Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.

com

Multiple sclerosis

RESEARCH PAPER

High consumption of coffee is associated with

decreased multiple sclerosis risk; results from two
independent studies
Open Access
Scan to access more
free content
1

Institute of Environmental
Medicine, Karolinska Institutet,
Stockholm, Sweden
2
Department of Neurology,
Johns Hopkins University,
Baltimore, Maryland, USA
3
Division of Epidemiology,
Genetic Epidemiology and
Genomics Lab, School of Public
Health, University of California,
Berkeley, California, USA
4
Kaiser Permanente Division of
Research, Oakland, California,
USA
5
Neuroimmunology Unit,
Department of Clinical
Neuroscience and Center for
Molecular Medicine, Karolinska
Institutet at Karolinska
University Hospital, Solna,
Sweden
Correspondence to
Dr A K Hedstrm, Institute of
Environmental Medicine,
Karolinska Institutet, Nobels
vg 13, Stockholm 17177,
Sweden; [email protected]
AKH, EMM, LFB and LA
contributed equally to this
paper.
Received 2 September 2015
Revised 6 January 2016
Accepted 11 January 2016
Published Online First
3 March 2016

A K Hedstrm,1 E M Mowry,2 M A Gianfrancesco,3 X Shao,3 C A Schaefer,4 L Shen,4

T Olsson,5 L F Barcellos,3 L Alfredsson1
ABSTRACT
Objective Previous studies on consumption of caffeine
and risk of multiple sclerosis (MS) have yielded
inconclusive results. We aimed to investigate whether
consumption of coffee is associated with risk of MS.
Methods Using two population-representative case
control studies (a Swedish study comprising 1620 cases
and 2788 controls, and a US study comprising 1159
cases and 1172 controls), participants with different
habits of coffee consumption based on retrospective
data collection were compared regarding risk of MS, by
calculating ORs with 95% CIs. Logistic regression
models were adjusted for a broad range of potential
confounding factors.
Results Compared with those who reported no coffee
consumption, the risk of MS was substantially reduced
among those who reported a high consumption of
coffee exceeding 900 mL daily (OR 0.70 (95% CI 0.49
to 0.99) in the Swedish study, and OR 0.69 (95% CI
0.50 to 0.96) in the US study). Lower odds of MS with
increasing consumption of coffee were observed,
regardless of whether coffee consumption at disease
onset or 5 or 10 years prior to disease onset was
considered.
Conclusions In accordance with studies in animal
models of MS, high consumption of coffee may decrease
the risk of developing MS. Caffeine, one component of
coffee, has neuroprotective properties, and has been
shown to suppress the production of proinammatory
cytokines, which may be mechanisms underlying the
observed association. However, further investigations are
needed to determine whether exposure to caffeine
underlies the observed association and, if so, to evaluate
its mechanisms of action.

BACKGROUND
http://dx.doi.org/10.1136/
jnnp-2015-312431

To cite: Hedstrm AK,

Mowry EM,
Gianfrancesco MA, et al. J
Neurol Neurosurg Psychiatry
2016;87:454460.
454

Coffee contains more than a thousand biologically

active compounds of which caffeine, a stimulant of
the central nervous system, is the most studied
agent. Caffeine intake has been inversely associated
with cardiovascular disease,1 stroke2 and type 2
diabetes mellitus,3 and a recent doseresponse
meta-analysis showed that coffee consumption was
associated with a decreased risk of mortality from
all causes and cardiovascular disease.4 5 In animal
models of Alzheimers disease, caffeine exerts protective effects at least in part by protecting against
bloodbrain barrier leakage.6 By upregulation of
adenosine A1 receptors, consumption of caffeine
attenuates neuroinammation and demyelination in

animal models of MS.79 The relationship between

consumption of caffeine and risk of MS has been
investigated in several casecontrol studies that generated inconsistent results,1012 while one prospective study found no apparent association.13 Using
data from two large casecontrol studies, we aimed
to investigate whether consumption of coffee is
associated with risk of MS.

