Content uploaded by Simón Cárdenas-Robledo
Author content
All content in this area was uploaded by Simón Cárdenas-Robledo on Aug 11, 2021
Content may be subject to copyright.
Eur J Neurol. 2021;00:1–11. wileyonlinelibrary.com/journal/ene
|
1© 2021 European Academy of Neurology
Received: 5 July 2021
|
Accepted: 19 July 2021
DOI : 10.1111/ene.15 03 0
ORIGINAL ARTICLE
Multiple sclerosis is associated with higher comorbidity and
health care resource use: A population- based, case– control
study in a western Mediterranean region
Simón Cárdenas- Robledo1,2 | Susana Otero- Romero1,3 |
Maria Angels Passarell- Bacardit4 | Pere Carbonell- Mirabent1 | Jaume Sastre- Garriga1 |
Xavier Montalban1 | Mar Tintoré1
1Neurology- Neuroimmunology Service,
Multiple Scler osis Center of Cat alonia
(Cemcat), Hospital Uni versit ari Vall
d'Hebron, Barcelona, Spain
2Depar tment of Neurology, Multiple
Sclerosis Center (CEMHUN), Hospit al
Universitario Nacional de Colombia,
Bogotá, Colombia
3Preventive Medicine and Epidemiology
Service, Hospital Universi tari Vall
d'Hebron, Barcelona, Spain
4Atención Primar ia / IDIAP Jordi Gol
Primar y Care Research Institute, Institu t
Catal á de la Salut, Cat alunya Central,
Barcelona, Spain
Correspondence
Susana Otero- Romero, Preventive
Medicine and Epidemiology Service,
and Neurology- Neuroimmunology
Service, Multiple Sc lerosis Center of
Catal onia, Hospita l Univer sitar i Vall
d'Hebron, Pas seig de la Val l d'Heb ron,
119- 129, Barcelona 08035, Spain.
Email: sotero@cem-cat.org
Funding information
The database for this stu dy was provided
by the Institut Un iversitari d’Investigació
en Atenció Primàr ia (IDIA P) Jordi G ol
through the grant: 4a Co nvocatò ria d'Ajut
SIDIAP, 2014.
Abstract
Background and purpose: Comorbidities are common in multiple sclerosis (MS), and have
been associated with worse outcomes and increased health care resource usage. We
studied the frequency of comorbidities and adverse health behaviors (AHBs) in MS pa-
tients in the Mediterranean region of Catalonia.
Methods: This population- based, case– control study used primary health care infor-
mation covering 80% of Catalonia's population. Cases were matched by age/sex with
randomly chosen controls (ratio = 1:5). Demographic information, comorbidities, AHBs,
annual visits, sick leave days, and medication dispensing were studied. The association of
comorbidities with MS and the profile of comorbidities according to sex within MS cases
were assessed with multivariate logistic regression models, after adjusting for confound-
ing variables. Health care resource usage was analyzed in MS cases compared to controls,
and within MS cases in those with compared to those without comorbidities.
Results: Five thousand five hundred forty- eight MS cases and 27,710 controls (70% fe-
male, mean age = 48.3 years) were included. Stroke (odds ratio [OR] = 1.54, 95% con-
fidence interval [CI] = 1.17– 1.99), epilepsy (OR = 2.46, 95% CI = 1.94– 3.10), bipolar
disorder (OR = 1.67, 95% CI = 1.17– 2.36), and depression (OR = 1.83, 95% CI = 1.70–
1.98) were more frequent in MS. Cases were more prone to smoking but less to alcohol
intake. Among cases, psychiatric comorbidities were more frequent in women, whereas
cardiovascular diseases and AHBs were more frequent in men. MS patients, particularly
with comorbidities, had higher health care resource usage than controls.
Conclusions: Psychiatric comorbidities, stroke, epilepsy, and AHBs are more common
in MS patients than in the general population in the western Mediterranean region of
Catalonia. The presence of comorbidities increases the health care resource usage in MS
patients.
KEYWORDS
comorbidity, health behavior, health services, multiple sclerosis, social class
2
|
CÁRDENAS- RO BLEDO Et AL.
INTRODUCTION
Com orbidit y can be defined as the occu rrence of any additio nal clin-
ical entity occurring during the clinical course of a patient with the
index disease under study [1]. The study of comorbidity in multi-
ple sclerosis (MS) patient s has attracted much at tention in the past
decade due to its high frequency and impact on the course of the
disease. Comorbidities may modify the clinical course of MS by in-
creasing the relapse rate [2] and the risk of disability progression
[3,4]. They also have implications for treatment choice [5,6], rate of
health care utilization [7], and quality of life [8]. Therefore, correctly
identif ying and treating coexisting diseases in MS patients can im-
prove patient care and outcomes. Most patients with MS have at
least one comorbid condition [9]. The frequency of comorbidities
increases with age and lower socioeconomic status, and has been
consistently found to be higher in MS patients than in control sub-
jects [9,10].
The most frequent comorbidities in MS are psychiatric disorders
and common cardiovascular disease risk factors, such as hyperten-
sion and hyperlipidemia [11]. There is wide variation in the burden
of chronic diseases across different populations, and information on
the geographical pattern of comorbidities in MS patient s is lacking,
as most studies addressing this issue have been performed in North
America [4,12,13] and Northern Europe [14– 17].
Studies on comorbidity in MS patients from Southern Europe,
especially Spain, are scarce [18] and have focused on particular
diseases [19– 23]. Moreover, classic definitions of comorbidity do
not include health behaviors, such as smoking and alcohol intake.
However, these behaviors can affect disease activity and should be
considered as a type of comorbidity [24]. Here, we examined the
frequency and pattern of a wide range of comorbidities, including
adverse health behaviors, in MS patients compared with the general
population in the Mediterranean region of Catalonia. We also as-
sessed the effect of MS comorbidities on health care resource use.
METHODS
Population
This is a cross- sectional population- based case– control study of pa-
tients treated within the primary health care net work in Catalonia
(Spain). The study was fully approved by the ethics committee of the
Foundation University Institute for Primar y Health Care Research
Jordi Gol. The Catalan Institute of Health is the main health care
provider in Catalonia and manages 279 (77.9%) of 358 primary care
centers (PCCs) in the region. Each PCC is constituted of at least
three basic care units (BCUs), each comprising one primary care
physician (PCP) and one nurse, who are assigned a common group
of patients. The population assigned to the PCCs is >8 0% of the
population of Catalonia (nearly 6 million people, and approximately
15% of the Spanish population). All the primar y care health care staff
(>10,00 0) use the same electronic health record (EHR), which has
been universal since 2005. Health care information recorded in in-
dividual subjects’ EHR from these PCCs is automatically transferred
to the primar y care research database SIDIAP [25], which contains
longitudinal and anonymized dat a in relation to demographic varia-
bles (date of birth, sex, nationalit y, PCC), laboratory tests (every test
performed in the PCCs since 2005), ambulatory medication (both
prescriptions and dispensed medication, health care professional
and PCC responsible fo r th e pr es cription), and healt h prob le ms codi-
fied according to the International Classification of Diseases, 10th
revision (ICD- 10), with date of diagnosis, as well as clinical variables
such as blood pressure, weight, and smoking, among others [25].
