JECH Online First, published on August 20, 2013 as 10.

1136/jech-2013-203104
Theory and methods

Meta-analysis of prevalence
Jan J Barendregt,1 Suhail A Doi,1 Yong Yi Lee,1 Rosana E Norman,2 Theo Vos3
1

University of Queensland,
School of Population Health
2
University of Queensland,
School of Population Health
and Queensland Children's
Medical Research Institute
3
University of Queensland,
School of Population Health
and University of Washington,
Institute of Health Metrics and
Evaluation
Correspondence to
Dr Jan J Barendregt, School of
Population Health, University of
Queensland, Herston Rd,
Herston, QLD 4006, Australia;
j.barendregt@sph.uq.edu.au
Received 12 July 2013
Revised 24 July 2013
Accepted 27 July 2013

ABSTRACT
Meta-analysis is a method to obtain a weighted average
of results from various studies. In addition to pooling
effect sizes, meta-analysis can also be used to estimate
disease frequencies, such as incidence and prevalence. In
this article we present methods for the meta-analysis of
prevalence. We discuss the logit and double arcsine
transformations to stabilise the variance. We note the
special situation of multiple category prevalence, and
propose solutions to the problems that arise. We
describe the implementation of these methods in the
MetaXL software, and present a simulation study and
the example of multiple sclerosis from the Global Burden
of Disease 2010 project. We conclude that the double
arcsine transformation is preferred over the logit, and
that the MetaXL implementation of multiple category
prevalence is an improvement in the methodology of the
meta-analysis of prevalence.

distribution. With the main meta-analysis methods

based on the inverse variance method (or modications thereof ), the binomial equation for variance
(expressed as a proportion) can be used to obtain
the individual study weights:
p(1  p)
1
N
where p is the prevalence proportion, and N the
population size.
Thus, the pooled prevalence estimate P then
becomes (according to the inverse variance
method):
Var(p)

pi
Var(pi )
P i
P 1
i Var(pi )

s
X 1
SE(P)
Var(pi )
i

with SE:

INTRODUCTION
The large majority of meta-analyses are devoted to
establishing the effects of interventions, and therefore aim to get a pooled estimate of effect size, be
that relative risk, odds ratio (OR), risk difference,
or weighted or standardised mean difference.
However, meta-analysis methods can be useful as
well to get a more precise estimate of disease frequency, such as disease incidence rates and prevalence proportions. For example, the Global Burden
of Disease 2010 project aimed to obtain disease frequency estimates of a large number of diseases and
conditions, often based on a limited number of
studies, of usually varying quality.
This article looks at the meta-analysis of prevalence, using a concrete example from the Global
Burden of Disease 2010 study: the distribution of
severity in the prevalence of multiple sclerosis
(MS). We rst discuss the specic properties of
prevalence as a variable and the options to deal
with these in the meta-analysis, next look at the
multiple category case, then discuss the implementation of these methods in the MetaXL software,
present the simulation study and the example, and
conclude.

PREVALENCE AS A VARIABLE

To cite: Barendregt JJ,

Doi SA, Lee YY, et al. J
Epidemiol Community Health
Published Online First:
[please include Day Month
Year] doi:10.1136/jech2013-203104

While many epidemiologists will habitually speak

of prevalence rate, prevalence is dened as a proportion: the number of cases of a disease in a
population, divided by the population number.
This denition implies that (1) prevalence is always
between 0 and 1 (inclusive), and (2) the sum over
categories always equals 1.
The denition of prevalence is the same as the
denition of the binomial distribution (number of
successes in a sample), and therefore the standard
assumption is that prevalence follows a binomial

Barendregt
JJ, et al.Article
J Epidemiolauthor
Community(or
Health
2013;0:15.
doi:10.1136/jech-2013-203104
Copyright
their
employer)
2013. Produced

The CI of the pooled prevalence can then be

obtained by:
4
CIg (P) P + Za=2 SE(P)
where Za=2 denotes the appropriate factor from the
standard Normal distribution for the desired condence percentage (eg, Z0.025=1.96).