METHODS
Design and study population
The present report was based on data from two
casecontrol studies of environmental and genetic
risk factors for MS. The rst study is EIMS
(Epidemiological
Investigation
of
Multiple
Sclerosis) with a study base comprising the Swedish
population aged 1670 years. Potential incident
cases of MS, including those with clinically isolated
syndrome, were recruited via 40 study centres,
including all university hospitals in Sweden. All
cases were examined and diagnosed by a neurologist located at the unit where the case was entered.
The treating neurologist provided information
regarding the patients onset of disease, and
whether the patient fullled the McDonald criteria.14 For each potential case, two controls were
randomly selected from the national population
register, matched by age in 5-year strata, sex and
residential area. Only cases that fullled the
McDonald criteria were included in the analyses.
Cases that did not full the criteria at the time of
this report were excluded, but not their corresponding controls. All aspects of the study were
approved by the Regional Ethical Review Board at
Karolinska Institutet.
The other casecontrol study recruited prevalent
cases identied among members of Kaiser
Permanente Medical Care Plan, Northern
California Region (KPNC) using electronic health
records. KPNC is an integrated health services
delivery system with a membership of 3.3 million
that comprises about 2530% of the population of
a 22-county service area in northern California.
Cases of MS, aged 18 years through 69 years, were
required to be current KPNC members who had
received the MS diagnosis by a neurologist.
Diagnoses were validated by chart review and radiology and pharmacy records, according to
McDonald criteria.15 The treating neurologist of
each potential MS case was contacted to gain
approval regarding contacting the case; potential

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

Multiple sclerosis
cases who did not have MS, were deemed unable to participate
due to being too severely ill or impaired, or who were no
longer KPNC members, were thus excluded by their neurologists. Controls were randomly selected from KPNC members;
they did not have a diagnosis of MS or related conditions and
were individually matched to cases on sex, birth date, race/ethnicity and zip code of the case residence. The study protocol was
approved by the Institutional Review Boards of the KP Division
of Research and the University of California, Berkeley. Details
of the research resource have been described elsewhere.15

Data collection
In EIMS, information on lifestyle factors and different exposures was collected using a standardised questionnaire given to
the cases shortly after they had received their diagnosis, and was
sent by mail to the controls. During the study period (April
2005 to March 2013), completed questionnaires were obtained
from 2055 cases who fullled the McDonald criteria,13 and
from 4518 matched controls, which is equivalent to 93% of
invited cases and 73% of invited controls. In November 2013,
complementary questions were sent to all participants who had
answered the standardised questionnaire during the aforementioned period. Among other questions, participants were asked
to report their coffee consumption during different age periods
by asking How many cups of coffee did you usually drink per
day when you were 1519, 2029, 3039 and 40+ years old?.
For each age period, the answer options were 0, 13, 34, 56,
78 and 8 or more cups of coffee daily. Those who reported 7
8 or 8 or more cups of coffee daily were combined into one category since few participants consumed more than 8 cups of
coffee daily. The complementary questions were answered by
82% of the cases and 66% of the controls. Information regarding coffee consumption was missing for 57 cases and 166 controls, and these individuals were excluded. Cases younger than
15 years of age at disease onset and their corresponding controls
were also excluded. The present study is thus comprised of
1620 cases and 2788 controls.
In the KPNC study, participants completed a computerassisted telephone interview at study entry administered by
trained staff regarding lifestyle factors and various exposures;
recruitment began in mid-2007. As of a recent data freeze in
August 2014, the study included a total of 1479 cases and 1185
controls. Within this data set, there were 1163 cases and 1178
matched controls. The study participation proportions were
approximately 80% for cases and 66% for controls. Information
on consumption habits of caffeinated coffee was obtained by
asking participants the highest number of servings per day they
have ever been consumed for 6 months or more (answer
options were 0, 1, 23 and 4 or more cups of caffeinated coffee
daily), and those who reported consuming one or more servings
of any caffeinated beverage were asked to estimate at what age
they started drinking caffeine on a regular basis. Information
regarding coffee consumption was missing for four cases and six
controls, and these individuals were excluded. The results based
on the KPNC study are thus comprised of 1159 cases and 1172
controls.
For each case in both studies, the year of the initial appearance of symptoms indicative of MS was dened as the index
year. Coffee consumption was considered prior to the index
year in the cases and during the same period of time in the corresponding controls. Participants from both data sets were categorised into groups based on their daily consumption of coffee
(number of cups of coffee). The size of one cup of coffee varies
between the countries. In Sweden, one cup of coffee was

dened as 150 mL, whereas a cup of coffee was dened as

237 mL (8 oz) in the KPNC study.