Data quality in the SIDIAP database is monitored at the BCU level by
means of a quality score, developed and validated by comparing the
observed and expected frequencies of common health problems in
BCUs assigned with more than 500 subjects [26]. Only the popula-
tion assigned to the BCUs in the highest quintile of registry qual-
ity score (approximately 2 million, 39% of the SIDIAP population) is
eligible when extracting data for clinical research; this is called the
SIDIAPQ [25]. Thus, the data available for this study are considered
of high quality, and the sample included in the SIDIAPQ [25] is con-
sidered to be highly representative of the population of Catalonia in
terms of geographic, age, and sex distribution [26].
Inclusion criteria and study sample
We included patients recorded within the SIDIAP database who had
visited a primary care practice at least once between 1 Januar y 2006
and 5 October 2016. Cases were defined as patient s >18 years old
with confirmed diagnosis of MS identified by the associated ICD-
10 code G35. The control group included subjects >18 years old,
who did not have an ICD- 10 code of any demyelinating disease (G36,
G360, G368, G369, G37, G373, G378, or G379). Control subjects
were chosen randomly and were matched by age, sex, PCC, and
socioeconomic status, with a control:case ratio of 5. Subjects were
excluded when basic demographic data required for matching were
unavailable.
Variables
Basic demographic data (age, sex, place of residence, socioeconomic
status) were drawn for every subject. For MS patients, disease du-
ration was estimated based on the date of the appearance of the
ICD code for MS. Socioeconomic status was determined by using
the MEDEA index [27], which summarizes five socioeconomic indi-
cators (the proportions of those aged at least 16 years who [i] are
unemployed, [ii] are employed as manual workers, [iii] are employed
as temporary workers, [iv] are illiterate or have not completed pri-
mary education, and [v] are younger than 29 years with insufficient
education) found in every cens us section. Accor ding to thes e indic a-
tors in their place of residence, subjects are classified in quintiles,
the first quintile (Q1) being the least deprived and the last quintile
|
3
COMORBIDITIES AND MULTIPLE SCLEROSIS
(Q5) the most deprived. This classification is done automatically ac-
cording to the subject's address. For the purpose of our analysis,
this was the last classification registered at the time of the data ex-
traction. The choice of comorbidities was based on the frequency
in the general population and that reported in MS patients, as well
as those included in common comorbidity scales; comorbidities
were obtained by identifying the ICD- 10 codes associated with each
subject (Table S1). For general groups of disease, we chose the ICD
code highest in the hierarchy, aiming to capture all the subcategories
of that code (e.g., E78 for disorders of lipoprotein metabolism and
other lipidemias, which covers 20 specific types of dyslipidemia).
Comorbidities were analyzed by the presence of at least one, at least
two, or more than two diagnoses. We also analyzed individual co-
morbidities and comorbidities by diagnosis groups (neurologic, psy-
chiatric, cardiovascular, cancer, respiratory, autoimmune). Following
recommendations of the International Workshop on Comorbidity in
Multiple Sclerosis [24] we included adverse health behaviors as part
of our analysis. For this, we identified the ICD- 10 codes related to
smoking and alcohol intake associated with each subject (Table S1),
which are generated when the patients report some form of alcohol
or tobacco consumption. Alcohol intake was defined as any form of
consumption, and smokers were identified by having smoked at any
time. In patients with available data on height and weight, body mass
index (BMI) was calculated to evaluate overweight/obesity. This was
obtained from the last measurements reported in the database.
Overweight was defined as BMI between 25 and 30, Grade I obesity
as BMI between 30 and 35, and Grade II obesity as BMI > 35.
Health care resource use was determined by analyzing the fol-
lowing variables: yearly number of visits to (i) PCPs and (ii) nurses,
(iii) yearly sick leave days (in those patients not on disabilit y pension
or retired), and (iv) yearly overall and symptomatic medication dis-
pensations (antidepressants, bladder, pain, and fatigue medications).
Finally, in patients with hypertension or diabetes, we assessed
whether a satisfactor y control of those diseases was achieved, ac-
cording to the recommendations of local guidelines as follows:
(i) For hyper tension: at least one annual visit and at least one annual
blood draw comprising serum glucose, creatinine, ureic nitrogen,
electrolytes, glomerular filtration rate, and one urinalysis.
(ii) For diabetes: at least one annual visit, biannual blood draw with
serum glucose and A1c glyc ated hemoglobin, and one annual
blood draw with triglycerides, total, high- density lipoprotein, and
low- density lipoprotein cholesterol, and creatinine and glomeru-
lar filtration rate.
Statistical analysis
Comorbidity profile in cases and controls
We compared the presence of comorbidities and adverse health
behaviors (presence of overweight/obesit y, smoking, and alcohol
intake) by means of univariate binomial logistic regression, built
to calculate odds ratios (ORs) with 95% confidence intervals (CIs).
To control for the effect of confounding variables and the effect
of clustering of comorbidities in the risk of individual diseases, we
built multivariate logistic regression models for each comorbidity
for which a significant risk was found in the univariate models, in-
cluding the other significant comorbidities as covariates. For these
models, we also included age, sex, and socioeconomic status, taking
into account that these variables were likely unbalanced between
those with and without specific comorbidities and their confounding
effect was likely not accounted for by the sampling procedure. To
further assess the influence of sex and age on the risk of comor-
bidities driven by MS, univariate logistic regressions were built after
stratification by sex and age groups. Health care resource use was
compared between both groups by calculating effect size using
Cohen d test.
Comorbidity profile within MS cases
We then analyzed the profile of comorbidities and adverse health
behaviors among MS cases. For this, we used the same approach as
above, first building univariate logistic regression models and then
adjusting for age, sex, disease duration, and socioeconomic status as
well as the comorbidities found significant in the univariate analysis.
OR s and 95% CI s we re ca lcula te d fro m th ese mode ls . Hea lth care re-
source use variables were compared in MS cases with comorbidities
against those without comorbidities, following the same procedure
described above.
Variables were described using propor tions, central tendency
(mean or median, according to normality evaluation), and dispersion
(SD or interquartile range, respectively) measures. Skewness, kurto-
sis, and Kolmogorov– Smirnov test s were used to assess normality
of continuous variables. Descriptive analyses were performed using
Pearson χ2 or Fish er exac t test for categori cal variable s, and St udent
t or Mann– Whitney U test for continuous variables as appropriate.
Statistical significance was set at 0.05. To correct for the likelihood
of false- positive results given the high number of comparisons and
hypothesis tests possible due to the number of comorbidities stud-
ied, we calculated the overall maximal false discover y rate (m- FDR;
i.e., the rate of false- positive results when referring to all p- values
≤ 0.05 as significant) and q- values (i.e., the expected proportion of
fal se- positive resu lt s among res ults as or mo re ext reme than the ob-
served one) for each hypothesis test.
Statistical analysis was performed using R version 4.0.2, with the
packages qvalue, sjPlot, and finalfit.
RESULTS
The selection process yielded a study sample of 5548 MS cases and
27,740 control subjects. Mean (SD) disease duration in MS cases
was 12.0 (7.6) years and did not differ between men and women
4
|
CÁRDENAS- RO BLEDO Et AL.
(p = 0.192). MS patients with comorbidities had slightly longer
mean disease duration than those without comorbidities (12.55 vs.
11.24 years, p < 0.001, Cohen d = 0.17). The sample subset with
available data on height and weight comprised 3022 (54.5%) cases
and 15,870 (57.2%) controls. Demographic characteristic s are de-
scribed in Table 1.