TRANSFORMATIONS
While this works ne for prevalence proportions
around 0.5, increasing problems arise when the
proportions get closer to the limits of the 0..1
range. The rst problem is mostly cosmetic: the
equation for the CI does not preclude condence
limits outside the 0..1 range. While this is annoying, the second problem is much more substantial:
when the proportion becomes small or big, the
variance of the study is squeezed towards 0 (see
equation (1)). As a consequence, in the inverse variance method, the study gets a large weight. A
meta-analysis of prevalence according to the
method described above therefore puts undue
weight on the studies at the extreme of the 0..1
range.
The way to deal with these problems is to transform the prevalence to a variable that is not constrained to the 0..1 range, has an approximately
normal distribution, and by being unconstrained
avoids the squeezing of variance effect. The
meta-analysis can be carried out on the transformed proportions, using the inverse of the variance of the transformed proportion as study
weight. For nal presentation, the pooled transformed proportion and its CI are back transformed

by BMJ Publishing Group Ltd under licence.

Theory and methods

to a proportion. We discuss two transformations that have been
used: the logit and the double arcsine.

Logit
The logit transformation is given by1:


p
l ln
1p

with variance:
Varl

1
1
+
Np N1  p

The back transformation to a proportion is done using:

exp(l)
7
exp(l) 1
However, while the logit transformation solves the problem of
estimates falling outside the 0..1 limits, unfortunately, it does
not succeed in stabilising the variance. Rather, there is a reversal
of the variance instability of the non-transformed proportions:
studies with proportions close to 0 or 1 get their variance estimates grossly magnied, while the variance for proportions
around 0.5 is squeezed. So a large study with a small prevalence
gets completely outweighed by a small study with prevalence
around 0.5. The variance instability that plagues nontransformed proportions thus persists even after logit
transformation.
p

Double arcsine
The double arcsine transformation is given by2:
r
r
n
n1
sin1
8
N1
N1
with n the number of people in the category. The variance of t
is given by:
t sin1

1
N 0:5
The back transformation to a proportion is done using3:
Var(t)

8
2

12 30:5 9
0
>
>
>
>
1
>
>
>
>
sin
t

<
7
6
B
C 7 =
sin
t
6
B
C
p 0:5 1  sgn(cos t)61  @sin t
A 7
>
5 >
4
N
>
>
>
>
>
>
:
;
10
with sgn being the sign operator. A much simpler but less
accurate alternative back transformation is:
 t 2
11
p sin
2
The double arcsine transformation addresses both the problem
of condence limits outside the 0..1 range and that of variance
instability, and it is therefore preferred over the logit
transformation.

MULTI-CATEGORY PREVALENCE
The discussion so far has implicitly been about single category
prevalence (those with the disease present which of course implies
2

a second category of those not in the rst). But in some instances

k-category prevalences may be meta-analysed where k>2 (representing different health states associated with the disease such as
mild, moderate and severe), and this complicates matters.
The rst thing to mention is that a multi-category prevalence from
a study is a single (vector-valued) observation, which suggests that a
single study weight across all categories applies. This is no problem
in the case of the double arcsine transformation, because its variance
depends on the population size (N) only. But for the untransformed
and logit transformation options the variance of both p and logit(p)
depends on p itself, and this implies that the same study would get a
different weight in each category, which seems hard to justify.
A second issue is the calculation of the heterogeneity statistic
Cochrans Q. With multiple categories, the standard calculation
for Q will return a different value for each category, again something that is hard to justify. Moreover, Q is used for further calculations, in the random effects model for 2, and in the quality
effects model for the overdispersion correction (see below).
The third issue is that because of the non-linearities introduced by the logit and double arcsine transformations, the backtransformed pooled proportions from multiple categories will
not necessarily add up to 1. We discuss solutions to these three
issues in the next section.

