Full Terms & Conditions of access and use can be found at
Critical Reviews in Food Science and Nutrition
ISSN: 1040-8398 (Print) 1549-7852 (Online) Journal homepage: http://www.tandfonline.com/loi/bfsn20
The effects of caffeine intake on weight loss: a
systematic review and dos-response meta-analysis
of randomized controlled trials
Reza Tabrizi, Parvane Saneei, Kamran B Lankarani, Maryam Akbari, Fariba
Kolahdooz, Ahmad Esmaillzadeh, Somayyeh Nadi-Ravandi, Majid Mazoochi
& Zatollah Asemi
To cite this article: Reza Tabrizi, Parvane Saneei, Kamran B Lankarani, Maryam Akbari, Fariba
Kolahdooz, Ahmad Esmaillzadeh, Somayyeh Nadi-Ravandi, Majid Mazoochi & Zatollah Asemi
(2018): The effects of caffeine intake on weight loss: a systematic review and dos-response meta-
analysis of randomized controlled trials, Critical Reviews in Food Science and Nutrition, DOI:
To link to this article: https://doi.org/10.1080/10408398.2018.1507996
Published online: 18 Oct 2018.
Submit your article to this journal
Article views: 171
View Crossmark data
The effects of caffeine intake on weight loss: a systematic review and
dos-response meta-analysis of randomized controlled trials
, Parvane Saneei
, Kamran B Lankarani
, Maryam Akbari
, Fariba Kolahdooz
, Somayyeh Nadi-Ravandi
, Majid Mazoochi
, and Zatollah Asemi
Health Policy Research Center, Institute of Health, Student Research Committee, Shiraz University of Medical Sciences, Shiraz, Iran;
Security Research Center, Department of Community Nutrition School of Nutrition and Food Science, Isfahan University of Medical Sciences,
Health Policy Research Center, Shiraz University of Medical Sciences, Shiraz, Iran;
Indigenous and Global Health Research,
Department of Medicine, University of Alberta, Edmonton, Canada;
Department of Community Nutrition School of Nutritional Sciences and
Dietetics, Tehran University of Medical Sciences, Tehran, Iran;
Health Information Management Research Center, Kashan University of
Medical Sciences, Kashan, Iran;
Department of Cardiology School of Medicine, Kashan University of Medical Sciences, Kashan, Iran;
Research Center for Biochemistry and Nutrition in Metabolic Diseases, Kashan University of Medical Sciences, Kashan, Iran
This systematic review and meta-analysis of randomized controlled trials (RCTs) was performed to
summarize the effect of caffeine intake on weight loss. We searched the following databases until
November 2017: MEDLINE, EMBASE, Web of Science, and Cochrane Central Register of Controlled
Trials. The relevant data were extracted and assessed for quality of the studies according to the
Cochrane risk of bias tool. We estimated an intake-status regression coefficient (Beta) for each
primary study and estimated the overall pooled Beta and SE using random effects meta-analysis
on a double-log scale. Heterogeneity between studies was assessed by the Cochran Q statistic
and I-squared tests (I
). Thirteen RCTs with 606 participants were included in the meta-analyses.
The overall pooled Beta for the effect of caffeine intake was 0.29 (95%CI: 0.19, 0.40; Q ¼124.5,
¼91.2%) for weigh, 0.23 (95%CI: 0.09, 0.36; Q ¼71.0, I
¼93.0%) for BMI, and 0.36 (95% CI: 0.24,
0.48; Q ¼167.36, I
¼94.0%) for fat mass. For every doubling in caffeine intake, the mean reduction
in weight, BMI, and fat mass increased 2 Beta-fold (20.29 ¼1.22, 20.23 ¼1.17, and 20.36 ¼1.28),
which corresponding to 22, 17, and 28 percent, respectively. Overall, the current meta-analysis
demonstrated that caffeine intake might promote weight, BMI and body fat reduction.
Caffeine; weight loss;
An estimated 64% of American populations are overweight or
obese [body mass index (BMI) 25 kg/m
] (Flegal et al. 2002).
eases, including coronary heart diseases (CHD), hypertension,
type 2 diabetes mellitus (T2DM), pulmonary dysfunction, and
non-metabolic such as osteoarthritis, and certain types of can-
cer (Kromhout 1983;Lynchetal.2009). Common treatments
for managing obesity include lifestyle changes such as weight
loss, appropriate diet, and increased physical activity, as well as
the appropriate use of pharmacological agents to reduce the
specific risk factors (Villareal et al. 2011).
Modest weight loss, 5–10% of the initial body weight,
would result in beneficial health effects (Wing et al.