Comorbidity profile in cases and controls
The presence of at least one comorbidity was slightly more frequent
in MS cases than in controls (60.0% vs. 56.8%, p < 0.001). After ad-
justing for sex, age, and socioeconomic status, we found the risk of at
least one comorbidity to be higher in MS cases (OR = 1.19, 95% CI =
1.11– 1.28, p < 0.0 01), whereas there was no significant difference
for two and three or more comorbidities (OR = 1.08, 95% CI = 0.9 9–
1.162, p = 0.064 and OR = 1.04, 95% CI = 0 . 9 5 – 1 . 1 3 , p = 0.371,
respectively). In the multivariate models, the risk of comorbidities
in general was significantly associated with increasing socioeco-
nomic deprivation, and especially with increasing age (Figure 1a).
Neurologic (OR = 1.24, 95% CI = 1.08– 1.43, p = 0.0 03) and psychi-
atric (OR = 1.43, 95% CI = 1.38– 1.52, p < 0.001) comorbidities were
significantly more frequent in MS cases than in controls.
When exploring individual comorbidities, we found stroke, epilepsy,
bipo lar disorder, and major depressive disorder to be signific antly more
frequent in MS, whereas hypertension, hyperlipidemia, and chronic
kidney dise as e were mo re frequ ent in co nt ro ls (Table 2). The findings of
the univariate logistic regression for anemia and diabetes became non-
significant after adjustment for confounding variables. Detailed results
of uni- and multivariate analyses are found in Table S2a.
We also found a slightly increased risk of ever smoking and current
smoking in MS cases compared to controls, as well as of alcohol intake
(Table 2). In contrast, we did not find an increased risk of overweight/
obesity (OR = 1.06, 95% CI = 0. 52– 2 . 20, p = 0.869) in MS patients.
The risk of comorbidities driven by MS was found to differ ac-
cording to age groups, being higher in cases aged <40 years (OR =
1.38, 95% CI = 1.23– 1.54, p < 0.001) and 40– 59 years (OR = 1.23,
95% CI = 1.14– 1.34, p < 0.0 01), but was lower in those between 60
and 79 years old (OR = 0.65, 95% CI = 0.55– 0.76, p < 0.001) and
80 years and older (OR = 0.27, 95% CI = 0.12– 0.63, p = 0.001). The
risk of at least two and at least three comorbidities according to age
showed a similar pattern (see Figure 2 and Table S2c). After strat-
ifying according to sex, the risk profile of comorbidities remained
unchanged, except for the associations of anemia and diabetes with
MS, which were not evident in females but more pronounced in
males (the frequencies and ORs of all the individual comorbidities
and adverse health behaviors are presented in Table S2b).
A satisfactory control of hypertension and diabetes was more
frequent in MS patients (hypertension: unadjusted OR = 1. 22, 95%
CI = 1.09– 1.36, p < 0.0 01; diabetes: unadjusted OR = 1.29, 95% CI
= 1.01– 1.65, p = 0.039). However, after adjusting for sex, age, and
socioeconomic status and adequate control of the other comorbid-
ity, this association was lost for hypertension (OR = 0.83, 95% CI =
0. 62– 1 . 11, p = 0.209) and remained marginally significant for diabe-
tes (OR = 1.38, 95% CI = 1.00– 1.897, p = 0.049).
We found statistically significant differences in all the variables
of health care resource use, with more PCP and nurse visit s, as well
as sick leave days, in MS cases as compared with controls (Table 4).
Medication dispensation was higher overall for MS patients com-
pared with controls. The effec t sizes, estimated by Cohen d test,
were nevertheless small.
Comorbidity profile within MS cases
The overall risk of comorbidities among MS patients was strongly
associated with increasing age and to a lesser extent with the female
sex in the multivariate model (Figure 1b).
The pattern of comorbidities dif fered between sexes, with car-
diovascular diseases (hypertension, hyperlipidemia, and chronic
ischemic cardiomyopathy) found more frequently in male patients,
and depression, anxiety, and anemia found more frequently in fe-
male patients (Table 3 and Table S3). Overweight/obesity was not
significantly associated with sex in the univariate analysis. In the
univariate models, both current and ever smoking status was more
frequent in male patients, but after adjusting we found an increased
TAB LE 1 Demographic characteristics of the c ase and control
groups
Characteristic MS, n = 554 8
Controls,
n = 27, 740
Age, years, mean (SD) 48.33 (12.79) 48.34 (12.79)
Age, years, n (%)
<40 1499 (27.0) 7494 (27.0)
4 0 – 5 9 2918 (52.6) 14,582 (52.6)
6 0 – 7 9 1062 (19.1) 5319 (19.1)
≥80 69 (1.2) 345 (1.2)
Sex, female n (%) 3863 (69.6) 19,31 5 (69. 6)
Female:male ratio 2.29 2.29
MEDEA index, n (%)a
Q1b664 (15.6) 3320 (15.6)
Q2 1024 (24.1) 5120 (24.1)
Q3 1045 (24.6) 5225 (24.6)
Q4 885 (20.8) 4425 (20. 8)
Q5 630 (14.8) 3150 (14.8)
Province of residence, n (%)
Barcelona 4197 (75.6) 20,940 (75.5)
Girona 432 (7.8) 2177 (7.8)
Lleida 417 (7.5) 2041 (7.4)
Tarragona 461 (8.3) 2269 (8.2)
Unknown 41 (0.7) 313 (1.1)
Abbreviation: MS, multiple sclerosis.
aSocioeconomic status index MEDEA.[27]
bQ1 is the quintile with the least socioeconomic deprivation.
|
5
COMORBIDITIES AND MULTIPLE SCLEROSIS
risk for current smoking in female patients. We found a significantly
higher risk of alcohol intake in female MS patients compared to males
(Table 3 and Table S3). The increased risk of comorbidities associated
with age was also observed for some individual comorbidities among
MS patients. This was true for cardiovascular diseases (hyperten-
sion, diabetes, dyslipidemia, ischemic cardiomyopathy, stroke),
depression, chronic kidney disease, and breast cancer. Anxiety, mi-
graine, and asthma never theless showed the opposite behavior (see
FIGURE 1 Overall risk of comorbidities
in multivariate models. Forest plots
show odds ratios (ORs) of at least one
(green), at least two (red), and at least
three (blue) comorbidities in multiple
sclerosis (MS) compared to controls (a)
and among MS cases (b), after adjusting
for, sex, age, disease duration (only for
MS cases), and socioeconomic status. OR
(points) and 95% confidence intervals
(lines) are depicted. Estimates considered
nonsignificant ( p > 0.05) are depicted
as empty circles. For age, the younger
group (<40 years) is used as reference.
