaptive thermogenesis. Am J Clin Nutr 2013;97:990-994.
7. Anderson JW, Konz EC, Frederich RC, Wood CL. Long-term weight-loss 33. Astrup A, Gøtzsche PC, van de Werken K, et al. Meta-analysis of resting metabolic
maintenance: a meta-analysis of US studies. Am J Clin Nutr 2001;74:579-584. rate in formerly obese subjects. Am J Clin Nutr 1999;69:1117-1122.
8. Douketis JD, Thabane L, Macie C, Williamson DF. Systematic review of long-term 34. Rosenbaum M, Hirsch J, Gallagher DA, Leibel RL. Long-term persistence of
weight loss studies in obese adults: clinical significance and applicability to clinical adaptive thermogenesis in subjects who have maintained a reduced body weight.
practice. Int J Obes Relat Metab Disord 2005; 29:1153-1167. Am J Clin Nutr 2008;88:906-912.
9. Loveman E, Frampton GK, Shepherd J, et al. The clinical effectiveness and cost- 35. Bosy-Westphal A, Braun W, Schautz B, M€ uller MJ. Issues in characterizing resting
effectiveness of long-term weight management schemes for adults: a systematic energy expenditure in obesity and after weight loss. Front Physiol 2013;4:47.
review. Health Technol Assess 2011;15:1-182.
36. Bonomi AG, Soenen S, Goris AH, Westerterp KR. Weight-loss induced changes in
10. Norris SL, Zhang X, Avenell A, Gregg E, Schmid CH, Lau J. Long-term non-
physical activity and activity energy expenditure in overweight and obese subjects
pharmacological weight loss interventions for adults with prediabetes. Cochrane
before and after energy restriction. PLoS One 2013;8:e59641.
Database Syst Rev 2005;2:CD005270.
37. MacLean PS, Bergouignan A, Cornier M-A, Jackman MR. Biology’s response to
11. WuT Gao X, Chen M van Dam RM. Long-term effectiveness of diet-plus-exercise
dieting: the impetus for weight regain. Am J Physiol Regul Integr Comp Physiol
interventions vs. diet-only interventions for weight loss: a meta-analysis. Obes Rev
2011;301:R581-R600.
2009;10:313-323.
12. Franz MJ, VanWormer JJ, Crain AL, et al. Weight-loss outcomes: a systematic 38. Muller MJ, Bosy-Westphal A. Adaptive thermogenesis with weight loss in humans.
review and meta-analysis of weight-loss clinical trials with a minimum 1-year Obesity 2013;218-228.
follow-up. J Am Diet Assoc 2007;107:1755-1767. 39. Rosenbaum M, Kissileff HR, Mayer LES, Hirsche J, Leibela RL. Energy intake in
13. Diabetes Prevention Program Research Group, Knowler WC, Fowler SE, et al. 10- weight-reduced humans. Brain Res 2010;1350:95-102.
year follow-up of diabetes incidence and weight loss in the Diabetes Prevention 40. Sumithran P, Prendergast LA, Delbridge E, et al. Long-term persistence of
Program Outcomes Study. Lancet 2009;14:1677-1686. hormonal adaptations to weight loss. N Engl J Med 2011;365:1597-1604. :
14. Look AHEAD Research Group. Eight-year weight losses with an intensive lifestyle 41. Rodrıguez-Rodrıguez E, Aparicio A, Bermejo LM, L opez-Sobaler AM, Ortega RM.
intervention: the Look AHEAD study. Obesity 2014;22:5-13. Changes in the sensation of hunger and well-being before and after meals in overweight/
15. Barone Gibbs B, Kinzel LS, Pettee GK, Chang YF, Kuller LH. Short- and long- obese women following two types of hypoenergetic diet. Public Health Nutr 2009;12:44-50.
term eating habit modification predicts weight change in overweight, 42. Chaput JP, Drapeau V, Hetherington M, Lemieux S, Provencher V, Tremblay A.
postmenopausal women: results from the WOMAN study. J Acad Nutr Diet 2012; Psychobiological effects observed in obese men experiencing body weight loss
112:1347-1355. plateau. Depress Anxiety 2007;24:518-521.
16. Flechtner-Mors M, Ditschuneit HH, Johnson TD, Suchard MA, Adler G. Metabolic 43. Doucet E, Imbeault P, St-Pierre S, et al. Appetite after weight loss by energy
and weight loss effects of long-term dietary intervention in obese patients: four-year restriction and a low-fat diet-exercise follow-up. Int J Obes Relat Metab Disord
results. Obes Res 2000;8:399-402. 2000;24:906-914.
17. Chae JS, Paik JK, Kang R, et al. Mild weight loss reduces inflammatory cytokines,
44. Rzehak P, Meisinger C, Woelke G, et al. Weight change, weight cycling and
leukocyte count, and oxidative stress in overweight and moderately obese
mortality in the ERFORT male cohort study. Eur J Epidemiol 2007;22:665-673.
participants treated for 3 years with dietary modification. Nutr Res 2013;33:
195-203. 45. Thomas DM, Bouchard C, Church T, Slentz C, Kraus WE, Redman LM. Why do
18. Wadden TA, Neiberg RH, Wing RR, Clark JM, Delahanty LM, Hill JO. Look individuals not lose more weight from an exercise intervention at a defined dose?
AHEAD Research Group. Four-year weight losses in the Look AHEAD study: An energy balance analysis. Obes Rev 2012;13:835-847.
factors associated with long-term success. Obesity (Silver Spring) 2011;19:1987- 46. Wyatt HR, Phelan S, Wing RR, Hill JO. Lessons from patients who have
1998. successfully maintained weight loss. Obesity Manage 2005;1:56-61.

1538 Obesity | VOLUME 23 | NUMBER 8 | AUGUST 2015 www.obesityjournal.org