Confounding factors and other covariates

All analyses in EIMS were adjusted for relevant demographic
factors as well as for factors previously associated with MS risk.
These factors included the design variables age, sex, residential
area, as well as ancestry, smoking habits (number of pack
years),16 exposure to passive smoking,17 sun exposure habits,18
and body mass index at age 20 years.19 20 Age was categorised
into the same eight intervals that were used in the matching procedure: 1619, 2024, 2529, 3034, 3539, 4045, 4549
and 5070 years of age. Assessment of ancestry was based on
whether the participant was born in Sweden or not, and
whether either of the participants parents had immigrated to
Sweden. A participant who was born in Sweden and whose
parents had not immigrated, was classied as Swedish. Smoking
was adjusted for by using a continuous variable ( pack years of
smoking). Exposure to passive smoking was dichotomised into
ever-exposed or never-exposed before the index year. On the
basis of three questions regarding ultraviolet radiation (UVR)
exposure where each answer option was given a number ranging
from 1 (the lowest exposure) to 4 (the highest exposure), we
constructed an index by adding the numbers together, and thus,
acquired a value between 3 and 12. UVR exposure was adjusted
for as a continuous variable. Body mass index at age 20 years
was calculated by dividing self-reported weight in kilograms by
self-reported height in metres squared, and adjusted for as a
continuous variable.
Additional covariates included educational level (compulsory
school, vocational upper secondary school, theoretical upper
secondary school, other education and university), socioeconomic status (skilled and unskilled workers, assistant, intermediate, and higher non-manual employees), alcohol
consumption at inclusion in the study (yes or no), a history of
infectious mononucleosis (yes or no) and HLA-DRB1*15 status
(positive or negative), but these factors did not signicantly
inuence the association between coffee consumption and MS
and were not retained in the nal models.
In the KPNC study, sex, birth date, race/ethnicity and zip
code of the case residence were taken into consideration by
matched analysis. Adjustment was made for similar potential
confounders as in the EIMS study, including smoking habits,
passive smoking, sun exposure habits and body mass index in
ones 20s21 (average of self-reported highest and lowest nonpregnancy weight during ones 20s, divided by self-reported
height in metres squared at the time of interview). Smoking was
dichotomised into ever-smokers and never-smokers before the
index year. Exposure to passive smoking was dichotomised into
those who had and had not been exposed to passive smoking
after the age of 18 years.
Sun exposure was dichotomised into high exposure (always
or almost always sunbathed in the summer at the age of
10 years), and low exposure (sometimes, rarely or never sunbathed in the summer at the age of 10 years). Body mass index
in ones 20s was entered as a continuous variable. Adjustments
were also made for educational level (college or no college education, or unknown), a history of infectious mononucleosis
(yes, no or unknown), and HLA-DRB1*15 status ( positive,
negative, or unknown), but these factors had only minor inuence on the results. Similar to analyses for EIMS, these factors
were not retained in the nal models as there was no signicant
change in the association of coffee consumption and the odds
of MS.

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

455

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

Multiple sclerosis
Statistical analysis
Using logistic regression, the occurrence of MS among participants with different coffee consumption habits was compared
with those who never drank coffee, by calculating ORs with
95% CIs. In EIMS, we investigated the inuence of coffee consumption at the index year, and 5 and 10 years prior to the
index year. We performed both matched and unmatched analyses of EIMS. In the unmatched analysis we were able to
include a higher number of controls, and therefore had
increased statistical power. Only the results from the unmatched
analyses are presented in this report since these were in close
agreement with those from the matched analyses but had tighter
CIs. A trend test for a doseresponse relationship regarding
coffee consumption and risk of MS was performed by using a
numerical integer variable for coffee consumption, ranging from
0 to 4 in EIMS, and 0 to 2 in KPNC.
In KPNC, we investigated the inuence of coffee consumption
at the index year, and 5 years prior to the index year, using conditional logistic regression. We assumed that consumption of coffee
began at the age the subjects rst began consuming caffeinated
beverages. For each individual, the reported highest number of
servings per day that coffee had ever been consumed for
6 months or more was considered to represent typical consumption during the whole period of regular coffee consumption. A
trend test for a doseresponse relationship regarding coffee consumption and risk of MS was performed by using a numerical
variable for coffee consumption ranging from 0 to 2. To differentiate if results were related to an overall effect of caffeine or were
likely attributable to coffee intake specically, we also conducted
analyses based on servings of tea and soda separately, using the
same assumptions as we did for coffee for consumption before
onset. This analysis was performed on KPNC data only. Data on
non-caffeinated beverages were neither collected for the KPNC
study participants nor for the EIMS participants. In Sweden, consumption of non-caffeinated beverages is rare.
We also conducted a meta-analysis in which the results of the
two casecontrol studies were combined by calculating a
weighted average of the two study-specic adjusted ORs
(weights were proportional to the inverse of the variance for
each OR). All analyses were conducted using Statistical Analysis
System (SAS) V.9.2.