For socioeconomic status, quintiles are
organized with increasing deprivation,
with the least deprived (Q1) as reference
Comorbidity
Univariate Multivariate
OR (95% CI) pOR (95% CI) p
Hyperlipidemia 0.91 (0.85– 0.98) 0.012 0.89 (0.82– 0.96) 0.002
Diabetes 0.85 (0.75– 0.96) 0.009 0.90 (0.79– 1.02) 0.090
Hypertension 0.89 (0.82– 0.96) 0.003 0.89 (0.81– 0.97) 0.010
Stroke 1.51 (1.17– 1.94) 0.001 1.54 (1.17– 1.99) 0.001
Bipolar disorder 1.88 (1.32– 2.63) <0.001 1.67 (1.17– 2.3 6) 0.004
Major depressive disorder 1.79 (1.66– 1.93) <0.001 1.83 (1.70– 1.98) <0.001
Epilepsy 2.57 (2.03– 3.23) <0.001 2.46 (1 .94– 3.10) <0.001
Anemia 1.16 (1.02– 1.31) 0.021 1.12 (0.98– 1.26) 0.083
Chronic kidney disease 0.57 (0.40– 0.78) <0.001 0.60 (0.42– 0.84) 0.004
Current smoker 1.18 (1.10– 1.26) <0.0 01 1.10 (1.02– 1.19) 0.017
Ever smoker 1.27 (1.19– 1.35) <0.001 1.23 (1.15– 1.33) <0.001
Alcohol consumption 1.63 (1.48– 1.81) <0.001 1.66 (1.47– 1.87) <0.0 01
Note: Results are from univariate and multivariate logistic regression models after adjusting for sex,
age, and socioeconomic st atus, as well as for the increased risk of the other comorbidities. Only
comorbidities and adver se health behaviors with significant associations with MS in the univariate
models are shown. For details of the remaining comorbidities see Table S2a.
Abbreviations: CI, confidence interval; OR, odds ratio.
TAB LE 2 Risk of individual
comorbidities in multiple sclerosis patients
compared to controls
6
|
CÁRDENAS- RO BLEDO Et AL.
Figure S1 for details). MS patients in the two most socioeconom-
ically deprived quintiles were at higher risk of having at least one
comorbidity (Q4: OR = 1.25, 95% CI = 1.02– 1.54, p = 0.031; Q5: OR
= 1.46, 95% CI = 1.16– 1.82, p = 0.001) compared with those in the
least deprived quintile, after adjusting for sex, age, and disease du-
ration. Additionally, MS patients in the most deprived quintile were
significantly more likely to be ever smokers compared with the least
deprived quintile (OR = 1.2 9, 95% CI = 1.02– 1.64, p = 0.03 5). We di d
not find significant associations bet ween disease duration and the
presence of comorbidities or adverse health behaviors, af ter adjust-
ing for the remaining variables.
We found a significant but small excess of nurse and PCP vis-
its, as well as increased medication dispensations, especially of an-
tidepressants and pain medication, in MS cases with comorbidities
compared with those without comorbidities. However, there were
no differences in the annual number of sick leave days ( Table 4).
Assessment of FDR
The m- FDR was estimated in 3.41%, and we found no q- value > 0.05
among the hypothesis tests considered significant.
DISCUSSION
This large population- based, case– control study furthers knowledge
on the frequency and pattern of comorbidities in MS, and the impact
they have on the use of health care resources in the Mediterranean
region, where evidence is scarce [19– 23]. Our results are consist-
ent with those published for other geographical areas [18], showing
that MS patients have a higher risk of comorbidities than the general
population and specifically, a significantly higher risk of epilepsy,
stroke, and psychiatric disorders. Additionally, we found that MS
patients use more health care resources [28], especially those with
comorbidities [7].
More than half of the MS patients had at least one additional
condition, which is important for their overall prognosis and care
[29,30]. The excess in comorbidities in MS compared with the gen-
eral population is seen in young age groups but is not present in those
>60 years old. This was surprising, given the well- established rela-
tion of comorbidities with older age [31], but it could be a reflection
of the attention older patients receive, which is focused on advanced
MS- related issues (such as increasing disability, pain, and sphincter
dysfunction), and comorbidities being somewhat neglected. It is also
possible that comorbidities are correctly identified and even treated
outside the primary health care network and therefore not identi-
fied as such in the study database. Other explanations are that older
and disabled patients may have had more infrequent primary care
follow- up, thus lowering the likelihood of comorbidity detec tion, or
that MS patients with comorbidities are less likely to reach an ad-
vanced age given their higher mort ality risk [32], and thus only the
healthier MS population was t aken into account in our study.
Consistent with previous evidence [33,34], we found a higher
risk of stroke in MS cases than in controls. This is a recognized as-
sociation, with the caveat that there is risk of miscoding given the
overlap between symptoms in stroke and early MS. Our result s re-
garding depression are also in line with those previously reported.
Comorbidity
Univariate OR
(95% CI) p
Multivariate OR
(95% CI) p
Hyperlipidemia 0.77 (0.67– 0.88) <0.001 0.79 (0.66– 0.93) 0.006
Diabetes 0.76 (0.61– 0.97) 0.023 0.87 (0.66– 1.63) 0. 35
Hypertension 0.80 (0.69– 0.93) 0.003 0.76 (0.62– 0.92) 0.037
Chronic ischemic heart disease 0.43 (0. 26– 0.72) 0.001 0.40 (0.21– 0.75) 0.009
Stroke 0.59 (0.37– 0.93) 0.022 0.58 (0.33– 1.02) 0.090
Peptic ulcer 0.34 (0.22– 0.53) <0.001 0.35 (0.21– 0.58) <0.001
Major depressive disorder 1.76 (1.51– 2.07) <0.001 1.56 (1.29– 1.89) <0.001
Anxiety disorder 1.96 (1.67– 2.31) <0.001 1.72 (1.41– 2.09) <0.001
Anemia 5.97 (4.02– 9.33) <0.001 5.75 (3.68– 9.50) <0.001
Asthma 1.45 (1.06– 2.02) 0.023 1.52 (1.05– 2.23) 0.041
Chronic bronchitis 0.41 (0.26– 0.63) <0.001 0.44 (0.26– 0.74) <0.001
Colon cancer 0.34 (0.12– 0.91) 0.031 0.25 (0.07– 0.79) 0.024
Current smoker 0.77 (0.68– 0.88) <0.001 1.43 (1.09– 1.89) 0.011
Ever smoker 0.52 (0.46– 0.59) <0.001 0.41 (0.32– 0.52) <0.001
Alcohol consumption 2.82 (2.33– 3.42) <0.001 2.68 (2.14– 3.37) <0.001
Note: Results are from univariate and multivariate logistic regression models after adjusting for age,
disease duration, and socioeconomic st atus, as well as for the presence of the other comorbidities.
Only comorbidities and adverse health behaviors with significant associations with sex in the
univariate models are shown. For det ails of the remaining comorbidities see Table S3.
Abbreviations: CI, confidence interval; MS, multiple sclerosis; OR, odds ratio.
TAB LE 3 Risk of individual
comorbidities in female MS patients
compared to male MS patients
|
7
COMORBIDITIES AND MULTIPLE SCLEROSIS
Depression is one of the comorbidities most frequently associated
with MS [35– 38], and it s impact in terms of disability [39] and qual-
ity of life [40– 42] is well recognized. In addition, we found that the
main difference in medication- dispensing between MS patients with
or without comorbidities was in the dispensing of antidepressants.
It has been shown that a significant proportion of MS patients use
at least one antidepressant regularly [43,44]. The higher frequency
of epilepsy in MS patients has also been established in several pop-
ulations [22,45– 47], and is associated with a worse prognosis and
mortality [45]. Due to the nature of our study, we were unable to
establish which disease preceded the other, and whether the pres-
ence of epilepsy was associated with increased disability. Consistent
FIGURE 2 The effect of age in the
risk of comorbidities driven by multiple
sclerosis (MS). Forest plot shows adjusted
odds ratios (ORs) of at least one (green),
at least two (red), and at least three
(blue) comorbidities in MS compared to
controls af ter stratifying by age group.