THE METAXL IMPLEMENTATION

MetaXL is a new, freely available, software program for
meta-analysis in Microsoft Excel, available from http://www.
epigear.com. It implements the standard xed effects (inverse
variance, MantelHaenszel, Peto) and random effects models.
MetaXL is unique in that it also implements a quality effects
model, which allows explicit accounting for study quality in the
meta-analysis.46
MetaXL implements three variants of meta-analysis of prevalence: untransformed, logit transformed and double arcsine
transformed prevalence, with the double arcsine transformation
as the default option. The user can specify as many categories as
desired, and can choose between xed effects (inverse variance
only), random effects and quality effects models.
The back-transformation in equation (10) can become numerically unstable when sin t is close to 0, and to avoid this, the double
arcsine back transformation is implemented as follows in MetaXL.
Let:

s


t 2
sin
2

12

where t is the pooled t. Then:



8
s
1s
>
>
s
if
,
2
or
,2
>
>
v
v
>
>
8
>
2
>

12 30:5 9
0
>
>
>
<
>
>
1
>
>
>
sin t 
<
=
7 >
6
P
B
C
sin
t
7
6
>
B
C
>
0:5
1

sgn(cos
t)
1

sin
t

7
6
>
@
A 5 >
>
1
>
4
>
>
>
>
>
>
>
>
>
>
:
;
>
v
:
otherwise

13

where P is the pooled prevalence and v is the pooled variance.

Instead of the harmonic mean, as suggested by Miller,3 we use
1=v because our simulation studies suggest that this is a more
stable estimate for N than the harmonic mean. This works
because the variance of a double arcsine transformed proportion
is approximately 1/N.

Barendregt JJ, et al. J Epidemiol Community Health 2013;0:15. doi:10.1136/jech-2013-203104

Theory and methods

The lower (LLC) and upper (ULC) limit are given by:

Prevalence and heterogeneity

8
P
>
>
0
if , 2
>
>
>
v
>
8
>
2

12 30:5 9
>
0
>
>
< >
>
>
1
>
>
>
sin t 
<
=
7 >
6
LLC
B
C
7
6
sin
t
>
C 7
>
1B
sin t
> 0:5>1  sgn(cos t)6
@
A
>
1
5 >
4
>
>
>
>
>
>
>
>
>
>
;
v
> :
:
otherwise

14

and
8


>
1P
>
>
,2
1
if
>
>
v
>
>
8
>
2
>

12 30:5 9
0
<
>
>
>
>
1
>
>
>
>
ULC
sint 
<
7
6
B
C 7 =
>
6
sint
>
B
C
>
0:5
1

sgn(cost)
1

sint

6
>
@
A 7
>
1
>
5 >
4
>
>
>
>
>
>
>
>
>
>
:
;
:
v
otherwise

15

p
where t t + Za=2 v.
The three issues caused by generalising to more than two categories are dealt with as follows. To obtain a single study weight
across categories for the untransformed and logit transformed
prevalences, MetaXL uses the inverse of the average of the category variances.
For the second issue, to obtain a single estimate for Cochrans
Q, we have chosen to take the maximum category Q. This is
because if we believe that the effect size variations across studies
in one category of proportions is not independent of variations
in the other categories, the maximum category Q value would
be the best and most conservative estimate of a common Q.
For the random effects model, a single 2 is calculated in the
usual way from this common Q and this 2 is added to the previously computed average variance across categories. Inverting
this inated variance obtains the random effects weight for each
study. The pooled variance for each category is computed by
rst adding 2 to category variances, invert to a weight, and sum
the category weights over studies. The pooled category variance
is then obtained from the reciprocal of the summed category
weights.
In the quality effects model, the redistribution of inverse variance weights is done using a quality parameter between 0
(lowest quality) and 1 (highest quality), individualised to each
study rather than the common 2 parameter across studies as is
done with the random effects model.5 6 In addition to this, the
quality effects model applies an overdispersion correction based
on Cochrans Q 2 statistic, resulting in a more conservative CI
in the presence of heterogeneity.4 Note that, unlike in the
random effects model, this increase does not affect the study
weights. For the multi-category prevalence analysis, the overdispersion correction is applied to each of the categories separately,
thus boosting the CIs in case of heterogeneity.
The third and nal issue is that, due to the transformations,
the back transformed proportions may not sum to 1. The difference is usually small, and MetaXL deals with it by offering the
option to normalise the prevalences in each category after
pooling and back transformation. The CIs, however, are not
adjusted.