1992).Caffeine has been widely used as an practical approach
in obesity management (Astrup 2000). Caffeine increases both
noradrenaline and dopamine release, and therefore stimulates
the neuronal activity in several brain regions (Zheng and
fat. Caffeine may increase fat oxidation through inhibiting
phosphodiesterase and the suppression of negative effects of
adenosine on increased noradrenaline release (Dulloo,
Seydoux, and Girardier 1992). Despite reported anti-obesity
effects and weight maintenance of caffeine in some clinical
trials (Boozer et al. 2002; Coffey et al. 2004;Molnaretal.
2000; Westerterp-Plantenga, Lejeune, and Kovacs 2005), few
studies did not show any beneficiary effect of caffeine for
body weight and weight maintenance after weight loss (Hursel
and Westerterp-Plantenga 2009; Lee et al. 2005).Therefore, the
effect of caffeine consumption to improve weight loss diet and
reduce percentage of body fat remains controversial.
We are aware of no systematic review and meta-analysis
of RCTs about the effect of caffeine intake on weight loss.
This meta-analysis was performed to summarize the avail-
able evidence of RCTs to investigate the effect of caffeine
intake on weight loss.
Search strategy and study selection
This meta-analysis was undertaken according to PRISMA
(Preferred Reporting Items for Systematic Reviews and
CONTACT Zatollah Asemi firstname.lastname@example.org Research Center for Biochemistry and Nutrition in Metabolic Diseases, Kashan University of Medical
Sciences, Kashan, Iran.
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/bfsn.
ß2018 Taylor & Francis Group, LLC
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION
Meta-Analyses) guideline. Two authors were independently
conducted the search, data selection, data extraction, and
evaluation of risk of bias. In the case of a disagreement, it is
resolved by consensus and/or discussion with a third author.
The following online databases were searched for relevant
RCTs studies published through November 2017: Cochrane
Library, EMBASE, MEDLINE, and Web of Science data-
bases. In addition, ongoing trials were searched from data-
bases including the International Standard Randomized
Controlled Trial Number Register and Meta-register for
RCTs. In addition, we did not publish the review protocol.
We conducted searches on gray literature using databases
including institute for scientific and technical information
(INIST) and the healthcare management information con-
sortium (HMIC), also to find other unpublished studies, we
contacted with experts and centers of related field. Trials
retrieved that examined the effect of caffeine intake on
weight, BMI and/or body fat by using the following MeSH
and text words [(“caffeine”[Mesh] OR “coffee”[Mesh] OR
caffeine [tiab] OR coffee [tiab]) OR caffeinated beverages
[tiab] AND (“body weight”[Mesh] OR “body mass
index”[Mesh] OR “body fat”[Mesh] OR body weight [tiab]
OR body mass index [tiab] OR body fat [tiab]) AND
(randomized clinical trial [pt] OR controlled clinical trials
[pt] OR randomized [tiab] OR placebo [tiab] OR randomly
[tiab] OR trial [tiab])].
The reference lists of all known related studies, including
original research papers and review articles, were reviewed as
an additional manual search. Trials applied that were pub-
lished in the English language without time restrictions for
Figure 1. Literature search and review flowchart for selection of studies.
2 R. TABRIZI ET AL.
Table 1. The effect of caffeine on weight, BMI and body fat reduction: overview of selected studies for dose-response meta-analysis.
Authors (y) Country
year Age (y) Gender
intervention) RCT type
(name and daily dose)
Colker et al. (1999) USA 1999 Range: >21 y F/M 7/9 Parallel No 975 C aurantium, 528 mg
C, 900 mg St. Johns
6 Weight, body fat Overweight
Molnar et al. (2000) Hungary 2000 Mean: 16 ± 1 F/M
13/16 Parallel Yes <80 kg:300 mg C/30 mg
Or >80 kg: 600 mg
C/60 mg E
450 Placebo 20 Weight, BMI,
Boozer et al. (2002) USA 2002 Range: 18-80 F/M 38/39 Parallel Yes 192 mg C/90 mg E 192 Placebo 24 Weight, body fat Obese
Boozer et al. (2002) USA 2002 Range: 25-55 F/M 24/24 Parallel No 240 mg C/72 mg E 100 Placebo 8 Weight, body fat Overweight/
Coffey et al. (2004) USA 2004 Range: 18-65 F/M 50/52 Parallel No 360 mg C/60 mg E/
90 mg salicin
360 Placebo 12 Weight, BMI,
Greenway et al. (2004) USA 2004 Range: 18-65 F/M 20/20 Parallel Yes 210 mg C/72mg E 210 Placebo 12 Weight Healthy
Hackman et al. (2006) USA 2006 Range: 25-74 F 23/19 Parallel No 100 mg C/40 mg E 100 Placebo 36 Weight, body fat Overweight/
Thom (2007) Norway 2007 Mean: 24.2 ±3.2 F/M 15/15 crossover No 10 g of Coffee Slender 528 10 g of decaf-
12 Weight, body fat Overweight/
2009 Mean: 44 ± 2 F/M 20/20 Parallel No 150 mg C/270 mg
150 Placebo 13 Weight, BMI,
Bakuradze et al. (2011) Germany 2011 Range: 20-44 M 33/33 Consecuti-
No 750 ml freshly
720 Water 8 Weight, BMI,
Liu et al. (2013) USA 2013 Range: 18-60 F/M 26/17 Parallel Yes 600 mg C/60 mg E 600 60 mg Leptin 24 Weight, body fat Obese
Bracale et al. (2014) Italy 2014 Mean:
36.3 ± 10.3
F 7/6 Parallel Yes 60 mg C/600 mg E 60 Placebo 4 Weight, BMI Obese
Davoodi et al. (2014) Iran 2014 Mean:
39.22 ± 5.81
F 30/30 Parallel Yes 5mg C/kg BW 425 Nothing 6 Weight, BMI,
F, Female; M, Male.
