Content uploaded by Behnam Sadeghirad
Author content
All content in this area was uploaded by Behnam Sadeghirad
Content may be subject to copyright.
Public Health Nutrition: page 1 of 11 doi:10.1017/S1368980012005046
1
Islamic fasting and weight loss: a systematic review and
2
meta-analysis
3
Behnam Sadeghirad
1
, Shahrzad Motaghipisheh
2
, Fariba Kolahdooz
3
,
4
Mohammad J Zahedi
4
and Ali A Haghdoost
2,5,
*
5
1
Kerman Neuroscience Research Center, University of Medical Sciences, Kerman, Islamic Republic of Iran:
6
2
Research Center for Modeling in Health, University of Medical Sciences, Avicenna Ave., Jahad Blvd, Postal
7
Code 7619813159, Kerman, Islamic Republic of Iran:
3
Aboriginal & Global Health Research Group,
8
Department of Medicine, University of Alberta, Edmonton, Canada:
4
Gastroenterology, Physiology Research
9
Center, Internal Medicine Department, University of Medical Sciences, Kerman, Islamic Republic of Iran:
10
5
London School of Hygiene and Tropical Medicine,Q1 London, UK
11
Submitted 17 April 2012: Final revision received 20 June 2012: Accepted 30 July 2012
13
14
Abstract
15
Objective: Studies on the effects of Ramadan fasting on weight changes have been
16
contradictory. We brought together all published data to comprehensively
17
examine the effects in a systematic review and meta-analysis.
18
Design: Relevant studies were obtained through searches of PubMed and CINAHL
19
and by independent screening of reference lists and citations without any time
20
restriction. All searches were completed between October and November 2011.
21
Setting: Changes in body weight during and after Ramadan were extracted from
22
thirty-five English-language studies and were meta-analysed. Most of the studies
23
were conducted in West Asia (n 19); the remainder were conducted in Africa
24
(n 7), East Asia (n 3) and North America/Europe (n 4).
25
Subjects: Healthy adults.
26
Results: Fasting during Ramadan resulted in significant weight loss (21?24 kg;
27
95 % CI 21?60, 20?88 kg). However, most of the lost weight was regained within
28
few weeks and only a slight decrease in body weight was observed in the fol-
29
lowing weeks after Ramadan compared with that at the beginning of Ramadan.
30
Weight loss at the end of Ramadan was significant in both genders (21?51 kg for
31
men and 20?92 kg for women); but againQ2 the weight loss lasted no longer than
32
2 weeks after Ramadan. Weight loss during Ramadan was greater among Asian
33
populations compared with Africans and Europeans.
34
Conclusions: Weight changes during Ramadan were relatively small and mostly
35
reversed after Ramadan, gradually returning to pre-Ramadan status. Ramadan
36
provides an opportunity to lose weight, but structured and consistent lifestyle
37
modifications are necessary to achieve lasting weight loss.
Keywords
Body weight
Ramadan fasting
Energy intake
Weight loss
Systematic review
38
39
Many religions recommend fasting and adherents find it
40
spiritually and physically helpful. Ramadan, the ninth
41
lunar month of the Islamic Hijri calendar, is a period
42
during which healthy adult Muslims are expected to
43
refrain from eating, drinking, smoking, sexual activity and
44
indulging in anything that is excessive or ill natured from
45
dawn to dusk
(1)
. Islamic fasting during Ramadan does not
46
require energy restriction; however, as intake of food and
47
fluid becomes less frequent and is exclusively nocturnal –
48
between sunset and sunrise – changes in body mass and
49
weight might be expected
(1,2)
. Since the Islamic calendar
50
is based on a lunar cycle, every year Ramadan moves 11 d
51
back and makes the fasting period vary from 11 to 18 h
52
depending on geographical location of the country and
53
the fasting season
(1)
.
54
Several nutritional changes occur and the quality of
55
ingested nutrients differs during Ramadan compared
56
with the rest of the year
(3)
; moreover, changes in meal
57
frequency, food composition, energy intake and sleep
58
duration have been reported
(1,4)
. Therefore, different
59
effects of Ramadan fasting on lifestyle and metabolic
60
indices have been reported in both the normal population
61
and specific groups
(1,5,6)
.
62
Within the last 30 years, results describing the effects of
63
Ramadan fasting on body weight have been inconclusive.
64
There are studies that reported a definite loss of body
65
weight or body fat during Ramadan
(7–13)
; while there are
66
as many studies that showed either no significant loss of
67
body weight
(14–17)
or a slight increase in body weight
68
during or after Ramadan
(3,18–20)
.
Public Health Nutrition
*Corresponding author: Email ahaghdoost@kmu.ac.ir; Ali-Akbar.Haghdoost@lshtm.ac.uk r The Authors 2012
69
We perfor med a meta-analysis based on a systematic
70
review to clarify the results of published literature
71
designed to investigate the effects of Ramadan fasting on
72
body weight. In addition, we extracted data on energy
73
intake, diet composition, fasting time and participant
74
age to examine the effects of these variables on weight
75
changes due to Ramadan fasting.
76
Methods
77
Search strategy
78
All literature published in peer-reviewed journals
79
addressing the effects of Ramadan fasting on body weight
80
and composition were potential contributors to the data-
81
base. We searched MEDLINE through PubMed and CINAHL
82
(Cumulative Index to Nursing and Allied Health Literature)
83
through EBSCO (Elton B Stephens Company) host.
84
Due to the religious nature of Ramadan fasting, the data
85
available in the literature rely on weight changes before
86
and after Ramadan, and no randomized controlled trial
87
has been done. Therefore, a search strategy that combined
88
a highly sensitive filter for observational studies along with
89
subject-specific terms was used. The MeSH (Medical Subject
90
Headings) terms and text words (and their combinations
91
and truncated synonyms) were adapted as appropriate to
92
search each database. The combination of keywords in
93
searches of both databases was as follows: (‘Ramadan’ OR
94
‘fast*’ OR ‘Muslim fast*’ OR ‘Islam fast*’ OR ‘Islamic fast*’)
95
AND (‘body weight’ OR ‘weight reduction’ OR ‘weight loss’
96
OR ‘weight gain’ OR ‘body fat distribution’ OR ‘waist
97
circumference’ OR ‘BMI’ OR ‘BMI’ OR ‘body composition’).
98
Attempts to contact authors to access missing data were
99
also extensively performed. All searching procedures were
100
completed between October and November 2011.
101
Additional studies were identified by searching the
102
reference lists of identified papers and by searching
103
the Science Citation Index for eligible papers that were
104
commonly cited to identify subsequent studies that had
105
cited them. The search was confined to full papers pub-
106
lished in the English language with no limit on the date of
107
publication.