Heterogeneity is the main issue with meta-analysis. When study

results are heterogeneous, we cannot assume that the same phenomenon has been measured in a sufciently similar way and
that differences in results are due to sampling error only. The
original random effects model assumed that the observed phenomenon follows a Normal distribution, and that individual
studies are random draws from this distribution.7
We have argued that, while the observed phenomenon may
well have a Normal distribution, the largest part of the heterogeneity is most likely due to differences in the way the studies
were carried out: differences in case denitions, biases, etc.46 8
The quality effects model was developed to take these differences explicitly into account by way of a study quality score,
thus giving lower weights to studies of lower quality.
In Burden of Disease studies, the aim is to get a best estimate
of disease frequency (in this case prevalence), based on available
data. When studies produce heterogeneous results, the rst
response is to look for covariates that may explain the heterogeneity, and stratify studies into more homogeneous subgroups
accordingly. But within those subgroups, the assumption is that
the studies have been measuring the same phenomenon, albeit
with varying quality. Therefore, the quality effects model is, in
our opinion, clearly the best choice for this purpose. However,
this will not be discussed further as the purpose of this paper is
to compare properties of various transformations of proportions, and not to compare meta-analysis models.

SIMULATION
We carried out a simulation study to compare bias and coverage
between the transformation methods, based on a hypothetical
dataset. We dened nine studies of increasing size, with the
smallest comprising 20 subjects, and each subsequent study
having 20 more subjects, resulting in 180 subjects for the largest
study. We assumed the number of cases in each study to have a
binomial distribution, with study size and prevalence proportion
as the parameters.
For the xed effects model, we assumed a prevalence of 0.05,
and for the random effects model we assumed the prevalence to
have a Normal distribution with parameters 0.05 and 0.005, for
mean and SD, respectively. For the quality effects model, we
assumed that quality associated deviations from the true prevalence would also follow a Normal distribution with parameters
0.05 and 0.005. From the deviation we calculated a quality
score between 0 and 1 using a quadratic function. (Note that
the quality effects model does not require that the quality score
is a function of the difference between the true effect and the
study estimate, but for this simulation this provides a convenient
way to calculate a quality score.)
We implemented the numbers and functions in Excel, using
the Ersatz Monte Carlo simulation add-in, and drew randomly
from the distributions 1000 times.9 Each time we recalculated
the meta-analysis for the three transformation options for each
model. We calculated the mean of the central estimates, the bias
in this mean and the mean squared error, and determined the
coverage proportion, given the pooled CIs.
Results of the simulation study are presented in table 1.
Without transformation, the effect of the squeezed variance for
a prevalence of this size can clearly be seen: the mean pooled
prevalences are much lower than the true prevalence of 0.05,
dragged down by the large weights applied to the samples with
low prevalence. For the logit transformation, a reverse (but
much smaller) effect can be seen: the lowest sampled

Barendregt JJ, et al. J Epidemiol Community Health 2013;0:15. doi:10.1136/jech-2013-203104

Theory and methods

Table 1 Results from the simulation study: mean, bias, mean
squared error (MSE) and coverage, by transformation option and
model
Transformation
Model