RESULTS
In the EIMS and KPNC cohorts, coffee consumption was associated with sex, smoking, passive smoking and adolescent body
mass index among both cases and controls. Characteristics of
cases and controls, by the amount of daily coffee consumption
at the index year, are presented in table 1.
In EIMS, coffee consumption, both during the index year or 5
or 10 years prior to the index year, was associated with reduced
odds of developing MS as compared with participants who
reported no coffee consumption. The adjusted OR was 0.70
(95% CI 0.49 to 0.99, p=0.04) among those who drank more
than six cups of coffee (more than 900 mL) daily at the index
year. The corresponding OR for those who reported high coffee
consumption 5 or 10 years prior to the index year were 0.72
(95% CI 0.51 to 1.03, p=0.08) and 0.71 (95% CI 0.47 to 1.06,
p=0.09), respectively (table 2). Similar results were observed in
the KPNC study. Among those who had started drinking coffee
at any point before the index year and consumed four or more
cups of coffee (more than 948 mL) daily, the OR of developing
MS was 0.69 (95% CI 0.50 to 0.96, p=0.05) compared with
those who never drank coffee. Similarly, drinking four or more
cups of coffee daily at least 5 years prior to the index year was
456

associated with reduced odds of developing MS (OR 0.64, 95%

CI 0.45 to 0.91, p=0.04) (table 2). No evidence for association
was observed between increasing amounts of tea or soda intake
and MS. Doseresponse relationships were observed that
showed reduced odds of MS with increasing coffee consumption
(table 2). When results from both studies were combined in
meta-analysis, the OR was 0.71 (95% CI 0.55 to 0.92) when
participants with the highest consumption of coffee (>900 mL
of coffee daily in the Swedish study and >948 mL of coffee
daily in the US study) were compared with participants who
never drank coffee. The corresponding OR was similar when
the analysis was stratied by smoking status (OR 0.72 (95% CI
0.45 to 1.13) among never-smokers and OR 0.65 (95% CI 0.44
to 0.96) among ever smokers).

DISCUSSION
Compared with participants who reported no coffee consumption, the odds of MS were reduced among those who reported
high coffee consumption. These results are in line with similar
observations in studies using animal models of MS.7 These
results are also interesting in light of the fact that coffee and caffeine have both been associated with a reduced risk of
Parkinsons disease.22 23 There are several potential mechanisms
by which coffee consumption may be important in MS. By
upregulation of adenosine 1A receptors, caffeine treatment
exerts a protective effect against experimental autoimmune
encephalomyelitis.7 8 Furthermore, in vitro caffeine treatment
of human monocytoid cells increased the expression of adenosine 1A receptors and reduced proinammatory cytokine production.6 Coffee consumption has also been associated with
reduced progression of disability in relapsing onset MS,
although this study was cross-sectional, and a causal relationship
could not be conrmed.24
The relationship between caffeine consumption and MS risk
has been investigated in several studies that generated inconsistent results.1013 In a casecontrol study comprising 93 cases and
186 controls of which 92 were hospital controls and 94 population controls, an increased risk of MS was observed among subjects who consumed coffee before age 15 years, where no
association was found between risk of MS and coffee intake after
age 15 years.10 A hospital-based, casecontrol study using 210
incident cases and 210 individually matched controls observed
an increased risk of MS with increasing coffee intake.11 An
inverse association between different eating patterns, which
included coffee, was observed in a casecontrol study comprising
75 cases and 75 controls recruited from the cases relatives, or
matched for age, economic and educational levels.12 Coffee consumption was not investigated separately in this study, however.
The only prospective cohort study (Nurses Health Study, NHS)
that has been carried out found no association between caffeine
intake and MS risk.13 It is possible that the much smaller
number of MS cases present in the NHS study (n=282, of
whom only 44 drank three or more cups of coffee daily, (ie, the
highest exposure category in NHS drank less than those in the
highest category of the present investigation) or other differences
in the patient populations (eg, inclusion of only female nurses in
this cohort) explains the discrepancy in ndings.
Both casecontrol studies used in our analyses have some limitations. Since information on exposure was gathered retrospectively, recall bias may be a concern, and given the lack of
validated instruments, misclassication bias may also have
occurred. However, the relationship between coffee consumption and MS risk had not been investigated until recently, and
since the results of previous studies have been inconsistent, the

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

Characteristics of MS cases and controls, by the amount of daily coffee consumption at the index year, and of non-responders with regard to the complementary questions

EIMS

Women (n, %)
Men (n, %)
Swedish (n, %)
Smoking (n, %)
Passive smoking (n, %)
Adolescent BMI (kg/m2, SD)
UVR exposure (SD)
Age at disease onset (SD)
Age at inclusion in study (SD)
Age at complementary questions (SD)
Total

No coffee
consumption

12 Cups of coffee
daily

34 Cups of coffee
daily

56 Cups of coffee
daily

More than 6 cups of

coffee daily

Controls

Cases

Controls

Cases

Controls

Cases

Controls

Cases

Controls

282 (81)
66 (19)
277 (80)
114 (33)
126 (36)
22.6 (4.3)
6.2 (1.9)

456 (79)
121 (21)
432 (75)
143 (25)
181 (31)
21.9 (3.9)
6.6 (2.1)

393 (76)
127 (24)
408 (78)
255 (49)
214 (41)
22.2 (3.6)
6.4 (1.8)