OR estimates (points) and 95% confidence
intervals (lines) are depicted. For detailed
data, see Table S2c
TAB LE 4 Health care resource use by patients with MS versus controls, and within MS cases, those with and without comorbidities
Whole sample MS cases
MS, mean
(SD)
Controls,
mean (SD) paEffect sizeb
With
comorbidities,
mean (SD)
Without
comorbidities,
mean (SD) paEffect sizeb
Yearly sick leave
days
11.0 (46.8) 6.7 (32.6) <0.001 0.106 10.6 (46.2) 11.62 (47.6) 0.426 0.022
Yearly nurse visits 5.0 (5.3) 3.8 (4.5) <0.001 0.24 4 3.57 (9.8) 2.23 (7.6) <0.001 0.15 3
Yearly PCP visits 3.0 (9.0) 1.7 (4.3) <0.001 0.18 4 5.77 (5.5) 3.94 (4.7) <0.0 01 0.358
Overall medication
dispensations
466.2 (911.3) 184. 5 (511.5) <0.001 0.381 596.0 (1034.3) 270.6 (637.6) <0.001 0.379
Bladder 11.5 (62.1) 1.8 (21.7) <0.0 01 0.208 12. 5 (59.4) 9.1 (65.4) 0.045 0.054
Fatigue 5.3 (61.3) 0.2 (77.8) <0.001 0.073 6.2 (68.4) 3.8 (48.6) 0.159 0.040
Depression 19.8 (60) 8.5 (36.8) <0.001 0.227 27.5 (59.3) 8.1 (31.5) <0.001 0.358
Pain 56.8 (226.6) 12.5 (95.6) <0.001 0.255 70.9 (242.9) 35.5 (197.6) <0.0 01 0.159
Abbreviations: MS, multiple sclerosis; PCP, primary care physician.
aStudent t- t e s t .
bCohen d- t e s t .
8
|
CÁRDENAS- RO BLEDO Et AL.
with our result s, bipolar disorder has previously been found to be
associated with MS [48– 50]. This association is highly relevant, be-
cause of the impact that the coexistence of both diseases has on
quality of life [51,52], and the challenges that bipolar disorder poses
for the treatment of chronic conditions [53,54]. Another interesting
finding is the lack of association between MS and anemia, contrary
to what has been reported previously, both before [55,56] and after
[23,56] the onset of MS, with a number of interpret ations. Like MS,
anemia is common in young women, so it is possible that both have
common underlying risk factors. Other explanations include anemia
as an adverse ef fect of MS treatment, anemia of autoimmune origin,
and a higher likelihood of diagnosis of anemia in patients complain-
ing of symptoms such as fatigue, both before and after diagnosis.
However, that adjusting for other comorbidities made this associ-
ation nonsignificant suggests that it was related to unmasking bias.
When stratified by sex, our results are in line with the obser ved in-
creased risk of anemia in male compared with female MS patient s,
which can be related to the overall higher frequency of anemia in
women. This was found also for bipolar disorder, for which a higher
risk was found in male cases compared to controls. This is probably
due to th e higher freq ue nc y of both anemia an d so me forms of bip o-
lar disorder in women in general [57].
We find noteworthy that our results do not show an increased
risk of cancer or infectious comorbidities such as bronchitis, given
that it is likely that a significant proportion of patients did receive
disease- modifying therapy at some point before the data extraction.
However, because MS treatments are prescribed and dispensed at
the hospital level, information on their frequency was not available
for our study. For the same reason, we were unable to obtain in-
formation regarding outcomes of MS (such as relapses, disability,
or quality of life) and the influence of comorbidities on them. On
the other hand, we found a slightly higher likelihood of adhering to
follow- up guidelines for diabetes in MS patients than in controls,
which may be a reflection of more frequent clinical and laboratory
assessments in those patients. However, our data preclude drawing
conclusions on the impact of MS in relevant outcomes of the co-
morbidities. Interestingly, despite finding a high rate of overweight/
obesity in the general population, our results show an even higher
frequency in MS patients, which is similar to that reported in other
populations [21,58,59]. This finding was unexpected, given the
“Mediterranean lifest yle” considered to be prevalent in our milieu.
The frequency of current or ever smoking among MS cases
was also found to be high, and associated with high socioeconomic
deprivation, as has been previously reported [58]. Risk of smoking
was not especially high in women, which is important in relation to
the increasing female- to- male ratio of MS incidence [60]. However,
our study does not assess trends in adverse health behaviors over
time, so this is an area relevant for future studies. Our results show
a frequency of alcohol consumption in MS patients similar to that
previously described [58,61], and lower than in controls, which is
consistent with data of other case– control studies [62]. However, it
is impor tant to note that our design does not allow for the discrimi-
nation of nonproblematic alcohol intake from abuse or dependence,
which deserves further study, given their high frequency in MS pop-
ulations [58]. In addition, our results support the notion that adverse
health behaviors are more frequent in male MS patients [58].
According to studies in different countries, including Spain, low so-
cioeconomic status is a well- established risk factor for comorbidities
in the general population [63] and this has been described for the MS
population as well [9]. This was confirmed by our study showing that
MS patients in the two most deprived quintiles were at highest risk
of having at least one comorbidity. When analyzing health behaviors,
this was only true for ever smoker status and not for current smok-
ers or those who consume alcohol. We find this surprising in light of
studies showing the high frequency of adverse health behaviors in MS
patient s and of smoking among low income groups [58,64]. It could be
explained by a higher rate of ever smokers in the control group com-
pared wi th the gener al popu lat ion [65] acros s all socioecon omic st rata.
MS patients use more health care resources compared with their
age- and sex- matched controls in terms of clinical visits, medica-
tion consumption, and sick leave days. This is consistent with other
studies and is correlated with increased disabilit y [28,66] and an
increased number of comorbidities [7]. In our primar y health care
network, patient s with MS had a mean of three PCP and three nurse
visits per year, and these numbers increased when we focused on
MS patients with comorbidities. We found, however, that the num-
ber of sick leave days did not vary between MS patients with and
without comorbidities, which might indicate that this variable is
mainly driven by MS itself. Although the measurement of sick leave
days is highly reliable due to its automated extraction from the EHR ,
we were unable to discriminate patients on disability pension from
those who were retired for the purpose of the anal ysis. Because dis-
ability pension is in itself a form of health care resource usage in
universal health care systems such as ours and our study was unable
to assess it, it remains an impor tant issue for further research.
Our findings on the frequency of comorbidities are in line
with others from countries with comparable health care systems
[33,34,37,38,45,46,51]. Regarding health care resource usage, our
findings are broadly similar to those in France, in terms of yearly vis-
its to PCPs and nurses, which are increased with a higher comorbid-
ity burden [67].
We believe our study has a number of important methodologi-
cal streng ths, such as that it is a largely unbiased population- based
sample covering >80% of the population of Catalonia. The remaining
20% corresponds to other public health care providers but does not
impl y dif ferential characteristics of the population in ter ms of socio-
economic status or other aspects that could potentially introduce a
selection bias. Comparability between study groups is guaranteed
thanks to the random selection of matched control subjects. Finally,
the data have been systematically collected at the source following
quality controls and gathered in the SIDIAPQ database, which po-
tentially limit s error due to recall and referral bias. Previous studies
conducted with this database confirm that it is highly representative
of the population [26], and accurately reflects the real landscape of
health conditions in terms of the most common diseases and risk
factors in Catalonia [68]. This study is an additional example of how
|
9
COMORBIDITIES AND MULTIPLE SCLEROSIS
the structured gathering of administrative data can adequately iden-
tify comorbidities in MS patients [13].