Variable

None

Logit

Double arcsine

Fixed effects

Mean
Bias
MSE
Coverage
Mean
Bias
MSE
Coverage
Mean
Bias
MSE
Coverage

0.02810
0.02190
0.00077
0.39400
0.04215
0.00785
0.00014
0.82200
0.03111
0.01889
0.00064
0.63200

0.05428
0.00428
0.00007
0.92500
0.05373
0.00373
0.00007
0.94700
0.05263
0.00263
0.00006
0.95500

0.05206
0.00206
0.00006
0.94300
0.05162
0.00162
0.00006
0.96100
0.05145
0.00145
0.00006
0.95800

Random effects

Quality effects

prevalences get less weight, and the mean prevalence therefore

is overestimated. Both logit and double arcsine do much better
than the no transformation option across all variables, with the
double arcsine doing slightly better than the logit for most variables, but much better for bias.
For prevalences close to 0.5, there is hardly any difference
between the results from the three transformation options (data
not shown).

stages mild, moderate, and severe. A total of 18 studies were

found to contain usable data on the distribution of MS severity.
A quality score was given to each study based on the method of
measuring functional impairment, with higher preference given
to studies that assessed impairment using both a physician and a
standardised scale. Kurtzkes Expanded Disability Status Scale
(EDSS) is considered by experts to be the gold standard for
measuring MS functional impairment and studies using this
were given the highest quality score of 1. The majority of
studies in table 2 used EDSS, as demonstrated by the frequent
score of 1. Three studies assessed MS severity with no emphasis
on scores or scales used and were given a nal quality score
of 0.6.
Table 2 shows the studies in this analysis, their distribution
over the three categories, their quality and population size
scores. From the table, it is clear that substantial heterogeneity
exists, and the I2 for these studies is 95.2% (95% CI 93.6% to
96.4%).
Table 3 shows the pooled results, with the normalisation option
of MetaXL switched on. The pooled estimates do not differ much
between models and transformation methods, with the random
effects and to a lesser extent the quality effects model producing a
bit larger proportion for the mild category at the expense of the
severe one. In the transformed analyses the CIs of the random
effects model are the widest: because the quality scores are mostly
1, the quality effects model CIs are not as wide.

Table 3 Pooled results and CIs for three categories, by

transformation method and model

EXAMPLE: MULTIPLE SCLEROSIS

In this section we present an example from the Global Burden
of Disease 2010 Study where systematic reviews of the scientic
literature and quantitative meta-analyses were conducted in
order to estimate the severity distribution of various disorders at
the global level. We look at the distribution of cases of MS over

Model

Transformation Category Pooled LCI

HCI

Range

Inverse variance

None

Logit

Arcsine

Table 2 Study size, prevalence by category and quality scores of

multiple sclerosis studies

Study name

Country

Mild Moderate Severe Quality

Casetta 199810
Balasa 200711
Tsai 200412
Houzen 200313
Chancellor 200314
Cabre 200115
Bencsik 199816
Lobinska 200417
Bencsik 200118
Benedikz 200219
Alvarez 199220
Arruda 200121
Al-Araji 200522
Pittock 2004 (1)23
Pittock 2004 (2)23
Modrego Pardo 199724
McDonnell 199825
Tola 199926

Italy
Romania
Taiwan
Japan
New Zealand
Caribbean
Hungary
Poland
Hungary
Iceland
Chile
Brazil
Iraq
USA
USA
Spain
Ireland
Spain

394
152
43
31
86
62
130
204
248
290
100
200
300
162
201
46
258
54

0.561
0.441
0.651
0.645
0.512
0.645
0.785
0.632
0.149
0.710
0.550
0.610
0.400
0.525
0.627
0.609
0.322
0.574

0.175
0.250
0.233
0.129
0.337
0.306
0.162
0.260
0.540
0.179
0.330
0.225
0.373
0.296
0.224
0.217
0.477
0.296