C, caffeine; E, ephedrine; OBLI, online basic lifestyle information; OBWM, online behavioral weight management; BEV, fortified diet cola beverage, soluble fiber dextrin and caffeine.
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 3
publication. Two authors (RT, MA) independently selected
studies in a two-steps process. In the first stage, authors
screened the titles and/or abstract for eligible trials. In the
second stage, the full-texts of related studies were retrieved to
assess the eligibility of selected studies using inclusion and
exclusion criteria. Studies did not contain proper data to be
included in the met-analysis, though presented the other
inclusion criteria were considered for a qualitative analyses to
help identify confounding parameters. Trials that met the
following criteria were selected for meta-analysis: (1) human
RCTs; (2) intervention group consumed caffeine or caffein-
ated coffee, whereas the control group received placebo; and
(3) the trials reported mean changes or mean difference
of weight and/or BMI and/or body fat loss with standard
deviation (SD) for the intervention and control groups or
reported enough data acquirable to assess Beta (B) and its
Standard Error (SE) for the assumed linear regression on the
Data extraction and quality assessment
Two independent authors (ZA and MA) extracted data from
each trial. The Cochrane Collaboration risk of bias tool was
used to assess the quality of all relevant RCTs based on the
following domains: random sequence generation, allocation
concealment, blinding of participants and outcome assess-
ment, incomplete outcome data, and selective outcome
reporting, and other sources of bias. The following data were
extracted: first authors’name, publication year, age, gender,
country, sample size, study design, energy restriction, the
dose of caffeine intake in intervention and control groups,
the duration of intervention, the mean change and standard
deviation on weight, BMI, and body fat in intervention and
control groups at the end of the intervention. We converted
the reported dosage of caffeine intake into milligram per day
We estimated a caffeine intake-status regression coefficient
(Beta) for each primary study on the baseline value when
the assumption of a linear correlation on the log
exists; caffeine intake compared with mean change weight,
BMI, and body fat. Algebraically derivation of an estimate
from primary study of the Beta and its SE, we compared
findings from studies with heterogeneously reported associa-
tions and the associated effects. The overall pooled Beta and
its SE were estimated by using random effects meta-analysis
following DerSimonian and Laird method (DerSimonian
and Laird 1986). In the meta-regression model, the equation
intercepts were calculated by using the meanX (the mean of
the caffeine intake on the ln scale) and the meanY (the
change of mean for weight, BMI, and fat mass status on the
ln scale) for every trial and weighted these by multiplying
with the weighting agent of the trial. Finally, we took the
mean of the weighted meanX and mean of the weighted
meanY as the coordination point at that the regression lines
were hung up. This point, together with Betas, presented the
intercept. All statistical transformation to provide Beta and
its SE were conducted using the Microsoft Excel version 7.0
(Microsoft, Inc). For each trial, we weighted these by multi-
plying with the weighting agent of the trial. Heterogeneity
between studies was assessed by Cochran Q test and I-
squared statistic (I
higher than 50 percent with P-value
<0.05 represented significant heterogeneity. Potential source
of heterogeneity between studies such as duration of study
(12 weeks vs. <12 weeks), geographic area (US vs. non-
US), gender (female vs. male), and energy restriction (no vs.
yes) were examined by using subgroup analysis based on
stratified random effects meta-analysis. We used the median
of intervention duration, which was 12 weeks to do sub-
group analyses in order to have equal distribution of data
for comparison analyses. For linear dose response analyses,
we used the transformations method to derive coherent
single-study calculates from available summary statistics
(Souverein et al. 2012). The present study applied a base–e
logarithmic transformation on the caffeine intake and mean
change of weight, BMI, and body fat before estimation of
trial-specific Betas, therefore the overall Beta provides the
difference in the loge-transformed and predicted mean
change of weight, BMI, and body fat for each 1 unit differ-
ence in the loge transformed value in caffeine intake.