678 (78)
189 (22)
660 (76)
358 (41)
344 (40)
21.5 (3.3)
6.7 (1.9)

353 (74)
126 (26)
385 (81)
284 (59)
228 (48)
22.4 (3.4)
6.3 (1.8)

632 (70)
265 (30)
722 (81)
467 (52)
412 (46)
21.7 (2.9)
6.6 (1.9)

119 (61)
77 (39)
162 (83)
134 (68)
116 (59)
22.8 (3.6)
6.0 (1.9)

210
109
268
197
140
21.9
6.5

42
32
62
58
47
23.4
5.6

72
56
110
86
72
22.6
6.0

348

577

523

867

479

897

196

319

Non-responders

Total

Cases

Controls

Cases

Controls

Cases

445 (68)
206 (32)
495 (77)
363 (56)
270 (41)
22.9 (4.0)
6.1 (1.9)
33.4 (10.4)
37.3 (10.6)

1342
589
1360
847
740
22.1
6.5

1192 (74)
428 (26)
1294 (80)
845 (52)
731 (45)
22.4 (3.7)
6.2 (1.8)
35.4 (10.8)
39.8 (11.5)
43.8 (11.7)
1620

2048 (73)
740 (27)
2192 (79)
1251 (45)
1149 (41)
21.7 (3.2)
6.6 (1.9)
40.5 (11.5)
44.3 (11.7)
2788

(69)
(31)
(71)
(44)
(39)
(3.5)
(2.0)

37.4 (10.8)

(66)
(34)
(85)
(62)
(44)
(2.6)
(1.8)

(57)
(43)
(84)
(78)
(64)
(3.8)
(2.0)

74

(56)
(44)
(86)
(67)
(56)
(3.6)
(1.9)

128

KPNC
Total

Women (n, %)
Men (n, %)
Smoking (n, %)
Passive smoking (n, %)
Adolescent BMI (kg/m2, SD)
Sun exposure (n, %)
Age at disease onset
Age at inclusion in study
Total

No coffee consumption

4 Cups of coffee daily

<4 Cups of coffee daily

Cases

Controls

Cases

Controls

Cases

Controls

Cases

Controls

944 (81)
215 (19)
521 (45)
522 (45)
23.0 (4.4)
133 (11.5)
31.1 (9.8)
50.8 (8.2)
1159

955 (81)
217 (19)
416 (35)
500 (43)
22.3 (3.7)
121 (10.3)

49.8 (8.7)
1172

441
70
169
215
23.0
46

426
78
91
162
22.5
48

361
96
231
208
22.9
61

375
81
193
221
21.7
48

142
49
121
99
23.3
26

154
58
132
117
22.7
25

511

(86)
(14)
(33)
(42)
(4.7)
(9)

504

(85)
(15)
(18)
(32)
(3.9)
(10)

457

(79)
(21)
(51)
(46)
(4.0)
(14)

(82)
(18)
(42)
(48)
(3.0)
(11)

456

191

(73)
(27)
(62)
(55)
(4.4)
(12)

212

457

Multiple sclerosis

BMI, body mass index; EIMS, Epidemiological Investigation of Multiple Sclerosis; KPNC, Kaiser Permanente Medical Care Plan, Northern California Region; MS, multiple sclerosis; UVR, ultraviolet rays.

(74)
(26)
(63)
(52)
(4.2)
(14)

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

Table 1

EIMS
Coffee consumption at index

Coffee consumption 5 years prior to index

Coffee consumption 10 years prior to index

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

Number of cups, 1 cup=150 mL

Cases/controls*

OR

OR

Cases/controls*

OR

OR

Cases/controls*

OR

OR

0
12
34
56
7+
p for trend

348/577
523/867
479/897
196/319
74/128

1.0 (reference)
1.01 (0.85 to 1.20)
0.91 (0.76 to 1.09)
1.05 (0.84 to 1.33)
0.99 (0.72 to 1.36)
0.80

1.0 (reference)
0.97 (0.80 to 1.17)
0.82 (0.68 to 0.99)
0.86 (0.67 to 1.10)
0.70 (0.49 to 0.99)
0.01

372/623
484/804
393/755
178/275
65/119

1.0 (reference)
1.03 (0.86 to 1.23)
0.91 (0.75 to 1.10)
1.13 (0.89 to 1.44)
0.95 (0.68 to 1.33)
0.96

1.0 (reference)
1.04 (0.86 to 1.24)
0.84 (0.69 to 1.02)
0.94 (0.73 to 1.21)
0.72 (0.51 to 1.03)
0.05

384/607
389/738
326/561
136/225
52/86

1.0 (reference)
0.85 (0.71 to 1.02)
0.96 (0.78 to 1.18)
1.00 (0.77 to 1.31)
1.00 (0.68 to 1.45)
0.78