0.264
0.309
0.116
0.226
0.151
0.048
0.054
0.108
0.310
0.110
0.120
0.165
0.227
0.179
0.149
0.174
0.202
0.130

1.00
1.00
0.60
1.00
1.00
1.00
1.00
1.00
0.60
1.00
0.60
1.00
1.00
1.00
1.00
1.00
1.00
1.00

Random effects

None

Logit

Arcsine

Quality effects

None

Logit

Arcsine

Mild
Moderate
Severe
Mild
Moderate
Severe
Mild
Moderate
Severe

0.539
0.281
0.179
0.511
0.297
0.192
0.525
0.289
0.185

0.523
0.266
0.167
0.477
0.271
0.173
0.502
0.270
0.170

0.556
0.297
0.192
0.515
0.305
0.201
0.537
0.302
0.198

0.033
0.031
0.026
0.038
0.034
0.029
0.036
0.033
0.028

Mild
Moderate
Severe
Mild
Moderate
Severe
Mild
Moderate
Severe

0.551
0.280
0.169
0.560
0.278
0.162
0.555
0.279
0.166

0.470
0.199
0.090
0.458
0.207
0.115
0.464
0.204
0.106

0.632
0.360
0.249
0.630
0.346
0.212
0.632
0.355
0.232

0.162
0.161
0.160
0.172
0.139
0.097
0.168
0.151
0.126

Mild
Moderate
Severe
Mild
Moderate
Severe
Mild
Moderate
Severe

0.552
0.272
0.176
0.527
0.286
0.187
0.540
0.279
0.181

0.441
0.192
0.129
0.464
0.244
0.156
0.461
0.224
0.143

0.663
0.352
0.223
0.562
0.317
0.212
0.607
0.331
0.219

0.223
0.160
0.094
0.097
0.073
0.056
0.145
0.107
0.076

HCI, higher CI; LCI, lower CI.

Barendregt JJ, et al. J Epidemiol Community Health 2013;0:15. doi:10.1136/jech-2013-203104

Theory and methods

CONCLUSION
We discussed the meta-analysis of prevalence, and two transformation options used: logit and double arcsine transformed
proportions. We signalled issues with the meta-analysis of multiple categories, and proposed solutions for these issues.
We discussed the implementation of these methods in
MetaXL, including a normalisation option for multiple category
prevalence, and presented results from a simulation study and
an example. We conclude that the double arcsine is the preferred transformation and that the MetaXL implementation of
multiple category prevalence is an improvement in the methodology of the meta-analysis of prevalence.

REFERENCES
1
2
3
4
5
6
7
8

What is already known on this subject

Meta-analysis of prevalence using inverse variance methods
has the problem that the variance becomes very small when
the prevalence is small or large, with the consequence that
such studies get a large weight in the meta-analysis.
Transformation methods can be used to avoid an undue
large weight for studies with small or large prevalence.

9
10
11
12
13
14
15
16

What this study adds

The double arcsine transformation is discussed and we show
it has properties that make it the clearly preferred option
over the often used logit transformation.
The generalisation of the methods to multiple category
prevalence and the implementation in the MetaXL software
package are introduced.
We conclude that the resulting MetaXL software is an
improvement in the meta-analysis of prevalence.

17
18

19

20
21

22

Contributors JJB and SAD developed the methods, YYL, RN and TV did the
meta-analyses, all authors contributed to writing.
Competing interests JJB owns Epigear International Pty Ltd, which sells the
Ersatz software used in the analysis.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement The data are available from the Epigear website, and
are part of the MetaXL download.