108
Selection criteria
109
We read the abstracts of all identified studies to exclude
110
those that were clearly not relevant. The full texts of
111
the remaining articles were read to determine if they met
112
the inclusion criteria. The extracted data were checked
113
independently by two reviewers (S.M. and B.S.) and
114
any discrepancies were resolved by consensus. Primary
115
and quantitative studies were included if a baseline esti-
116
mation of body weight either before Ramadan fasting or
117
within the final week of the month of Ramadan and body
118
weight after Ramadan were provided. Since any lifestyle
119
change needs a minimum period of time to induce
120
physiological change and weight change in particular,
121
we excluded studies that reported body weight changes
122
in a short time period (e.g. ,2 weeks). We also excluded
123
studies that were not representative of the general
124
population. Studies conducted on specific subgroups,
125
such as medically ill patients (e.g. diabetic, hyperlipi-
126
daemic or transplant patients), pregnant or breast-feeding
127
women, athletes, refugees and prisoners, were excluded
128
as well. There was 100 % agreement on inclusion and
129
exclusion of studies.
130
Quality assessment and data extraction
131
To assess the quality of included studies, a simple
132
checklist based on the Newcastle–Ottawa Scale and the
133
Agency for Health Care Research and Quality was gen-
134
erated
(21)
. Studies were evaluated for methodological
135
considerations of observational studies such as the
136
representativeness of the sample, comparison, exposure
137
ascertainment, outcome assessment, adequacy of follow-
138
up, consideration of confounders and type of analysis.
139
Two investigators independently extracted data,
140
reconciling differences by consensus. Information related
141
to body weight, BMI, average daily nutrient consumption
142
(i.e. intakes of carbohydrate, protein and fat) and total
143
daily energy intake of participants before, during and
144
after Ramadan, average fasting time during Ramadan,
145
publication and study date, study population, sample size,
146
age and gender of participants was extracted. There was
147
94 % agreement between data collectors across all items.
148
Statistical analysis
149
To compare the effects of Ramadan fasting on body
150
weight, the difference between baseline body weight
151
(weight in the first day of Ramadan or before that) and
152
follow-up weight (either in the final week of Ramadan
153
(Dweight1) or .2 weeks after Ramadan (Dweight2)) was
154
computed for each study and used as the study end point.
155
Additionally, the difference in body weight between the
156
final week of Ramadan and .2 weeks after Ramadan
157
(Dweight3) was computed.
158
For meta-analysis of the weight change between pre-
159
and post-Ramadan, other than final and baseline mea-
160
surements, the standard deviations of the weight change
161
are also necessary; however, the availability of the latter
162
in publications of interest is a challenging issue
(22,23)
.
163
In the current review, only four studies reported data on
164
variances of Dweight1 and only two studies reported data
165
on variances of Dweight2 and Dweight3. Since a large
166
proportion of the primary articles did not provide infor-
167
mation on variances of the Dweights, the imputation
168
method based on reported P values for Dweights was
169
used to calculate standard deviations of the weight
170
changes. A detailed description of this method and its
171
validity is reported elsewhere
(22,24)
.
172
The analysis and reporting of the current review
173
conformed as much as practicable to the MOOSE
174
(Meta-analyses Of Observational Studies in Epidemiology)
Public Health Nutrition
2 B Sadeghirad et al.
175
guideline
(25)
. We chose fixed- or random-effects model to
176
estimatethecombinedeffectsaccordingtotheresultsofthe
177
heterogeneity test (Cochrane Q). In addition, in order to
178
minimize the random variation among point estimations,
179
we adjusted findings of the studies using Bayesian analysis
180
in forest plots. This technique uses the combined estimation
181
as the prior probability and re-estimates the effect of each
182
study accordingly
(26)
.
183
We estimated t
2
as the indicator of heterogeneity using
184
the restricted likelihood method. A significance level of
185
P , 0?10 and I
2
. 50 % were considered as heterogeneity.
186
In the case of significant heterogeneity, the random-
187
effects model was used and the relevant factors were
188
explored in subgroup analysis where data were available.
189
We also used the meta-regression model to examine the
190
effects of energy intake, diet composition, age and aver-
191
age fasting time on the weight changes during and after
192
Ramadan fasting. Publication bias was examined visually
193
with a funnel plot of study precision v. effect size and by
194
using Begg and Mazumdar’s adjusted rank correlation
195
test. A sensitivity analyses was conducted to assess the
196
contribution of each study to the overall effect. The
197
analyses were performed using the Stata statistical soft-
198
ware program version 11.
199
Results
200
Description of studies
201
Out of 698 retrieved studies, thirty-five studies on the
202
effects of fasting during Ramadan on body weight met
203
the inclusion criteria and were subjected to meta-analysis;
204
the stages of evaluation and exclusion of the identified
205
studies are presented in Fig. 1.
206
The methodological quality of the included studies
207
was moderate, with a median score (stars) of 7 out of a
208
possible 9 (interquartile range: 6–8). No study was omitted
209
due to poor quality.
Public Health Nutrition
Citations identified by search strategy (n 814)
Studies excluded because retrieved twice
from different databases (n 116)
Studies excluded on basis of title and
abstract (n 617)
Studies retrieved for title and abstract
evaluation (n 698)
Articles reviewed in full text (n 81)
Studies on effects of Ramadan fasting
in medically ill patients (n 24)
Potentially appropriate studies on
effects of Ramadan fasting in healthy subjects to
be included in systematic review (n 42*)
Eligible studies included in systematic
review and meta-analysed (n 35)
Studies excluded by thorough article
evaluation for not fulfilling
inclusion criteria or not reporting
adequate data on body weight (n 7)
Studies on effects of Ramadan
fasting in athletes (n 13)
Studies on effects of Ramadan
fasting in obese subjects (n 4)
Fig. 1 Flowchart showing the selection of studies included in the present systematic review of Ramadan fasting effects on body
weight (*two studies reported the effects of Ramadan fasting in healthy and obese subjects, and in healthy and diabetic subjects)
Ramadan fasting and weight: a meta-analysis 3
210
Table 1 provides detailed information on the studies
211
used in the meta-analysis
(3,8–20,27–47)
and descriptive
212
statistics of their characteristics. The average number of
213
subjects per study was 35?8(SD 27?9) and the subjects’
214
average age was 29?4(SD 6?4) years old. The average time
215
for fasting during the month of Ramadan in the included
216
studies was 13?3(SD 1?6) h/d (Table 2). Most of the
217
included studies (54 %) provided findings for men and
218
women combined; however, of studies reporting data
219
only for one gender, twelve findings were recorded for
220
men and four for women. Included studies were under-
221
taken in different geographical locations. The studied
222
populations included West Asian (n 19; more than 60 %
223
of eligible studies), African (n 7), East Asian (n 3) and
224
North American/European (n 4).