Egger’s test was used to detect the existence of potential
publication bias for the primary outcome measure. To meas-
ure the pooled estimates, nonparametric test (Duval and
Figure 2. The methodological quality of included studies based on review authors’judgments about each risk of bias item presented as percentages across all
4 R. TABRIZI ET AL.
Figure 3. A–C. Random effects meta analysis of 13 randomized controlled trials that examined the association or effect of caffeine intake on change mean
(A) weight, (B) for BMI, (C) for body fat in intervention and control groups by using regression coefficients (Bets) for the liner association between loge-transformed
caffeine intake and loge transformed change mean in weight, BMI, and fat mass status (CI ¼95%).
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 5
Tweedie) was used. We used STATA version 12.0 (Stata
Corp., College Station, TX) and RevMan V.5.3 software
(Cochrane Collaboration, Copenhagen, Denmark) for data anal-
yses. P-Values <0.05 were considered as statistically significant.
Our initial search found 945 potential citations, after screen-
ing 13 trials with total of 606 participants were potentially
relevant and was included in the meta-analysis. Figure 1
shows the details of the study selection and Table 1 shows
the characteristics of the included studies that were pub-
lished between 1999 to 2014. Sample size varied between 13
to 102 participants. A consecutive design was performed in
one study, cross-over design in one study, and parallel group
design in the remaining 11 trials. To incorporating cross-
over trials, we were included only data from the first period.
Twelve trials have reported mean changes on weight
(Bakuradze et al. 2011; Boozer et al. 2002; Bracale et al.
2014; Coffey et al. 2004; Colker et al. 1999; Davoodi et al.
2014; Greenway et al. 2004; Hackman et al. 2006; Hursel
and Westerterp-Plantenga, 2009; Liu et al. 2013; Molnar
et al. 2000; Thom, 2007), six on BMI (Bakuradze et al. 2011;
Bracale et al. 2014; Coffey et al. 2004; Davoodi et al. 2014;
Hursel and Westerterp-Plantenga, 2009; Molnar et al. 2000),
and one on body fat loss (Greenway et al. 2004). The dur-
ation of intervention among trials varied between 4 and
36 weeks. The dosage of caffeine or caffeinated coffee in
intervention group was from 60 to 4000 mg/day (median:
360 mg/day). Seven trials were conducted in USA (Boozer
et al. 2002; Coffey et al. 2004; Colker et al. 1999; Greenway
et al. 2004; Hackman et al. 2006; Liu et al. 2013), and one in
each of the following countries, Iran (Davoodi et al. 2014),
Italy (Bracale et al. 2014), Germany (Bakuradze et al. 2011),
Netherlands (Hursel and Westerterp-Plantenga, 2009),
Hungary (Molnar et al. 2000), and Norway (Thom, 2007).
Figure 2 shows risk of bias of included trials.
The pooled analyses yielded a Beta coefficient of 0.29
(95%CI: 0.19, 0.40; Q ¼124.5, I
¼91.2%) for weight, 0.23
(95%CI: 0.09, 0.36; Q ¼71.0, I
¼93.0%) for BMI, and 0.36
(95%CI: 0.24, 0.48; Q ¼167.36, I
¼94.0%) for body fat
(Figures 3 and 4). The dose-response analyses showed that a
person who consumed 2 mg of caffeine per day compared to
1 mg of caffeine intake per day have 22% more reduction in
weight, 17% more reduction in BMI, and 28% more reduc-
tion in body fat.
The subgroup analysis for potential confounder variables
for heterogeneity including the dosage of caffeine intake, the
duration of the intervention, geographic area, gender, and
energy restriction are summarized in Table 2. The results of
subgroup analyses showed that the Betas were different in
some specific strata of suspected variables.
Egger’s regression tests indicated no significant publica-
tion bias for the effect of caffeine intake on mean reduction
in weight (B ¼3.23, P¼0.22), and BMI (B ¼3.79,
P¼0.29). There was evidence of possible publication bias
in the effect of caffeine intake and fat mass (B ¼7.81,
P¼0.001). The results showed that the summary regression
coefficient (Beta) on body fat significantly decreased between
before (Beta 0.36; 95%CI, 0.24, 0.24) and after (Beta 0.22;
95%CI, 0.10, 0.34) censored trials were included into
This systematic review and meta-analysis is the first report
of the effect of caffeine intake on weight, BMI and body fat
and showed that caffeine intake might promote weight, BMI
and body fat reduction.