1.0 (reference)
0.80 (0.67 to 0.97)
0.84 (0.68 to 1.04)
0.80 (0.60 to 1.06)
0.71 (0.47 to 1.06)
0.06

KPNC
Coffee consumption at index

Coffee consumption 5 years prior to index

Number of cups, 1 cup=237 mL

Cases/controls*

OR

OR

ca/co*

OR

OR

0
13
4+
p for trend

511/504
457/456
191/212

1.0 (reference)
1.00 (0.83 to 1.20)
0.90 (0.71 to 1.14)
0.45

1.0 (reference)
1.00 (0.77 to 1.28)
0.69 (0.50 to 0.96)
0.05

598/601
395/385
166/186

1.0 (reference)
1.04 (0.86 to 1.27)
0.91 (0.71 to 1.17)
0.62

1.0 (reference)
0.98 (0.76 to 1.27)
0.64 (0.45 to 0.91)
0.04

EIMS: Data on coffee consumption habits at index were missing for 57 cases and 166 controls. Unconditional logistic regression has been used.
KPNC: Data on coffee consumption habits at index were missing for 4 cases and 6 controls. Conditional logistic regression has been used.
*Number of exposed cases and controls.
Adjusted for age, gender, residential area and ancestry.
Adjusted for age, gender, residential area, ancestry, smoking habits, exposure to passive smoking, sun exposure habits and body mass index at age 20 years.
Matched OR (sex, birth date, race/ethnicity and zip code of the case residence).
Adjusted for smoking habits, exposure to passive smoking, sun exposure habits, and body mass index at age 20 years.
EIMS, Epidemiological Investigation of Multiple Sclerosis; MS, multiple sclerosis.

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

Multiple sclerosis

458

Table 2 ORs with 95% CIs of developing MS for participants in different categories based on coffee consumption using logistic regression

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

Multiple sclerosis
quality of the reported information on habits of coffee consumption would probably not differ between cases and controls
due to different perceptions on potential effects from coffee
consumption, and misclassication would likely have been nondifferential, thus biasing towards the null (when comparing the
category of highest exposure with the unexposed). While a
potential selection bias may result from the relatively high proportion of non-responders among the controls who answered
the complementary questions in EIMS, this bias is probably
modest because lifestyle habits such as the prevalence of
smoking and the pattern of alcohol consumption among the
controls who responded were consistent with that expected for
the general population in similar ages.25 Furthermore, there
were no signicant differences in lifestyle habits (smoking,
passive smoking, BMI and sun exposure) between those who
answered the complementary questions and those who did not,
indicating that selection did not take place in this step. The possibility that the inverse relationship between coffee consumption
and risk of MS might be due to reverse causation seems less
likely since higher consumption of coffee was associated with
decreased odds of MS even several years prior to the index year,
as observed in EIMS. Further, there were no signicant changes
in habits of coffee consumption among cases or controls during
the decade preceding the index year in the Swedish data. The
correlation between coffee consumption at index and 5 years
prior to index was 0.9 ( p<0.0001), for cases and controls. The
corresponding correlation between coffee consumption and
10 years prior to index was 0.8 ( p<0.0001; see also table 1).
The same was assumed to be true in the KPNC cohort for this
investigation. However, the possibility remains that underreporting of coffee intake prior to onset by MS cases, occurred
and led to an overestimate of the strength of the association, a
risk that may be higher in the KPNC cohort due to the longer
disease duration at the time of the study. Given the delay
between diagnosis of MS and data collection in both data sets
and, for KPNC, the adoption of the highest ever exposure as
representing pre-MS exposure, the possibility of reverse causation cannot entirely be ruled out. Additional limitations in the
KPNC cohort are that the amount consumed was assumed to be
consistent over time and that coffee consumption was assumed
to have begun at the age the participants rst began consuming
caffeinated beverages. However, using a similar method with the
same assumptions as for evaluating coffee, there was no apparent evidence of an association between greater consumption of
tea or soda intake and reduced odds of MS. Finally, while we
carefully adjusted for putative confounders in both casecontrol
studies, the possibility of residual confounding remains, as is
true of any study.
The main analysis based on EIMS was performed by unconditional logistic regression (adjusted for the matching factors) in
order to maximise the number of controls that could be
included in the analysis and, thereby, increase the precision of
our effect estimates. The rationale for doing so is that cases that
did not full McDonald criteria at the time of this report were
excluded, but their corresponding controls were not. Further,
since information on coffee consumption was collected by complementary questions sent out in 2013, non-response made
some triplets without a case or without controls. However, we
also performed conditional analyses, and the estimated ORs
were in close agreement with those from the unconditional analysis (OR 0.73 in the conditional vs OR 0.70 in the unconditional analysis).
The study also has a number of strengths. The careful recruitment of cases and controls from the same reference population