225
Weight variation and Ramadan fasting
226
One study did not report the body weight of participants
227
before Ramadan, whereas all of the included studies
228
reported body weight at the end of Ramadan. A total of
229
sixteen studies reported participants’ body weight after
230
Ramadan. The mean initial weight for all 1234 subjects
231
was 68?5(SD 6?5) kg and the average baseline BMI was 23?5
232
(SD 1?9) kg/m
2
(Tab le 2). When findings were pooled toge-
233
ther, the results of the meta-analysis showed a 1?24 kg weight
234
reduction durin g Ra madan (95 % CI 21?60, 20?88 kg;
235
P , 0?001). Figure 2 shows the forest plot of thirty-four stud-
236
ies which pertain to the effects of fasting on body weight
237
during Ramadan. The findings showed substantial hetero-
238
geneity in this category (I
2
5 79?1 %). The weight reduction
239
during Ramadan was followed by a significant increase in
240
weight after Ramadan (0?72 kg, 95 % CI 0?32, 1?13 kg;
241
P , 0?001; Fig. 3). These findings also had substantial
242
heterogeneity (I
2
5 81?7 %). When the reported body weight
243
after Ramadan was compared with baseline values, the
244
results of the meta-analysis showed a weight loss of 0?27 kg
245
(95 % CI 20?51, 20?04 kg; P 5 0?023; I
2
5 4?1%). Figure 4
246
shows the forest plot with mean differences and 95 % con-
247
fidence intervals, and the pooled estimate for the weight
248
change after Ramadan compared with baseline values using
249
a fixed-effects model. We did not find any sign of publication
250
bias for studies in any of these categories (Table 3).
251
When findings for weight variation were meta-analysed
252
separately for men and women, weight loss in both
253
genders at the end of Ramadan was statistically significant
254
(21?51 kg for men and 20?92 kg for women); however,
255
this weight loss did not last for .2 weeks after Ramadan.
256
In both men and women, body weight after Ramadan
257
compared with the baseline value showed a small non-
258
significant decrease (20?10 kg, P 5 0?83 and 20?55 kg,
259
P 5 0?37, respectively). During post-Ramadan weeks,
260
compared with the end of Ramadan, men gained 1?02 kg
261
in weight (95 % CI 0?42, 1?63 kg; P , 0?001), whereas
262
women’s weight remained unchanged (Table 3).
263
In East Asian populations, there was a significant
264
weight loss (21?56 kg, 95 % CI 22?10, 21?04; P , 0?001)
265
between baseline and the end of Ramadan. Weight
266
reduction during Ramadan was also statistically significant
267
in West Asian and African populations (21?24 kg, P , 0?001
268
and 21?13 kg, P 5 0?001, respectively). Although weight
269
loss in Europeans was less than that in other populations,
270
it was statistically significant (20?64 kg, P , 0?001). Meta-
271
analysis of studies undertaken in Africa and West Asia
272
showed that the lost weight during Ramadan was regained
273
after Ramadan; however, studies of East Asian populations
274
showed a reduction in participants’ body weight after
275
Ramadancomparedwithbaselinevalues(Table3).
276
Diet composition and energy consumption during
277
fasting periods were recorded in fourteen studies and
278
thirteen studies recorded baseline values before Ramadan.
279
In three studies, the total daily energy intake did not change.
280
During Ramadan, the number of studies with increased total
281
daily energy intake was equal to the number of studies
282
with decreased total daily energy intake during Ramadan.
283
Before Ramadan, the total average daily energy intake was Q4
284
8027 (SD 2110) kJ, which increased Q5to 8219 (SD 2527) kJ
285
during Ramadan month. The total energy intake continued
286
to increase even after Ramadan (9923 (SD 1818) kJ/d).
287
Overall, carbohydrate was the food group with the largest
288
consumption (.55 % of total energy intake) before, during
289
and after Ramadan.
290
Meal frequency during Ramadan, compared with the
291
periods before and after Ramadan, was reduced to
292
essentially two per day in almost all studies. The increase
293
in total daily energy intake during Ramadan was not due
294
to an increase in carbohydrate consumption (56?9(SD 10?1)
295
v.57?4(SD 5?5) % of total Q6energy intake). The percentage
296
of energy intake from protein and fat during Ramadan did
297
not change considerably compared with before or after the
298
month of Ramadan (Table 2).
299
The total energy intake was not similar in all geo-
300
graphical locations. The baseline daily energy intake was
301
considerably lower in West Asian populations. During the
302
month of Ramadan, West and East Asian populations
303
reduced their daily energy intake by 600–900 kJ/d,
304
whereas the daily energy intake of African populations
305
increased by about 1100 kJ/d (Table 4). The increase in
306
energy intake in African populations during Ramadan was
307
mainly due to an 84 g/d increase in carbohydrate con-
308
sumption (from 775?3(SD 626?0) to 859?3(SD 713?8) g/d)
309
and a 45 g/d increase in fat consumption (286?1(SD 268?4)
310
to 331?3(SD 296?5) g/d), while protein consumption
311
increased only by 27 g/d (174?3(SD 124?0) to 221?0
312
(SD 141?0) g/d). In both West and East Asian popula-
313
tions a decline in carbohydrate consumption during
314
Ramadan was the main reason for the reduction in daily
315
energy intake.
316
In meta-regression model, the effects of baseline BMI,
317
age, fasting time and energy intake on Dweights were
318
investigated; however, none of the these variables had a
319
considerable effect on the weight reduction during and
320
after Ramadan.