We acknowledge that the study has a number of limitations.
First, because the comorbidities were defined according to their
ICD- 10 code, it is possible that a unique diagnostic code describes
an outcome of several different diseases (e.g., heart failure can be
the result of valve disease, as well as of coronary arter y disease). We
did not perform sensitivity analyses using alternative codes, which
is a limitation. However, the ICD- 10 codes used in this study have
been shown to have high rates of agreement with clinical records,
at least for the comorbidities for which we found significant differ-
ences in risk [69– 73]. We did not assess and control for the number
or frequency of visit s with the PCP, which might be a source of as-
certainment bias, and could explain in par t the increased frequency
of comorbidities in younger MS patients. As discussed previously,
elder patients may not have been included in the analysis due to
immortal time bias, which limits the external validity of our results
in the elderly. This is problematic, given that this is the population
in which multimorbidity is more frequent. Furthermore, the infor-
mation gathered might not be reflective of the real situation at the
time of the database lockup (e.g., a depressive episode might have
been resolved but the code not deactivated). In addition, although
the SIDIAPQ database provides highly representative data from the
population in Catalonia, the PCCs from the most deprived areas are
overrepresented [74]; this, and the possibility that the less deprived
population seeks private attention more frequently, are a possible
source of bias that could limit the generalizability of our findings
regarding socioeconomic status. Finally, our study's transversal de-
sign does not allow conclusions regarding the temporal relationship
between MS and its duration and comorbidities or adverse health
behaviors. This is important because the comorbidity profile in MS
patients changes over time [56], which warrant s further research.
In conclusion, our results suggest that, similar to other geo-
graphical regions, MS patients in Catalonia have a higher frequency
of depression [37,38], bipolar disorder [51], epilepsy [45,46], stroke
[33,34] and smoking [58], and a male predominance of adverse
health behaviors [58]. MS patients in this Mediterranean region use
more health care resources in terms of visits to the clinic, medication
use, and sick leave days, the latter being conditioned by MS itself
and not by comorbidities. A high rate of over weight/obesity and
adverse health behaviors found both in MS patient s and controls
is worrisome and deser ves further study. Our findings could guide
monitoring strategies and lifestyle interventions that reduce the
burden of MS on patients and health care systems.
ACKNOWLEDGMENTS
We thank Mónica Hoyos- Flight, PhD for her valuable comments
on the manuscript. This work was possible thanks to an ECTRIMS
Clinical Training Fellowship awarded to S.C.- R.
CONFLICT OF INTEREST
S.C.- R. has received travel expenses for scientific meetings from
Biogen, Tecnofarma, Merck, and Genzyme; compensation for
consulting services or participation on advisory boards from Merck,
Roche, and Novartis; speaking honoraria from Novartis; and research
support from Biogen. S.O.- R. has received compensation for consult-
ing services from Biogen and Genz yme, and research support from
Novartis. M.A.P.- B. has nothing to declare. P.C.- M.'s yearly salary is
suppor ted by a grant from Biogen to Fundació privada Cemcat toward
statistical analysis. J.S.- G. has received compensation for consulting
services and speaking honoraria from Almirall, Bayer, Biogen, Celgene,
Sanofi, Merck, Novartis, Roche, Bial, Biopass, and Teva, is a member
of the editorial commit tee of Multiple Sclerosis Journal, and direc tor of
Revista de Neurología. X.M. has received speaking honoraria and travel
expenses for scientific meetings, has been a steering committee mem-
ber of clinical trials, or has participated on advisor y boards of clinical
trials in the past 3 years with Actelion, Alexion , Bayer, Bio gen, Celgene,
EMD Serono, EXCEMED, Genzyme, MedDay, Merck, MSIF, Nervgen,
NMSS, Novartis, Roche, Sanofi- Genzyme, Teva Pharmaceutical, and
TG Therapeutics. M.T. has received compensation for consulting ser-
vices and speaking honoraria from Almirall, Bayer Schering Pharma,
Biogen, Genzyme, Merck- Serono, Novartis, Roche, Sanofi- Aventis,
Viela- Bio, and Teva Pharmaceuticals, and is coeditor of Multiple
Sclerosis Journal– Experimental, Translational and Clinical.
AUTHOR CONTRIBUTIONS
Simón Cárdenas- Robledo: Data curation (equal), formal analysis
(equal), investigation (equal), methodology (equal), writing– original
draft (equal), writing– review & editing (equal). Susana Otero-
Romero: Conceptualization (equal), data curation (equal), investi-
gation (equal), methodology (equal), project administration (equal),
supervision (equal), validation (equal), writing– original draft (equal),
writing– review & editing (equal). Maria Angels Passarell- Bacardit:
Conceptualization (equal), data curation (equal), funding acquisition
(equal). Pere Carbonell- Mirabent: Formal analysis (equal), investiga-
tion (equal). Jaume Sastre- Garriga: Conceptualization (equal), fund-
ing acquisition (equal), investigation (equal), writing– review & editing
(equal). Xavier Montalban: Conceptualization (equal), funding ac-
quisition (equal), investigation (equal), super vision (equal), writing–
review & editing (equal). Mar Tintoré: Conceptualization (equal),
data curation (equal), investigation (equal), methodology (equal),
project administration (equal), supervision (equal), writing– original
draft (equal), writing– review & editing (equal).
DATA AVAIL AB ILI T Y STAT EME N T
The data that support the findings of this study are available from
the corresponding author upon reasonable request.
ORCID
Simón Cárdenas- Robledo https://orcid.org/0000-0002-7612-3985
Susana Otero- Romero https://orcid.org/0000-0002-1451-6927
Jaume Sastre- Garriga https://orcid.org/0000-0002-1589-2254
REFERENCES
1. Feinstein AR. The pre- therapeutic classification of co- morbidity in
chronic disease. J Chronic Dis. 1970;23( 7):455- 468.
10
|
CÁRDENAS- RO BLEDO Et AL.
2. Kowalec K , McKay K A, Pat ten SB, et al. Comorbidity in-
creases the risk of relapse in multiple sclerosis . Neurology.
2017;89(24):2455- 2461.
3. Tettey P, Siejka D, Simpson S, et al. Frequency of comorbidities and
their association with clinical disability and relapse in multiple scle-
rosis. Neuroepidemiology. 2016;46(2):106- 113.
4. Zhang T, Tremlett H, Zhu F, et al. Ef fects of physical comorbidi-
ties on disability progression in multiple sclerosis. Neurology.
2018;90(5):e419- e427.
5. Zhang T, Tremlett H, Leung S, et al. Examining the effec ts of co-
morbidities on disease- modifying therapy use in multiple sclerosis.
Neurology. 2016;86(14):1287- 1295.
6. Laroni A, Signori A , Maniscalco GT, et al. Assessing association of co-
morbidities with treatment choice and persistence in MS. Neurology.