in both cohorts mitigates concern that apparent associations are

due to anything other than case or control status. Further, our
analytical models were adjusted for a large number of potential
confounding factors, including currently established environmental risk factors for MS.
In conclusion, we observed a signicant association between
high consumption of coffee and decreased risk of developing
MS. Further studies are required to establish if it is in fact caffeine, or if there is another molecule in coffee underlying the
ndings, to longitudinally assess the association between consumption of coffee and disease activity in MS, and to evaluate
the mechanisms by which coffee may be acting, which could
thus lead to new therapeutic targets.
Acknowledgements The authors thank Farren Briggs, Allan Bernstein, Kalliope
Bellesis, Hong Quach, Gary Artim, Charlotte Anderberg, and Eva Johansson.
Contributors AKH, TO and LA were responsible for study concept and design of
the EIMS study. EMM, CAS and LFB were responsible for study concept and design
of the US study. LS assisted with data collection in the US study. AKH, MG and XS
performed the statistical analyses. AKH drafted the manuscript. All authors helped in
interpreting the results and critically revising the manuscript. LFB and CAS
supervised the US study, and TO and LA supervised the EIMS study. All authors
approved the nal version of the manuscript to be published.
Funding The work was supported by grants from the Swedish Research Council;
the Swedish Research Council for Health, Working Life and Welfare, Knut and Alice
Wallenberg foundation, the AFA foundation, the Swedish Brain foundation, the
Swedish Association for Persons with Neurological Disabilities, NIH/NINDS
K23067055, and NIH/NIEHS R01 ES017080, NIH/NINDS R01 NS079510; NIH/NIAID
R01 AI076544.
Competing interests TO received compensation for scientic advisory boards or
lectures for Biogen and Genzyme, unrestricted MS research grants from Biogen,
Bovartis, Genzyme, Allmiral and AstraZeneca, the Swedish Research Council
(07488), EU fp7 Neurinox, Knut and Alice Wallenberg Foundation, Margareta af
Ugglas Stiftelse, the Afa Foundation and the Swedish Brain Foundation. LA receives
research support from the Swedish Medical Research Council and Swedish Research
Council for Health, Working life and Welfare.
Patient consent Obtained.
Ethics approval The Regional Ethical Review Board at Karolinska Institutet, and
the Institutional Review Boards of the KP Division of Research and the University of
California, Berkeley.
Provenance and peer review Not commissioned; externally peer reviewed.
Open Access This is an Open Access article distributed in accordance with the
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which
permits others to distribute, remix, adapt, build upon this work non-commercially,
and license their derivative works on different terms, provided the original work is
properly cited and the use is non-commercial. See: http://creativecommons.org/
licenses/by-nc/4.0/

REFERENCES
1

2
3
4

OKeefe JH, Bhatti SK, Patil HR, et al. Effects of habitual coffee consumption on
cardiometabolic disease, cardiovascular health, and all-cause mortality. J Am Coll
Cardiol 2013;62:104351.
Larsson SC. Coffee, tea, and cocoa and risk of stroke. Stroke 2014;45:30914.
van Dam RM, Hu FB. Coffee consumption and risk of type 2 diabetes: a systematic
review. JAMA 2005;294:97104.
Crippa A, Discacciati A, Larsson S, et al. Coffee consumption and mortality from all
causes, cardiovascular disease, and cancer: a dose-response meta-analysis.
Am J Epidemiol 2014;180:76375.
Lofteld E, Freedman ND, Graubard BI, et al. Association of coffee consumption
with overall and cause specic mortality in a large U.S. prospective cohort.
Am J Epidemiol 2015;182:101022.
Chen X, Ghribi O, Geiger JD. Caffeine protects against disruptions of the
blood-brain barrier in animal models of Alzheimers and Parkinsons disease.
J Alzheimers Dis 2010;20(Suppl 1):S12741.
Tsutsui S, Schnermann J, Noorbakhsh F, et al. A1 adenosine receptor upregulation
and activation attenuates neuroinammation and demyelination in a model of
multiple sclerosis. J Neurosci 2004;24:15219.
Chen GQ, Chen YY, Wang XS, et al. Chronic caffeine treatment attenuates
experimental autoimmune encephalomyelitis induced by guinea pig spinal cord
homogenates in Wistar rats. Brain Res 2010;1309:11625.