Public Health Nutrition
4 B Sadeghirad et al.
Public Health Nutrition
Table 1 Description of studies included in the present systematic review of Ramadan fasting effects on body weight, and body weight, energy intake and diet composition before, during and
after Ramadan fasting
Age Fasting
At the beginning or
before Ramadan
End of
Ramadan
After
Ramadan
Author & date Country SS* (years) Gender duration- (h) Wt-
-
Energyy CARBJ Fatz Wt-
-
Energyy CARBJ Fatz Wt-
-
Energyy CARBJ Fatz
Fedail et al. (1982)
(8)
UK & Sudan 18 30?0 Both 69?567?7
Shoukry (1986)
(27)
30 Both 66?665?3
Husain et al. (1987)
(28)
Malaysia 21 20–45 Both 14?557?3 9213 55?7 7679 56?9
Hallak and Nomani (1988)
(9)
Syria 16 18–30 M 16?066?263?7
Takruri (1989)
(29)
137 Both 67?665?5
Nomani et al. (1990)
(10)
Syria 16 18–30 M 15?066?263?8
Sweileh et al. (1992)
(11)
7 22–35 Both 72?1 4409 651?1193?070?2 5087 755?4 222?7
Maislos et al. (1993)
(30)
Israel 24 27?0 Both 13?568?068?2
El Ati et al. (1995)
(14)
Tunisia 16 25–39 F 13?059?3 10 216 297?6100?558?9 10 203 243?1 119?658?6 11 011 318?2 109?2
Adlouni et al. (1998)
(31)
Morocco 32 25–50 M 12?069?6 11 237 347?2100?467?8 13 443 412?3 118?169?1 11 942 341?1 104?6
Bigard et al. (1998)
(32)
France 17 38?0 Both 18?073?171?1
Finch et al. (1998)
(15)
UK 41 35?3 Both 11?071?070?770?8
Maislos et al. (1998)
(18)
Israel 22 24?0 Both 12?568?067?068?0
Ramadan et al. (1999)
(33)
Kuwait 13 36?5M 14?573?672?7
Roky et al. (2001)
(34)
Morocco 8 20–28 M 12?067?065?067?0
Ramadan (2002)
(35)
Kuwait 16 35?0M 12?080?279?1
Fakhrzadeh et al. (2003)
(36)
Iran 91 20?8 Both 12?063?0 6337 62?2 4959
Kassab et al. (2003)
(37)
Bahrain 26 18–45 F 12?572?5 6975 71?7 6954 72?3
Kassab et al. (2004)
(38)
Bahrain 46 22?0F 12?580?8 6950 80?1 6996 80?5
Rahman et al. (2004)
(12)
Bangladesh 20 38?3M 12?064?162?1 8850 408?924?763?0 8935 433?317?9
Yucel et al. (2004)
(16)
Turkey 34 32?5 Both 13?068?768?6
Aksungar et al. (2005)
(39)
Turkey 24 30?0 Both 15?072?772?672?3
Al-Numair (2006)
(40)
Saudi Arabia 45 30–45 M 12?085?5 9002 350?055?083?2 7997 310?051?0
Dewanti et al. (2006)
(41)
Indonesia 37 39?0M 14?064?563?0
Lamine et al. (2006)
(19)
Tunisia 30 23?7 Both 12?061?3 6959 244?050?961?7 8357 251?082?662?1 7323 259?153?5
Ziaee et al. (2006)
(42)
Iran 81 22?7 Both 11?062?461?2
Al-Hourani and Atoum (2007)
(43)
Jordan 47 21?6F 12?057?5 5242 174?245?756?9 4902 164?313?9
Furuncuoglu et al. (2007)
(44)
Turkey 39 28?0 Both 14?064?262?8
Mansi (2007)
(13)
Jordan 70 21?3 Both 12?576?672?773?6
Haouari et al. (2008)
(17)
Tunisia 36 24?0M 12?070?9 9684 70?5 9600
Ibrahim et al. (2008)
(45)
United Arab Emirates 14 25–58 Both 16?070?5 7707 269?557?769?1 8056 268?065?5
Haghdoost and Poorranjbar (2009)
(20)
Iran 41 20?4M 14?071?671?172?3
Lamri-Senhadji et al. (2009)
(3)
Algeria 46 24?0 Both 12?061?1 10 493 60?612051 61?7 10 420
Hajek et al. (2011)
(46)
UK 87 34?3 Both 15?571?270?671?1
Unalacak et al. (2011)
(47)
Turkey 10 27?4M 13?571?971?1
M, males; F, females,
*SS, study sample size.
-Fasting duration, average fasting hours per day during Ramadan.
-
-
Wt, body weight (kg).
yEnergy, energy intake (kJ/d).
JCARB, carbohydrates intake (g/d).
zFat, fat intake (g/d).
Ramadan fasting and weight: a meta-analysis 5
Public Health Nutrition
Table 2 Description of variables based on studies included in the present systematic review of Ramadan fasting effects on body weight
Men Women Total
Variable Mean*
SD Range n- Mean* SD Range n- Mean* SD Range n-
Sample size 24?313?48–501927?614?5 9–47 10 35?827?9 7–137 35
Fasting duration (h/d) 13?41?8 11–18 19 12?81?2 11–15 10 13?31?6 11–18 31
Age (years) 29?37?320?4–39?019 25?04?820?8–33?510 28?16?520?4–42?033
Baseline BMI (kg/m
2
)-
-
24?12?221?8–28?612 23?11?421?0–25?3823?51?921?2–28?622
Body weight (kg)
Before Ramadan 70?86?662?9–85?518 63?09?549?9–80?810 68?56?557?3–85?534
End of Ramadan 69?76?661?3–83?218 62?29?648?3–80?110 67?26?255?7–83?235
After Ramadan 69?75?162?6–82?47 67?711?549?2–80?5669?55?856?9–80?516
Energy intake (kJ/d)
Before Ramadan 9062 1949 6021–11 237 6 7137 1594 5242–10 216 7 7931 1923 4409–11 237 13
End of Ramadan 8940 2822 5078–13 443 7 6870 2297 4814–11 716 7 7914 2640 4903–13 443 14
After Ramadan 10 913 1250 8935–11 942 3 10 233 644 9713–11 011 2 9977 1636 7327–11 942 5
% of daily energy intake from carbohydrate
Before Ramadan 58?96?050?5–63?93 56?54?848?8–63?0358?05? 048?8–63?98
End of Ramadan 59?88?751?4–77?14 54?37?039?9–60?0356?78?039?9–77?19
After Ramadan 62?012?550?8–81?13 56?06?348?4–61?0259?79?848?4–81?15
% of daily energy intake from fat
Before Ramadan 26?74?923?0–33?73 31?73?926?0–37?1328?54? 918?
3–37?18
End of Ramadan 23?57?510?6–33?04 33?06?524?0–43?2328?78?410?6–43?29
After Ramadan 24?210?97?6–34?13 30?65?726?0–37?4226?78?87?6–37?45
% of daily energy intake from protein
Before Ramadan 13?71?013?0–15?13 12?80?712?2–14?1313?51? 412?2–20?08
End of Ramadan 13?01?811?2–15?54 13?71?512?9–16?9313?51?911?
2–20?39
After Ramadan 13?71?511?3–15?03 13?50?613?0–14?2213?51? 111?3–15?05
*Frequency-weighted mean for all variables except sample size and fasting duration.