2017;89(22):2222- 2229. https://doi.org/10.1212/WNL.00000 00000
004686
7. McKay KA, Marrie RA, Fisk JD, Patten SB, Tremlett H. Comorbidities
are associated with altered health services use in multiple sclerosis:
a prospective cohor t study. Neuroepidemiology. 2 0 1 8 ; 5 1 ( 1 – 2 ) : 1 - 1 0 .
Available from https://www.kar ger.com/Artic le/FullT ext/488799
8. Warren SA, Turpin KVL, Pohar SL , Jones C A, Warren KG.
Comorbidit y and health- related quality of life in people with multi-
ple sclerosis. Int J MS Care. 20 09;11(1):6- 16.
9. Marrie RA, Horwitz R, Cutter G, Tyry T, Campagnolo D, Vollmer
T. Comorbidity, socioeconomic status and multiple sclerosis. Mult
Scler. 2008 ;14(8) :1091- 10 98.
10. Marrie RA , Fisk J, Tremlett H, et al. Differing trends in the incidence
of vascular comorbidity in MS and the general population. Neurol
Clin Pract. 2016;6(2):120- 128.
11. Marrie RA, Reider N, Cohen J, et al. A systematic review of the inci-
dence and prevalence of autoimmune disease in multiple sclerosis.
Mult Scler J. 2015;21(3):282- 293.
12 . Marrie RA , Yu BN, Leung S, et al. Rising prevalence of vascular co-
morbidities in multiple sclerosis: validation of administrative defi-
nitions for diabetes, hypertension, and hyperlipidemia. Mult Scler J.
2012;18(9):1310- 1319.
13. Marrie RA, Yu BN, Leung S, et al. The utility of administrative data
for surveillance of comorbidity in multiple sclerosis: a validation
stud y. Neuroepidemiology. 2013;40 (2):85- 92.
14. Bütepage G, Esawi A, Alexanderson K, et al. Cost- of- illness tra-
jectories among people with multiple sclerosis by comorbidity: a
register- base d prospective study in Sweden. Mult Scler J Exp Transl
Clin. 2020;6(4):205521732096859.
15. Binzer S, McKay KA, Brenner P, Hillert J, Manouchehrinia A .
Disability worsening among persons with multiple sclerosis and
depression: a Swedish cohor t study. Neurology. 2019;93(24):e221
6- e2223.
16. Roshanisefat H, Bahmanyar S, Hillert J, Olsson T, Montgomer y S.
Multiple sclerosis clinical course and cardiovascular disease risk -
Swedish cohort study. Eur J Neurol. 2014;21(11):1353- e88.
17. Nørgaard M, Veres K, Didden EM, Wormser D, Magyari M. Multiple
sclerosis and cancer incidence: a Danish nationwide cohort study.
Mult Scler Relat Disord. 2019;28:81- 85.
18. Hauer L, Perneczk y J, Sellner JA. A global view of comor bidit y in
multiple sclerosis: a systematic review with a focus on regional dif-
ferences, methodology, and clinical implications. J Neurol. 2020;1:3.
h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 0 0 4 1 5 - 0 2 0 - 1 0 1 0 7 - y
19. Marrosu MG, Cocco E, Lai M, Spinicci G, Pischedda MP, Contu P.
Patient s with multiple sclerosis and risk of t ype 1 diabetes mellitus
in Sardinia, Italy: a cohort study. Lancet. 2002; 359(9316 ):14 61- 1465.
20. Da Silva AM, Vilhena E, Lopes A, et al. Depression and anxiet y in
a Portuguese MS population: associations with physical disability
and severity of disease. J Neurol Sci. 2 0 1 1 ; 3 0 6 ( 1 - 2 ) : 6 6 – 7 0 . h t t p s : / /
doi.org/10.1016/j.jns.2011.03.042
21. Sicras- Mainar A, Ruíz- Beato E, Navarro- Artieda R , Maurino
J. Comorbidity and metab olic syndrome in patients with
multiple sclerosis from Asturias and Catalonia, Spain. BMC Neurol.
2017;17(1):134.
22. Nicoletti A, Sofia V, Biondi R, et al. Epilepsy and multiple sclerosis in
sicily: a population- based s tudy. Epilepsia. 2003;44(11):1445- 1448.
23. Koudriavtseva T, Renna R, Plantone D, Mandoj C, Piattella MC,
Giannarelli D. Association between anemia and multiple sclerosis.
Eur Neurol. 2015;73(3– 4):233- 237.
24. Marrie RA, Miller A , S ormani MP, et al. Recommendations for ob-
servational studies of comorbidity in multiple sclerosis. Neurology.
2016 ;86(15):14 46- 145 3.
25. Bolíbar B , Fina Avilés F, Morros R, et al. Base de datos SIDIAP: la
historia clínica informatizada de Atención Primaria como fuente
de información para la investigación epidemiológica. Med Clín.
2012 ;138 (14): 617- 621.
26. García- Gil MDM, Hermosilla E, Prieto- Alhambra D, et al.
Constr uction and validation of a scoring system for the selection
of high- quality data in a Spanish population primar y care database
(SIDIAP). Inform Prim Care. 2011;19(3):135- 145.
27. Domínguez- Berjón MF, Borrell C, Cano- Serral G, et al. Constructing
a deprivation index based on census data in large Spanish cities(the
MEDEA project). Gac Sanit. 200 8;22(3):179- 187.
28. Kobelt G, Thompson A, Berg J, Gannedahl M, Eriksson J. New in-
sights into the burden and costs of multiple sclerosis in Europe.
Mult Scler J. 2017;23(8):1123- 1136.
29. Marrie R A, Elliott L , Marriott J, et al. Effe ct of comorbidity on mor-
tality in multiple sclerosis. Neurology. 2015;85(3):240- 247.
30. Marrie RA , Hor witz R, Cutter G, Tyry T, Campagnolo D, Vollmer T.
Comorbidity delays diagnosis and increases disability at diagnosis
in MS. Neurology. 2009;72(2):117- 124.
31. Nguyen H, Manolova G, Daskalopoulou C, Vitoratou S, Prince M ,
Prina AM. Prevalence of multimorbidity in community settings: A
systematic review and meta- analysis of observational studies. J
Comorbidity. 2019;9:2235042X1987093.
32. Chou IJ, Kuo CF, Tanasescu R , et al. Comorbidity in multiple sclero-
sis: its temporal relationships with disease onset and dose effec t on
mort ality. Eur J Neurol. 2020;27(1):105- 112.
33. Jadidi E, Mohammadi M, Moradi T. High risk of cardiovascu-
lar diseases after diagnosis of multiple sclerosis. Mult Scler J.
2013;19(10):1336- 1340.
34. Thormann A , Magyari M, Koch- Henriksen N, Laursen B, Sørensen
PS. Vascular comorbidities in multiple sclerosis: a nationwide study
from Denmark. J Neurol. 2 016;263(12): 248 4- 2493.
35. Wood B, van der Mei I, Ponsonby A- L, et al. Prevalence and con-
currence of anxiety, depression and fatigue over time in mul-
tiple sclerosis. Mult Scler J. 2013;19(2):217- 224. https://doi.
org /10.117 7/13524 58512 4503 51
36. Beiske AG, Svensson E, Sandanger I, et al. Depression and anxiety
amongst multiple sclerosis patients. Eur J Neurol. 2008;15(3):239- 245.