23

24
25
26

Lipsey M, Wilson DB. Practical meta-analysis. Thousand Oaks: Sage Publications,

2001.
Freeman MF, Tukey JW. Transformations related to the angular and the square root.
Ann Math Stat 1950;21:60711.
Miller JJ. The inverse of the Freeman-Tukey double arcsine transformation. Am Stat
1978;32:138.
Doi SA, Barendregt JJ, Mozurkewich EL. Meta-analysis of heterogeneous clinical
trials: an empirical example. Contemp Clin Trials 2011;32:28898.
Doi SA, Thalib L. A quality-effects model for meta-analysis. Epidemiology
2008;19:94100.
Doi SA, Thalib L. An alternative quality adjustor for the quality effects model for
meta-analysis. Epidemiology 2009;20:314.
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials
1986;7:17788.
Al Khalaf MM, Thalib L, Doi SA. Combining heterogeneous studies using the
random-effects model is a mistake and leads to inconclusive meta-analyses. J Clin
Epidemiol 2011;64:11923.
Ersatz [ program]. 1.2 version. Brisbane: Epigear (http://www.epigear.com), 2012.
Casetta I, Granieri E, Marchi D, et al. An epidemiological study of multiple sclerosis
in central Sardinia, Italy. Acta Neurol Scand 1998;98:3914.
Balasa R, Feier C, Dan M, et al. The prevalence of multiple sclerosis in Mures
county, central Romania. Revista Romana de Neurologie 2007;6:804.
Tsai CP, Yuan CL, Yu HY, et al. Multiple sclerosis in Taiwan. J Chin Med Assoc
2004;67:5005.
Houzen H, Niino M, Kikuchi S, et al. The prevalence and clinical characteristics of
MS in northern Japan. J Neurol Sci 2003;211:4953.
Chancellor AM, Addidle M, Dawson K. Multiple sclerosis is more prevalent in
northern New Zealand than previously reported. Intern Med J 2003;33:7983.
Cabre P, Heinzlef O, Merle H, et al. MS and neuromyelitis optica in Martinique
(French West Indies). Neurology 2001;56:50714.
Bencsik K, Rajda C, Klivenyi P, et al. The prevalence of multiple sclerosis in the
Hungarian city of Szeged. Acta Neurol Scand 1998;97:31519.
Lobinska A, Stelmasiak Z. [Epidemiological aspects of multiple sclerosis in Lublin
(Poland)]. Neurol Neurochir Pol 2004;38:3616.
Bencsik K, Rajda C, Fuvesi J, et al. The prevalence of multiple sclerosis, distribution
of clinical forms of the disease and functional status of patients in Csongrad
County, Hungary. Eur Neurol 2001;46:2069.
Benedikz J, Stefansson M, Guomundsson J, et al. The natural history of untreated
multiple sclerosis in Iceland. A total population-based 50 year prospective study.
Clin Neurol Neurosurg 2002;104:20810.
Alvarez G, Castillo JL, Ruiz F, et al. Multiple sclerosis in Chile. Acta Neurol Scand
1992;85:14.
Arruda WO, Scola RH, Teive HA, et al. Multiple sclerosis: report on 200 cases from
Curitiba, Southern Brazil and comparison with other Brazilian series. Arq
Neuropsiquiatr 2001;59:16570.
Al-Araji A, Mohammed AI. Multiple sclerosis in Iraq: does it have the same features
encountered in Western countries? J Neurol Sci 2005;234:6771.
Pittock SJ, Mayr WT, McClelland RL, et al. Disability prole of MS did not change
over 10 years in a population-based prevalence cohort. Neurology 2004;
62:6016.
Modrego Pardo PJ, Latorre MA, Lopez A, et al. Prevalence of multiple sclerosis in
the province of Teruel, Spain. J Neurol 1997;244:1825.
McDonnell GV, Hawkins SA. An epidemiologic study of multiple sclerosis in
Northern Ireland. Neurology 1998;50:4238.
Tola MA, Yugueros MI, Fernandez-Buey N, et al. Prevalence of multiple sclerosis in
Valladolid, northern Spain. J Neurol 1999;246:1704.

Barendregt JJ, et al. J Epidemiol Community Health 2013;0:15. doi:10.1136/jech-2013-203104