-Number of studies included in each analysis.
-
-
BMI before or at the beginning of Ramadan.
6 B Sadeghirad et al.
321
Discussion
322
Religious fasting, including Islamic fasting, provides a
323
unique and interesting vantage point for evaluating the
324
effects of food restriction/modification
(6)
. Ramadan fast-
325
ing is a natural experiment that occurs annually for
326
1 month and requires specific practices, such as nocturnal
327
consumption of food and changes in meal frequency,
328
food quality, energy intake and sleep cycle. Different
329
studies have reported both positive and negative effects
330
of Ramadan fasting on overall health. The present review
331
summarizes the best available evidence on the effects of
332
Ramadan fasting on weight loss in a healthy population.
333
Weight loss may reduce cardiovascular morbidity and
334
mortality and may lead to prevention of type 2 diabetes
335
and improved control of hypertension
(48–50)
. At any given
Public Health Nutrition
Fedail (1982)
Shoukry (1986)
Husain (1987)
Hallak (1988)
Takruri (1989)
Nomani (1990)
Sweileh (1992)
El Ati (1995)
Finch (1998)
Adlouni (1998)
Bigrad (1998)
Maislos (1998)
Ramadan (1999)
Roky (2001)
Ramadan (2002)
Fakhrzadeh (2003)
Kassab (2003)
Rahman (2004)
Kassab (2004)
Yucel (2004)
Aksungar (2005)
Dewanti (2006)
Lamine (2006)
Al-Numair (2006)
Ziaee (2006)
Mansi (2007)
Furuncuoglu (2007)
AI-Hourani (2007)
Ibrahim (2008)
Haouari (2008)
Haghdoost (2009)
Lamri-Senhadji (2009)
Hajek (2011)
Unalacak (2011)
Combined
Change in body weight, Δweight1 (kg)
−3·0 −2·5 −2·0 −1·5 −1·0 −0·5
0·0 0·5 1·0
Fig. 2 Forest plot showing the change in body weight
Q3
during Ramadan (Dweight1; difference between baseline body weight
(weight in the first day of Ramadan or before that) and follow-up weight (weight in the final week of Ramadan)); values are means
with 95 % confidence intervals shown by horizontal bars. Studies are ordered alphabetically by name of first author and publication
date. The pooled or ‘combined’ mean difference was calculated by a random-effects model
Husain (1987)
Maislos (1993)
El Ati (1995)
Finch (1998)
Adlouni (1998)
Maislos (1998)
Roky (2001)
Kassab (2003)
Rahman (2004)
Kassab (2004)
Aksungar (2005)
Lamine (2006)
Mansi (2007)
Haghdoost (2009)
Lamri-Senhadji (2009)
Hajek (2011)
Combined
Change in body weight, ∆weight3 (kg)
−0·5
0·0 0·5
1·0
1·5 2·0
2·5
Fig. 3 Forest plot showing the change in body weight from the final week of Ramadan to .2 weeks after Ramadan (Dweight3);
values are means with 95 % confidence intervals shown by horizontal bars. Studies are ordered alphabetically by name of first
author and publication date. The pooled or ‘combined’ mean difference was calculated by a random-effects model
Ramadan fasting and weight: a meta-analysis 7
336
time, about 40–60 % of the adult population in Western
337
countries is attempting to lose or maintain weight
(51)
.
338
Our results demonstrated that studies have hetero-
339
geneous findings regarding the effects of fasting on body
340
weight during and after Ramadan. Overall, body weight
341
decreased during Ramadan, which suggests that Muslims
342
do not gain weight as a result of skipping daytime meals
343
during Ramadan but rather lose more than a kilogram of
344
weight; however, the lost weight was regained 6 weeks
345
from the beginning of Ramadan. Muslim men lost more
Public Health Nutrition
Husain (1987)
El Ati (1995)
Finch (1998)
Adlouni (1998)
Maislos (1998)
Roky (2001)
Kassab (2003)
Rahman (2004)
Kassab (2004)
Aksungar (2005)
Lamine (2006)
Mansi (2007)
Haghdoost (2009)
Lamri-Senhadji (2009)
Hajek (2011)
Combined
−7 −5 −3 −1
Change in body weight, ∆weight2 (kg)
1357
Fig. 4 Forest plot showing the change in body weight due to Ramadan (Dweight2; difference between baseline body weight
(weight in the first day of Ramadan or before that) and follow-up weight .2 weeks after Ramadan); values are means with 95 %
confidence intervals shown by horizontal bars. Studies are ordered alphabetically by name of first author and publication date. The
pooled or ‘combined’ mean difference was calculated by a fixed-effects model
Table 3 Results of the meta-analysis of the studies investigating the effect of fasting on weight changes before and after Ramadan
95 % CI
P value P value for
n* Difference (kg) Lower Upper for difference I
2
(%) publication bias-
Dweight1
Men 19 21?51 22?04 20?98 ,0?001 75?00?22
Women 10 20?92 21?37 20?48 ,0?001 10?10?65
African 7 21?13 21?81 20?46 0?001 52?50?10
West Asian 18 21?24 21?85 20?62 ,0?001 85?40?07
East Asian 3 21?56 22?10 21?04 ,0
?001 ,10?12
Europe 3 20?64 20?92 20?35 ,0?001 ,10?60
Total 34 21?24 21?60 20?88 ,0?001 79?10?33
Dweight2
Men 5 20?10 21?02 0?82 0?83 53?50?62
Women 4 20?55 21?76 0?65 0?37 ,10?90
African 5 20?05 20?92 0?82 0?90 ,10?32
West Asian 6 20?29 20?29 0?87 0?33 ,10?85
East Asian 2 20?88 21?53 20?23 0?008 – –
Europe 2 20?32 20?61 20?02 0?035 – –
Total 15 20?27 20?51 20?04 0?023 4?10?73
Dweight3
Men 5 1?02 0?42 1?63 0?001 56?90?62
Women 4 20?04 20?52 0?44 0?87 ,10?50
African 5 0?89
20?03 1?82 0?058 90?20?92
West Asian 7 0?55 20?04 1?14 0?069 70?70?88
East Asian 2 1?08 0?69 1?48 ,0?001 – –
Europe 2 0?46 0?01 0?92 0?049 – –
Total 16 0?72 0?32 1?13 ,0?001 81?70?47
Dweight1, difference between body weight at the end of Ramadan and baseline body weight before Ramadan; Dweight2, difference between body weight after
Ramadan (.2 weeks) and baseline body weight; Dweight3, difference between body weight after Ramadan and body weight at the end of Ramadan (when the
number of findings was ,3, only pooled analysis was performed).