Obesity is associated with multiple metabolic and non-
metabolic disorders such as CHD, T2DM, and certain types of
cancer (Kromhout, 1983; Lynch et al. 2009). The current meta-
analysis of RCTs demonstrated that caffeine consumption
resulted in a significant decrease in weight, BMI and body fat.
In a meta-analysis by Phung et al.(Phung et al. 2010), it was
documented that the administration of green tea catechins
(GTCs) with caffeine was correlated with a significant decrease
in BMI, body weight, and waist circumference. Among healthy
people intake of green tea extract containing 270 mg of
Figure 4. A–C. The change of mean weight, BMI, and fat mass status as a
function of dietary of caffeine intake (mg/d) calculated using random effects
meta-analyses of randomized controlled trails on )A) weight
[loge(y) ¼0.29 loge(x) 0.90], )B) BMI [loge(y) ¼0.23 loge(x) 1.93], )C)
body fat loge(y) ¼0.36 loge(x) 1.10].
6 R. TABRIZI ET AL.
epigallocatechin gallate and 150 mg of caffeine, were associated
with a significant increase in energy expenditure (by 4%)
compared with people who consumed caffeine alone. It was
also found that; there was a significant decrease in fat oxida-
tion (by 41%) for people who consumed green tea compared
with people who consumed caffeine (by 33%) (Dulloo et al.
1999). Few studies have reported the beneficial effects of caf-
feine on metabolic profiles. In a meta-analysis by Shi et al. (Shi
et al. 2016), caffeine intake significantly reduced insulin sensi-
tivity in healthy people. In another meta-analysis, the adminis-
tration of GTCs with or without caffeine led to a significant
drop in fasting glucose concentrations (Zheng et al. 2013).
Earlier, it was reported that the family of insulin-like growth
factors and their binding proteins involved in energy restriction
(Hamilton-Fairley et al. 1993), might interfere with reduction
in weight, BMI and body fat (Kiddy et al. 1989). Increased
insulin sensitivity would result in elevated IGF-binding pro-
tein-1 during short-term energy restriction (Moran et al. 2003).
In the current meta-analysis, we selected weight, BMI,
and body fat because they are known as the main diagnostic
variables in overweight and obese people, as well as being
the independent risk factors for CVD and diabetes (NHLBI
Obesity Education Initiative Expert Panel 1998). Despite the
statistical significance found between caffeine intake and
reduction in weight, BMI, and body fat in the current study,
the observed changes might not possibly be clinically rele-
vant. For instance, for anti-obesity agents available in the
market, subjects are considered to have failed their treat-
ment if they have not achieved a weight loss of 2 kg after
4 weeks of the therapy (NHLBI Obesity Education Initiative
Expert Panel 1998). It is noteworthy to mention that, in the
current meta-analysis, caffeine consumption provided an
average weight loss of <2 kg after 4 weeks of intervention
compared with the control group.
Caffeine intake may contribute to a decrease in anthropo-
metric measures through increased energy expenditure (Astrup
et al. 1990; Dulloo et al. 1989), and increased thermogenesis
(Astrup et al. 1990). Current evidence indicates the presence of
caffeine antagonize adenosine receptors (Graham 2001;Thong
and Graham 2002b) both in skeletal muscle (Graham 2001;
Han et al. 1998) and in the central nervous system, with the
latter results in an elevation in sympathetic activity (Thong
and Graham 2002a), which might results in weight loss.
The strengths of the current study include: (1) we made a
quantitative evaluation on how caffeine intake may influence
weight and BMI, based on the best available evidence from
RCTs; and (2) we combined all available dose in our dose-
response meta-analysis across a large range of exposure, and
the validity of the dose-response estimates have been increased.
There are several limitations in this meta-analysis, which
should be taken into consideration when assess the results.