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

459

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

Multiple sclerosis
9

10
11
12
13
14
15
16
17

460

Mills JH, Thompson LF, Mueller C, et al. CD73 is required for efcient entry of
lymphocytes into the central nervous system during experimental autoimmune
encephalomyelitis. Proc Natl Acad Sci USA 2008;105:932530.
Tola MR, Granieri E, Malag S, et al. Dietary habits and multiple sclerosis.
A retrospective study in Ferrara, Italy. Acta Neurol 1994;16:18997.
Pekmezovic T, Drulovic J, Milenkovic M, et al. Lifestyle factors and multiple sclerosis:
a case-control study in Belgrade. Neuroepidemiology 2006;27:21216.
Jahromi SR, Toghae M, Jahromi MJ, et al. Dietary pattern and risk of multiple
sclerosis. Iran J Neurol 2012;11:4753.
Massa J, OReilly EJ, Munger KL, et al. Caffeine and alcohol intakes have no
associations with risk of multiple sclerosis. Mult Scler 2013;19:538.
Thompson AJ, Montalban X, Barkhof F, et al. Diagnostic criteria for primary
progressive multiple sclerosis: a position paper. Ann Neurol 2000;47:8315.
Briggs F, Acuna B, Shen L, et al. Adverse socioeconomic position during the life course
is associated with multiple sclerosis. J Epidemiol Community Health 2014;68:6229.
Hedstrm AK, Hillert J, Olsson T, et al. Smoking and multiple sclerosis susceptibility.
Eur J Epidemiol 2013;28:86774.
Hedstrm AK, Brnhielm M, Olsson T, et al. Exposure to environmental tobacco
smoke is associated with increased risk for multiple sclerosis. Mult Scler
2011;17:78893.

18

19
20

21

22
23
24

25

Brnhielm M, Hedstrm AK, Kockum I, et al. Sunlight is associated with decreased

multiple sclerosis risk: no interaction with human leukocyte antigen-DRB1*15.
Eur J Neurol 2012;19:95562.
Munger KL, Chitnis T, Ascherio A. Body size and risk of MS in two cohorts of US
women. Neurology 2009;73:154350.
Hedstrm AK, Olsson T, Alfredsson L. High body mass index before age 20 is
associated with increased risk for multiple sclerosis in both men and women. Mult
Scler 2012;18:13346.
Gianfrancesco MA, Acuna B, Shen L, et al. Obesity during childhood and
adolescence increases susceptibility to multiple sclerosis after accounting for
established genetic and environmental risk factors. Obes Res Clin Pract 2014;8:
e43547.
Liu R, Guo X, Park Y, et al. Caffeine intake, smoking, and risk of Parkinson disease
in men and women. Am J Epidemiol 2012;175:12007.
Qi H, Li S. Dose-response meta-analysis on coffee, tea and caffeine consumption
with risk of Parkinsons disease. Geriatr Gerontol Int 2014;14:4309.
Dhooghe MB, Haentjens P, Nagels G, et al. Alcohol, coffee, sh, smoking
and disease progression in multiple sclerosis. Eur J Neurol 2012;19:
61624.
Internet-based information. Http://www.scb.se (accessed 2 Mar 2014).

Hedstrm AK, et al. J Neurol Neurosurg Psychiatry 2016;87:454460. doi:10.1136/jnnp-2015-312176

Downloaded from http://jnnp.bmj.com/ on December 15, 2016 - Published by group.bmj.com

High consumption of coffee is associated

with decreased multiple sclerosis risk;
results from two independent studies
A K Hedstrm, E M Mowry, M A Gianfrancesco, X Shao, C A Schaefer, L
Shen, T Olsson, L F Barcellos and L Alfredsson
J Neurol Neurosurg Psychiatry 2016 87: 454-460 originally published
online March 3, 2016

doi: 10.1136/jnnp-2015-312176
Updated information and services can be found at:
http://jnnp.bmj.com/content/87/5/454

These include:

Supplementary Supplementary material can be found at:

Material http://jnnp.bmj.com/content/suppl/2016/03/03/jnnp-2015-312176.DC1.
html

References

This article cites 24 articles, 10 of which you can access for free at:
http://jnnp.bmj.com/content/87/5/454#BIBL

Open Access

This is an Open Access article distributed in accordance with the Creative

Commons Attribution Non Commercial (CC BY-NC 4.0) license, which
permits others to distribute, remix, adapt, build upon this work
non-commercially, and license their derivative works on different terms,
provided the original work is properly cited and the use is
non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/

Email alerting
service

Receive free email alerts when new articles cite this article. Sign up in the
box at the top right corner of the online article.

Topic
Collections

Articles on similar topics can be found in the following collections

JNNP Patients' choice (119)
Open access (219)
Press releases (33)
Immunology (including allergy) (1943)
Multiple sclerosis (934)
Stroke (1449)

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions
To order reprints go to:
http://journals.bmj.com/cgi/reprintform
To subscribe to BMJ go to:
http://group.bmj.com/subscribe/