*Number of studies included in each analysis.
-Based on Begg’s test to estimate the chance of publication bias.
8 B Sadeghirad et al.
346
weight than women during Ramadan. A possible expla-
347
nation may be women’s prohibition from fasting during
348
their menstrual periods, which results in fewer days of
349
fasting during Ramadan. Women managed to maintain
350
their body weight after Ramadan but men gained about
351
1 kg during post-Ramadan weeks. It is suggested that fat
352
oxidation during Ramadan fasting may result in an adaptive
353
mechanism for weight maintenance in women
(13)
. In both
354
genders, post-Ramadan weight showed insignificant
355
change from baseline body weight.
356
During Ramadan, changes in the timing of food and
357
fluid intake along with the reduction of meal frequency
358
may result in various physiological changes
(1)
.Itis
359
believed that Ramadan fasting often leads to reduced
360
energy intake and weight loss occurs as a result
(9)
but our
361
findings do not support this hypothesis. Despite the
362
decreased meal frequency during Ramadan in almost all
363
studies, only one-third of included studies showed
364
decreased energy intake during Ramadan. The studies on
365
West and East Asian populations mostly reported that
366
total daily energy intake reduced during Ramadan, but in
367
African populations total daily energy intake was mostly
368
increased. However, the weight loss at the end of
369
Ramadan was not considerably different among African,
370
West Asian and East Asian populations. It is reported that
371
an increase in body weight after Ramadan might be due
372
to less physical activity and/or more energy intake
(52)
.
373
Since the number of studies with information on energy
374
intake and physical activity after Ramadan was inade-
375
quate, we were not able to examine this hypothesis.
376
It is obvious that dietary habits and food choices vary
377
among different cultures; hence, percentages of energy
378
intake from carbohydrate, protein and fat are not similar
379
in different Islamic countries. In relation to macronutrient
380
composition, it is reported that meals are often composed
381
of more fat and less carbohydrate during Ramadan com-
382
pared with the rest of the year
(14,30)
. On the other hand,
383
carbohydrate comprised the largest proportion of foods
384
consumed during Ramadan. The increased energy intake
385
in African populations and the reduced energy intake in
386
West and East Asian populations were both the result of
387
increased or decreased intake of all macronutrients.
388
However, the number of studies in different geographical
389
locations which reported macronutrient consumption
390
during Ramadan as well as information on weight varia-
391
tion was inadequate for a clear conclusion about their
392
effects on weight loss.
393
In different studies various explanations were given for
394
the discrepancies of the reported results; the length and
395
the temperature of the fasting days
(1,53)
, meal schedules
396
and frequencies
(1)
, eating behaviours and the quality and
397
quantity of food intakes
(6,14)
, fluid intakes
(5,54)
, sleeping
398
patterns and duration
(4)
, physical activity
(14,20)
and age
(6)
399
are all reported to have effect on weight loss during
400
Ramadan. The results of our meta-regression model showed
401
that duration of fasting, age, baseline BMI, total energy
402
intake and geographical location have no effect on weight
403
loss in healthy populations during Ramadan. Since the
404
number of studies on the effect s of macronutrient con-
405
sumption, sleep cycle, fluid intake or physical activity on
406
weight variation during Ramadan was not enough for meta-
407
regression, we did not include these studies in our review.
408
Changes in sleeping hours might change the serum
409
level of leptin, insulin and cortisol; these factors could
410
affect daily energy consumption
(38)
and indirectly may
411
explain some of the body weight variation during and
412
after Ramadan.
413
Some authors suggest that the body weight changes
414
could be attributed to variations of fluid intake
(11,54)
.
415
Dehydration certainly occurs during the fasting hours.
416
Acute changes in total body water are best characterized
417
by repeated measurements of body mass; however, with
418
small losses in body mass over a long period of time,
419
any other changes in body composition may bias the
420
measurement of dehydration. Thus, the separation or the
421
quantification of the effects of fasting on body weight
422
from dehydration as a result of fasting seems to be difficult
423
or impossible
(5)
.
Public Health Nutrition
Table 4 Frequency-weighted mean diet composition in different geographical locations
Africa West Asia East Asia
Variable Time Mean
SD n* Mean SD n* Mean SD n*
Energy intake (kJ/d) Before Ramadan 9769 1448 160 6829 1167 269 9213 2757 21
End of Ramadan 10 899 1827 160 6160 1307 269 8251 592 41
After Ramadan 10 145 1705 124 – – 8935 560 20
Carbohydrate intake (g/d) Before Ramadan 775?3 626?0 124 261?482?3 106 – –
End of Ramadan 859?3 713?8 124 239?969?1 106 408?949?920
After Ramadan 753?5 587?3 124 – – 433?342?620
Fat intake (g/d) Before Ramadan 286?1 268?4 124 51?25?0 106 – –
End of Ramadan 331?3 296?5 124 36?520? 8 106 64?620?520
After Ramadan 288?6 265?8 124 – – 60?121?520
Protein intake (g/d) Before Ramadan 174?3 124?0 124 55?014?5 106 – –
End of Ramadan 201?0 141?0 124 50?811? 9 106 24?73?020
After Ramadan 179?6 128?6 124 – – 17?93?620
*n, total number of participants in each category used as frequency weight.
Ramadan fasting and weight: a meta-analysis 9
424
Metabolic responses, for instance to a glucose load, are
425
slower in the evening than in the morning; moreover , gastric
426
emptying and blood flow are greater during the daytime
427
than at night, which leads to a faster absorption of foodstuffs
428
from the gastrointestinal tract
(55)
. Thus, another likely
429
explanationfortheweightlossmayberelatedtothelower
430
absorption of foods eaten nocturnally during Ramadan.
431
Smoking is prohibited during fasting hours, therefore
432
limiting the number of cigarettes per day might be related
433
to weight gain among smokers in Ramadan
(56)
. All of the
434
above explanations might be the good reasons for the
435
observed heterogeneity among studies.
436
Conclusion
437
The present systematic review shows that the lost weight
438
during Ramadan is relatively small and that body weight
439
variations during Ramadan fasting are mostly reversed
440
after Ramadan, gradually returning to pre-Ramadan sta-
441
tus. Ramadan provides an opportunity to lose weight, but
442
structured and consistent lifestyle modifications are
443
necessary to achieve life-lasting weight loss. Although our
444
findings regarding body weight variations as a result of
445
Ramadan fasting were heterogeneous, we believe that
446
restrictions in meal frequencies or energy intake alone
447
cannot explain these variations, and the combination of
448
different factors such dehydration, changes in diet, phys-
449
ical activity and even sleeping hours may be influential.