Firstly, the number of studies, which were included in low-
grade and high-grade subgroup analyses, was too small to
gain solid conclusions; therefore more relevant studies are
Table 2. The effects of caffeine intake on weight, BMI and body fat reduction based on subgroup analysis.
of trials Subgroups
(random effect) 95% CI I-squared (%) Overall I-squared (%)
Weight Dosage of caffeine (mg/day) 3 <200 0.18 0.04, 0.32 86.2 91.2
5 200- 450 0.44 0.21, 0.67 95.9
3 450 <0.24 0.15, 0.33 0.0
Duration of study (week) 8 12 0.25 0.12, 0.39 91.7
413 0.37 0.17, 0.58 92.3
Geographic area 7 US 0.33 0.23, 0.44 82.8
5 Non US 0.24 0.04, 0.45 94.2
Gender 3 Female 0.11 0.09, 0.30 89.3
1 Male 0.18 0.02, 0.33 –
8 Both 0.38 0.26, 0.50 88.2
Energy restriction 6 No 0.33 0.24, 0.42 70.0
6 Yes 0.27 0.10, 0.44 93.7
BMI Dosage of caffeine (mg/day) 2 <200 0.10 0.06, 0.14 0.0 93.0
3 200- 450 0.33 0.07, 0.60 95.7
1 450 <0.21 0.06, 0.36 –
Duration of study (week) 4 12 0.20 0.01, 0.40 93.0
213 0.28 0.07, 0.62 95.8
Geographic area 1 US 0.46 0.36, 0.55 –
5 Non US 0.17 0.07, 0.28 85.1
Gender 2 Female 0.09 0.03, 0.15 0.0
1 Male 0.21 0.06, 0.36 –
3 Both 0.34 0.06, 0.61 96.8
Energy restriction 3 No 0.26 0.01, 0.50 95.7
3 Yes 0.20 0.03, 0.42 91.7
Fat mass Dosage of caffeine (mg/day) 4 <200 0.29 0.13, 0.46 94.0 94.0
4 200- 450 0.34 0.11, 0.56 95.6
3 450 <0.53 0.10, 0.96 94.3
Duration of study (week) 6 12 0.34 0.17, 0.52 94.6
513 0.39 0.20, 0.58 94.6
Geographic area 6 US 0.45 0.25, 0.65 94.1
5 Non US 0.27 0.11, 0.44 94.8
Gender 2 Female 0.30 0.17, 0.77 96.4
1 Male 0.12 0.01, 0.26 –
8 Both 0.41 0.27, 0.56 94.3
Energy restriction 7 No 0.38 0.23, 0.53 92.8
4 Yes 0.33 0.11, 0.55 95.7
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 7
needed to further explore this association. Secondly, sub-
stantial heterogeneity was observed across studies, which
was expected considering differences in types of caffeine
(e.g., only caffeine vs. caffeine plus other compositions), and
participants’characteristics (e.g., gender, geographic region,
genetic background, and gene-environment interactions).
Thirdly, our search was limited to English language.
Overall, the current meta-analysis demonstrated that caf-
feine intake promoted weight, BMI and body fat reduction.
Additional prospective studies investigating the effect of caf-
feine supplementation on weight, BMI, and body fat loss
The current study was founded by a grant from the Vice-chancellor for
Research, Shiraz University of Medical Sciences, Shiraz, and Iran.
Kamran B Lankarani http://orcid.org/0000-0002-7524-9017
NHLBI Obesity Education Initiative Expert Panel. 1998. Clinical guide-
lines on the identification, evaluation, and treatment of overweight
and obesity in adults: executive summary. Expert panel on the iden-
tification, evaluation, and treatment of overweight in adults. The
American Journal of Clinical Nutrition 68(4):899–917.
Astrup, A. 2000. Thermogenic drugs as a strategy for treatment of
obesity. Endocrine 13(2):207–12.
Astrup, A., S. Toubro, S. Cannon, P. Hein, L. Breum, and J. Madsen.
1990. Caffeine: A double-blind, placebo-controlled study of its
thermogenic, metabolic, and cardiovascular effects in healthy volun-
teers. The American Journal of Clinical Nutrition 51(5):759–67.
Bakuradze, T., Boehm, N. Janzowski, C. Lang, R. Hofmann, T. Stockis,
J. P. Albert, F. W. Stiebitz, H. Bytof, G. Lantz. I. et al. 2011.
Antioxidant-rich coffee reduces DNA damage, elevates glutathione
status and contributes to weight control: Results from an interven-
tion study. Molecular Nutrition & Food Research 55(5):793–7.
Boozer, C. N., P. A. Daly, P. Homel, J. L. Solomon, D. Blanchard, J. A.
Nasser, R. Strauss, and T. Meredith. 2002. Herbal ephedra/caffeine
for weight loss: A 6-month randomized safety and efficacy trial.
International Journal of Obesity and Related Metabolic Disorders :
Journal of the International Association for the Study of Obesity
Bracale, R., M. L. Petroni, S. Davinelli, U. Bracale, G. Scapagnini,
M. O. Carruba, and E. Nisoli. 2014. Muscle uncoupling protein 3
expression is unchanged by chronic ephedrine/caffeine treatment:
Results of a double blind, randomised clinical trial in morbidly
obese females. PLoS One 9(6):e98244
Coffey, C. S., D. Steiner, B. A. Baker, and D. B. Allison. 2004. A
randomized double-blind placebo-controlled clinical trial of a prod-
uct containing ephedrine, caffeine, and other ingredients from
herbal sources for treatment of overweight and obesity in the
absence of lifestyle treatment. International Journal of Obesity and
Related Metabolic Disorders: Journal of the International Association
for the Study of Obesity 28(11):1411–9.
Colker, C. M., D. S. Kaiman, G. C. Torina, T. Perlis, and C. Street.
1999. Effects of citrus aurantium extract, caffeine, and St. John’s
Wort on body fat loss, lipid levels, and mood states in overweight
healthy adults. Current Therapeutic Research 60(3):145–53.
Davoodi, S. H., S. J. Hajimiresmaiel, M. Ajami, A. Mohseni-Bandpei,
S. A. Ayatollahi, K. Dowlatshahi, G. Javedan, and H. Pazoki-
Toroudi. 2014. Caffeine treatment prevented from weight regain
after calorie shifting diet induced weight loss. Iranian Journal of
Pharmaceutical Research : Ijpr 13(2):707–18.
DerSimonian, R., and N. Laird. 1986. Meta-analysis in clinical trials.
Controlled Clinical Trials 7(3):177–88.
Dulloo, A. G., C. Duret, D. Rohrer, L. Girardier, N. Mensi, M. Fathi,
P. Chantre, and J. Vandermander. 1999. Efficacy of a green tea
extract rich in catechin polyphenols and caffeine in increasing 24-h
energy expenditure and fat oxidation in humans. The American
Journal of Clinical Nutrition 70(6):1040–5.
Dulloo, A. G., C. A. Geissler, T. Horton, A. Collins, and D. S. Miller.
1989. Normal caffeine consumption: Influence on thermogenesis
and daily energy expenditure in lean and postobese human volun-
teers. The American Journal of Clinical Nutrition 49(1):44–50.
Dulloo, A. G., J. Seydoux, and L. Girardier. 1992. Potentiation of the
thermogenic antiobesity effects of ephedrine by dietary methylxan-
thines: Adenosine antagonism or phosphodiesterase inhibition?
Metabolism –Clinical and Experimental 41(11):1233–41.
Flegal, K. M., M. D. Carroll, C. L. Ogden, and C. L. Johnson. 2002.
Prevalence and trends in obesity among US adults, 1999–2000.
Graham, T. E. 2001. Caffeine and exercise: Metabolism, endurance and
performance. Sports Medicine (Auckland, N.Z.) 31(11):785–807.
Greenway, F. L., L. De Jonge, D. Blanchard, M. Frisard, and S. R.
Smith. 2004. Effect of a dietary herbal supplement containing caf-
feine and ephedra on weight, metabolic rate, and body composition.
Obesity Research 12(7):1152–7.
Hackman, R. M., P. J. Havel, H. J. Schwartz, J. C. Rutledge, M. R.
Watnik, E. M. Noceti, S. J. Stohs, J. S. Stern, and C. L. Keen. 2006.
Multinutrient supplement containing ephedra and caffeine causes
weight loss and improves metabolic risk factors in obese women: A
randomized controlled trial. International Journal of Obesity 30(10):
Hamilton-Fairley, D., D. Kiddy, V. Anyaoku, R. Koistinen, M. Seppala,
and S. Franks. 1993. Response of sex hormone binding globulin and
insulin-like growth factor binding protein-1 to an oral glucose toler-
ance test in obese women with polycystic ovary syndrome before
and after calorie restriction. Clinical Endocrinology 39(3):363–7.
Han, D. H., P. A. Hansen, L. A. Nolte, and J. O. Holloszy. 1998.
Removal of adenosine decreases the responsiveness of muscle glu-
cose transport to insulin and contractions. Diabetes 47(11):1671–5.
Hursel, R., and M. S. Westerterp-Plantenga. 2009. Green tea catechin
plus caffeine supplementation to a high-protein diet has no add-
itional effect on body weight maintenance after weight loss. The
American Journal of Clinical Nutrition 89(3):822–30.
Kiddy, D. S., D. Hamilton-Fairley, M. Seppala, R. Koistinen, V. H.
James, M. J. Reed, and S. Franks. 1989. Diet-induced changes in sex
hormone binding globulin and free testosterone in women with nor-
mal or polycystic ovaries: Correlation with serum insulin and insu-
lin-like growth factor-I. Clinical Endocrinology 31(6):757–63.