450
Hence, further studies are needed to investigate the role
451
of dehydration and changes in diet, sleeping hours and
452
physical activity on body weight during Ramadan.
453
Acknowledgements
454
Sources of funding: This work was supported by a grant
455
from the Kerman University of Medical Sciences, Iran
456
through grant number 90/348 (awarded to B.S.). Conflicts
457
of interest: The authors declare that there are no conflicts
458
of interest. Author’s contribution: A.A.H. and B.S. direc-
459
ted the study and were involved in the study design. S.M.,
460
M.J.Z. and B.S. were involved in designing the search
461
strategy, data collection and entry. B.S. and A.A.H. carried
462
out the analysis. F.K. and B.S. had primary responsibility
463
for drafting the manuscript and all authors contributed to
464
writing and revising the manuscript. Acknowledgements:
465
The authors are grateful to Dr N. Pourdamghan for her
466
excellent assistance in searching and gathering the lit-
467
erature for this review and Dr M. Karamuzian for her
468
comments on the manuscript.
469
References
470
1. Roky R, Houti I, Moussamih S et al. (2004) Physiological
471
and chronobiological changes during Ramadan intermittent
472
fasting. Ann Nutr Metab 48, 296–303.
473
2. Azizi F (2010) Islamic fasting and health. Ann Nutr Metab
474
56, 273–282.
475
3. Lamri-Senhadji MY, El Kebir B, Belleville J et al. (2009)
476
Assessment of dietary consumption and time-course of
477
changes in serum lipids and lipoproteins before, during
478
and after Ramadan in young Algerian adults. Singapore
479
Med J 50, 288–294.
480
4. Reilly T & Waterhouse J (2007) Altered sleep–wake cycles
481
and food intake: the Ramadan model. Physiol Behav 90,
482
219–228.
483
5. Leiper JB & Molla AM (2003) Effects on health of fluid
484
restriction during fasting in Ramadan. Eur J Clin Nutr 57,
485
Suppl. 2, S30–S38.
486
6. Trepanowski J, Canale R, Marshall K et al. (2011) Impact of
487
caloric and dietary restriction regimens on markers of
488
health and longevity in humans and animals: a summary of
489
available findings. Nutr J 10, 107.
490
7. Muazzam MG & Khaleque KA (1959) Effects of fasting in
491
Ramadhan. Am J Trop Med Hyg 62, 292–294.
492
8. Fedail SS, Murphy D, Salih SY et al. (1982) Changes in
493
certain blood constituents during Ramadan. Am J Clin Nutr
494
36, 350–353.
495
9. Hallak MH & Nomani MZ (1988) Body weight loss and
496
changes in blood lipid levels in normal men on hypocaloric
497
diets during Ramadan fasting. Am J Clin Nutr 48,
498
1197–1210.
499
10. Nomani MZ, Hallak MH & Siddiqui IP (1990) Effects of
500
Ramadan fasting on plasma uric acid and body weight in
501
healthy men. J Am Diet Assoc 90, 1435–1436.
502
11. Sweileh N, Schnitzler A, Hunter GR et al. (1992) Body
503
composition and energy metabolism in resting and
504
exercising Muslims during Ramadan fast. J Sports Med Phys
505
Fitness 32, 156–163.
506
12. Rahman M, Rashid M, Basher S et al. (2004) Improved
507
serum HDL cholesterol profile among Bangladeshi male
508
students during Ramadan fasting. East Mediterr Health J 10,
509
131–137.
510
13. Mansi KMS (2007) Study the effects of Ramadan fasting on
511
the serum glucose and lipid profile among healthy
512
Jordanian students. Am J Appl Sci 4, 565–569.
513
14. El Ati J, Beji C & Danguir J (1995) Increased fat oxidation
514
during Ramadan fasting in healthy women: an adaptative
515
mechanism for body-weight maintenance. Am J Clin Nutr
516
62, 302–307.
517
15. Finch GM, Day JE, Razak et al. (1998) Appetite changes
518
under free-living conditions during Ramadan fasting.
519
Appetite 31, 159–170.
520
16. Yucel A, Degirmenci B, Acar M et al. (2004) The effect of
521
fasting month of Ramadan on the abdominal fat distribu-
522
tion: assessment by computed tomography. Tohoku J Exp
523
Med 204, 179–187.
524
17. Haouari M, Haouari-Oukerro F, Sfaxi A et al. (2008)
525
How Ramadan fasting affects caloric consumption, body
526
weight, and circadian evolution of cortisol serum levels in
527
young, healthy male volunteers. Horm Metab Res 40,
528
575–577.
529
18. Maislos M, Abou-Rabiah Y, Zuili I et al. (1998) Gorging and
530
plasma HDL-cholesterol: the Ramadan model. Eur J Clin
531
Nutr 52, 127–130.
532
19. Lamine F, Bouguerra R, Jabrane J et al. (2006) Food intake
533
and high density lipoprotein cholesterol levels changes
534
during Ramadan fasting in healthy young subjects.
535
Tunis Med 84, 647–650.
536
20. Haghdoost AA & Poorranjbar M (2009) The interaction
537
between physical activity and fasting on the serum lipid
538
profile during Ramadan. Singapore Med J 50, 897–901.
539
21. Wells G, Shea B, O’Connell D et al. (2011) The New-
540
castle
Q7
–Ottawa Scale (NOS) for assessing the quality of
541
nonrandomised studies in meta-analysis. http://wwworhica/
542
programs/clinical_epidemiology/oxfordasp (accessed ?).
Public Health Nutrition
10 B Sadeghirad et al.
543
22. Thiessen Philbrook H, Barrowman N & Garg AX (2007)
544
Imputing variance estimates do not alter the conclusions of
545
a meta-analysis with continuous outcomes: a case study of
546
changes in renal function after living kidney donation.
547
J Clin Epidemiol 60, 228–240.
548
23. The Cochrane Collaboration (2011) Cochrane Handbook
549
for Systematic Reviews of Interventions, Version 5.1.0 [JPT
550
Higgins and S Green, editors]. http://www.cochrane-
551
handbook.org (accessed March 2011).
552
24. Follmann D, Elliott P, Suh I et al. (1992) Variance
553
imputation for overviews of clinical trials with continuous
554
response. J Clin Epidemiol 45, 769–773.
555
25. Stroup DF, Berlin JA, Morton SC et al. (2000) Meta-analysis
556
of observational studies in epidemiology: a proposal for
557
reporting. Meta-analysis Of Observational Studies in
558
Epidemiology (MOOSE) group. JAMA 283, 2008–2012.