Kromhout, D. 1983. Body weight, diet, and serum cholesterol in 871
Middle-aged men during 10 years of follow-up (The Zutphen
Study). The American Journal of Clinical Nutrition 38(4):591–8.
Lee, S., R. Hudson, K. Kilpatrick, T. E. Graham, and R. Ross. 2005.
Caffeine ingestion is associated with reductions in glucose uptake
independent of obesity and type 2 diabetes before and after exercise
training. Diabetes Care 28(3):566–72.
Liu, A. G., S. R. Smith, K. Fujioka, and F. L. Greenway. 2013. The
effect of leptin, caffeine/ephedrine, and their combination upon vis-
ceral fat mass and weight loss. Obesity (Silver Spring, Md.) 21(10):
Lynch, E., K. Liu, G. S. Wei, B. Spring, C. Kiefe, and P. Greenland.
2009. The relation between body size perception and change in
body mass index over 13 years: The coronary artery risk
8 R. TABRIZI ET AL.
development in young adults (CARDIA) study. American Journal of
Molnar, D., K. Torok, E. Erhardt, and S. Jeges. 2000. Safety and effi-
cacy of treatment with an ephedrine/caffeine mixture. The first dou-
ble-blind placebo-controlled pilot study in adolescents. International
Journal of Obesity and Related Metabolic Disorders: Journal of the
International Association for the Study of Obesity 24(12):1573–8.
Moran, L. J., M. Noakes, P. M. Clifton, L. Tomlinson, C. Galletly, and
R. J. Norman. 2003. Dietary composition in restoring reproductive
and metabolic physiology in overweight women with polycystic
ovary syndrome. The Journal of Clinical Endocrinology and
Phung, O. J., W. L. Baker, L. J. Matthews, M. Lanosa, A. Thorne, and
C. I. Coleman. 2010. Effect of green tea catechins with or without
caffeine on anthropometric measures: A systematic review and
Meta-analysis. The American Journal of Clinical Nutrition 91(1):
Shi, X., W. Xue, S. Liang, J. Zhao, and X. Zhang. 2016. Acute caffeine
ingestion reduces insulin sensitivity in healthy subjects: A systematic
review and meta-analysis. Nutrition Journal 15(1):103
Souverein, O. W., C. Dullemeijer, P. vanT Veer, and H. van der Voet.
2012. Transformations of summary statistics as input in meta-ana-
lysis for linear dose-response models on a logarithmic scale: A
methodology developed within EURRECA. BMC Medical Research
Thom, E. 2007. The effect of chlorogenic acid enriched coffee on glu-
cose absorption in healthy volunteers and its effect on body mass
when used long-term in overweight and obese people. The Journal
of International Medical Research 35(6):900–8.
Thong, F. S., and T. E. Graham. 2002a. Caffeine-induced impairment
of glucose tolerance is abolished by beta-adrenergic receptor block-
ade in humans. Journal of Applied Physiology 92(6):2347–52.
Thong, F. S., and T. E. Graham. 2002b. The putative roles of adenosine
in insulin- and exercise-mediated regulation of glucose transport
and glycogen metabolism in skeletal muscle. Canadian Journal of
Applied Physiology 27(2):152–78.
Villareal, D. T., S. Chode, N. Parimi, D. R. Sinacore, T. Hilton, R.
Armamento-Villareal, N. Napoli, C. Qualls, and K. Shah. 2011.
Weight loss, exercise, or both and physical function in obese older
adults. The New England Journal of Medicine 364(13):1218–29.
Westerterp-Plantenga, M. S., M. P. Lejeune, and E. M. Kovacs. 2005.
Body weight loss and weight maintenance in relation to habitual caf-
feine intake and green tea supplementation. Obesity Research 13(7):
Wing, R. R., R. W. Jeffery, L. R. Burton, C. Thorson, L. H. Kuller, and
A. R. Folsom. 1992. Change in waist-hip ratio with weight loss and
its association with change in cardiovascular risk factors. The
American Journal of Clinical Nutrition 55(6):1086–92.
Zheng, X., and H. Hasegawa. 2016. Administration of caffeine inhibited
adenosine receptor agonist-induced decreases in motor performance,
thermoregulation, and brain neurotransmitter release in exercising
rats. Pharmacology, Biochemistry, and Behavior 140 :82–9.
Zheng, X. X., Y. L. Xu, S. H. Li, R. Hui, Y. J. Wu, and X. H. Huang.
2013. Effects of green tea catechins with or without caffeine on gly-
cemic control in adults: a Meta-analysis of randomized controlled
trials. The American Journal of Clinical Nutrition 97(4):750–62.
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 9