559
26. Sutton AJ & Abrams KR (2001) Bayesian methods in meta-
560
analysis and evidence synthesis. Stat Methods Med Res 10,
561
277–303.
562
27. Shoukry MI (1986) Effect of fasting in Ramadan on plasma
563
lipoproteins and apoproteins. Saudi Med J 7, 561–565.
564
28. Husain R, Duncan MT, Cheah SH et al. (1987) Effects of
565
fasting in Ramadan on tropical Asiatic Moslems. Br J Nutr
566
58, 41–48.
567
29. Takruri HR (1989) Effect of fasting in Ramadan on body
568
weight. Saudi Med J 10, 491–494.
569
30. Maislos M, Khamaysi N, Assali A et al. (1993) Marked
570
increase in plasma high-density-lipoprotein cholesterol
571
after prolonged fasting during Ramadan. Am J Clin Nutr
572
57, 640–642.
573
31. Adlouni A, Ghalim N, Saı
¨
le R et al. (1998) Beneficial effect
574
on serum apo AI, apo B and Lp AI levels of Ramadan
575
fasting. Clin Chim Acta 271, 179–189.
576
32. Bigard AX, Boussif M, Chalabi H et al. (1998) Alterations in
577
muscular performance and orthostatic tolerance during
578
Ramadan. Aviat Space Environ Med 69, 341–346.
579
33. Ramadan J, Telahoun G, Al-Zaid NS et al. (1999) Responses
580
to exercise, fluid, and energy balances during Ramadan in
581
sedentary and active males. Nutrition 15, 735–739.
582
34. Roky R, Chapotot F, Hakkou F et al. (2001) Sleep during
583
Ramadan intermittent fasting. J Sleep Res 10, 319–327.
584
35. Ramadan J (2002) Does fasting during Ramadan alter body
585
composition, blood constituents and physical perfor-
586
mance? Med Princ Pract 11, 41–46.
587
36. Fakhrzadeh H, Larijani B, Sanjari M et al. (2003) Effect of
588
Ramadan fasting on clinical and biochemical parameters in
589
healthy adults. Ann Saudi Med 23, 223–226.
590
37. Kassab SE, Abdul-Ghaffar T, Nagalla DS et al. (2003) Serum
591
leptin and insulin levels during chronic diurnal fasting. Asia
592
Pac J Clin Nutr 12, 483–487.
593
38. Kassab S, Abdul-Ghaffar T, Nagalla DS et al. (2004)
594
Interactions between leptin, neuropeptide-Y and insulin
595
with chronic diurnal fasting during Ramadan. Ann Saudi
596
Med 24, 345–349.
597
39. AksungarFB,ErenA,UreSet al. (2005) Effects of intermittent
598
fasting on serum lipid levels, coagulation status and plasma
599
homocysteine levels. Ann Nutr Metab 49, 77–82.
600
40. Al-Numair K (2006) Body weight and some biochemical
601
changes associated with Ramadan fasting in healthy Saudi
602
men. J Med Sci 6, 112–116.
603
41. Dewanti L, Watanabe C, Sulistiawati et al. (2006) Un-
604
expected changes in blood pressure and hematological
605
parameters among fasting and nonfasting workers during
606
Ramadan in Indonesia. Eur J Clin Nutr 60, 877–881.
607
42. Ziaee V, Razaei M, Ahmadinejad Z et al. (2006) The
608
changes of metabolic profile and weight during Ramadan
609
fasting. Singapore Med J 47, 409–414.
610
43. Al-Hourani HM & Atoum MF (2007) Body composition,
611
nutrient intake and physical activity patterns in young
612
women during Ramadan. Singapore Med J 48, 906–910.
613
44. Furuncuoglu Y, Karaca E, Aras S et al. (2007) Metabolic,
614
biochemical and psychiatric alterations in healthy subjects
615
during Ramadan. Pak J Nutr 6, 209–211.
616
45. Ibrahim WH, Habib HM, Jarrar AH et al. (2008) Effect of
617
Ramadan fasting on markers of oxidative stress and serum
618
biochemical markers of cellular damage in healthy
619
subjects. Ann Nutr Metab 53, 175–181.
620
46. Hajek P, Myers K, Dhanji AR et al. (2012) Weight change
621
during and after Ramadan fasting. J Public Health (Oxf) 34,
622
377–381.
623
47. Unalacak M, Kara IH, Baltaci D et al. (2011) Effects of
624
Ramadan fasting on biochemical and hematological para-
625
meters and cytokines in healthy and obese individuals.
626
Metab Syndr Relat Disord 9, 157–161.
627
48. Poirier P, Giles TD, Bray GA et al. (2006) Obesity and
628
cardiovascular disease: pathophysiology, evaluation, and
629
effect of weight loss. Circulation 113, 898–918.
630
49. Look AHEAD Research Group; Pi-Sunyer X, Blackburn G,
631
Brancati FL et al. (2007) Reduction in weight and
632
cardiovascular disease risk factors in individuals with
633
type 2 diabetes: one-year results of the look AHEAD trial.
634
Diabetes Care 30, 1374–1383.
635
50. Nordmann AJ, Nordmann A, Briel M et al. (2006) Effects of
636
low-carbohydrate vs low-fat diets on weight loss and
637
cardiovascular risk factors: a meta-analysis of randomized
638
controlled trials. Arch Intern Med 166, 285–293.
639
51. Teixeira PJ, Silva MN, Coutinho SR et al. (2010) Mediators
640
of weight loss and weight loss maintenance in middle-aged
641
women. Obesity (Silver Spring) 18, 725–735.
642
52. Frost G & Pirani S (1987) Meal frequency and nutritional
643
intake during Ramadan: a pilot study. Hum Nutr Appl Nutr
644
41, 47–50.
645
53. Bakhotmah B (2011) The puzzle of self-reported weight
646
gain in a month of fasting (Ramadan) among a cohort
647
of Saudi families in Jeddah, Western Saudi Arabia. Nutr J
648
10, 84.
649
54. Gumaa KA, Mustafa KY, Mahmoud NA et al. (1978) The
650
effects of fasting in Ramadan. 1. Serum uric acid and lipid
651
concentrations. Br J Nutr 40, 573–581.
652
55. Sanders SW & Moore JG (1992) Gastrointestinal chrono-
653
pharmacology: physiology, pharmacology and therapeutic
654
implications. Pharmacol Ther 54, 1–15.
655
56. Hughes J (2007) Effects of abstinence from tobacco: valid
656
symptoms and time course. Nicotine Tob Res 9, 315–327.
Public Health Nutrition
Ramadan fasting and weight: a meta-analysis 11