37. Hoang H, Laursen B, Stenager EN, Stenager E. Psychiatric co-
morbidity in multiple sclerosis: the risk of depression and anxiety
before and after MS diagnosis. Mult Scler J. 2016;22(3):347- 353.
38. Marrie RA , Walld R, Bolton JM, et al. Estimating annual preva-
lence of depression and anxiety disorder in multiple sclerosis using
administrative data. BMC Res Notes. 2017;10(1) :619. ht tps://doi .
o r g / 1 0 . 1 1 8 6 / s 1 3 1 0 4 - 0 1 7 - 2 9 5 8 - 1
39. Fiest KM, Fisk JD, Patten SB, et al. Fatigue and comorbidities in
multiple sclerosis. Int J MS Care. 2016;18(2):96- 104.
40. Amato MP, Ponziani G, Rossi F, Liedl CL, Stefanile C, Rossi L. Quality
of life in multiple sclerosis: the impact of depression, fatigue and
disability. Mult Scler J. 2001;7(5):340- 3 44.
41. Hanna M, Strober LB. Anxiet y and depression in Multiple Sclerosis
(MS): antecedents, consequences, and differential impact on well-
being and quality of life. Mult Scler Relat Disord. 2020;44:102261.
https://doi.org/10.1016/j.msard.2020.102261
42. Schmidt S, Jöstingmeyer P. Depression, fatigue and disability are in-
dependently associated with quality of life in patients wit h multiple
|
11
COMORBIDITIES AND MULTIPLE SCLEROSIS
Sclerosis: results of a cross- sectional study. Mult Scler Relat Disord.
2019;35:262- 269. https://doi.org/10.1016/j.msard.2019.07.029
43. Cetin K, Johnson KL, Ehde DM, Kuehn CM, Amtmann D, Kraf t GH.
Antidepressant use in multiple sclerosis: epidemiologic study of a
large community sample. Mult Scler. 2007;13(8):10 46- 1053.
4 4. Mirsky MM, Marrie RA , Rae- Grant A. Antidepressant drug treat-
ment in association with multiple sclerosis disease- modifying ther-
apy. Int J MS Care. 2016;18(6):305- 310.
45. Chou IJ, Kuo CF, Tanasescu R, et al. Epilepsy and associated mortality
in patients with multiple sclerosis. Eur J Neurol. 2019;26 (2 ):3 42- e 23 .
46. Allen AN, Seminog OO, Goldacre MJ. Association bet ween multiple
sclerosis and epilepsy: large population- based record- linkage stud-
ies. BMC Neurol. 2013;13 (1):189.
47. Uribe- San- Mar tín R , Ciampi- Díaz E, Suarez- Hernández F, Vásquez-
Torres M, Godoy- Fernández J, Cárcamo- Rodríguez C. Prevalence
of epilepsy in a cohort of patient s with multiple sclerosis. Seizure.
2014;23(1):81- 83.
48. C art a MG, Moro MF, Lorefice L, et al. The risk of bipolar disorders
in multiple sclerosis. J Af fect Disord. 2014;155(1):2 55- 260.
49. Johansson V, Lundholm C, Hillert J, et al. Multiple sclerosis and
psychiatric disorders: comorbidity and sibling risk in a nationwide
Swedish cohort. Mult Scler J. 2014;20(14):1881- 1891.
50. Marrie RA, Patten SB, Tremlett H, et al. Sex dif ferences in
comorbidity at diagnosis of multiple sclerosis. Neurology.
2016;86(14):1279- 1286.
51. Jun- O’Connell AH, Butala A , Morales IB, et al. The prevalence
of bipolar disorders and association with qualit y of life in a co-
hort of patients with multiple sclerosis. J Neuropsychiatry Clin
Neurosci. 2017; 29(1):45 - 51. ht tp s://doi.org/10.1176/ap pi.ne uro
psych.15120403
52. C arta Mg, Moro Mf, Lorefice L , Picardi A, et al. Multiple sclerosis
and bipolar disorders: the burden of comorbidity and its conse-
quences on qualit y of life. J Affect Disord. 2 014;167:192- 197.
53. Smith Daniel J, Martin Daniel, McLean Gary, Langan Julie, Guthrie
Bruce, Mercer Stewart W. Multimorbidit y in bipolar disorder and under-
treatment of cardiovascular disease: a cross sectional study. BMC Med.
2013;11(1):1741- 7015. https://doi.o rg /10.118 6/1741- 7015- 11- 263
54. Mitchell AJ, Lord O, Malone D. Differences in the prescribing of
medication for physical disorders in individuals with v. without
mental illness: meta- analysis. Br J Psychiatry. Cambridge University
Press; 2012;201:435- 443. Available from: ww w.bnf.org
55. Yusuf FLA , Wijnands JMA, Kingwell E, et al. Fatigue, sleep disor-
ders, anaemia and pain in the multiple sclerosis prodrome. Mult
Scler J. 2021;27(2):290– 302.
56. Lo LMP, Taylor BV, Winzenberg T, Palmer AJ, Blizzard L, van der Mei
I. Change and onset- type differences in the prevalence of comor-
bidities in people with multiple sclerosis. J Neurol. 2020;268(2):602-
6 1 2 . h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 0 0 4 1 5 - 0 2 0 - 1 0 1 9 4 - x
57. Arnold LM. Gender differences in bipolar disorder. Psychiatr Clin
North Am. 2003;26(3):595- 620.
58 . Marrie R, Hor witz R, Cut ter G, Tyry T, Campagnolo D, Vollmer T.
High frequency of adverse health behaviors in multiple sclerosis.
Mult Scler J. 20 09;15(1):105- 13. https: //doi.o rg /10.117 7/13524
58508 096680
59. Marck CH, Neate SL, Taylor KL, Weiland TJ, Jelinek GA . Prevalence
of comorbidities, over weight and obesity in an international sample
of people with multiple sclerosis and associations with modifiable
lifest yle factors . PLoS One. 2016;11(2):e0148573.
60. Koch- Henriksen N, Sørensen PS. The changing demographic pattern
of multiple sclerosis epidemiology. Lancet Neurol. 2010;9(5):520-
5 3 2 . h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / S 1 4 7 4 - 4 4 2 2 ( 1 0 ) 7 0 0 6 4 - 8
61. Weiland TJ, Hadgkiss EJ, Jelinek G A, Pereira NG, Marck CH,
van der Meer DM. The association of alcohol consumption and
smoking with qualit y of life, disability and disease activit y in an in-
ternational sample of people with multiple sclerosis. J Neurol Sci.
2014;336(1– 2):211- 219.
62. Zhu T, Ye X, Zhang T, et al. Association between alcohol consump-
tion and multiple sclerosis: a meta- analysis of obser vational stud-
ies. Neurol Sci. 2015;36(9):1543- 1550.
63. Afshar S, Roderick PJ, Kowal P, Dimitrov BD, Hill AG. Multimorbidity
and the inequalities of global ageing: a cross- sec tional study of
28 countries using the World Health Surveys. BMC Public Health.
2015;15(1):1- 10.
6 4. C asetta B, V idela AJ, Bardach A, et al. A ssociation between cig-
arette smoking prevalence and income level: a systematic review
and meta- analysis. Nicotine Tob Res. 2017;19(12):14 01- 14 07.
65. Departament de Salut Generalitat de Catalunya. Enquesta de
salut de Catalunya 2014. Informe dels principals resultats, 2ndda.
Barcelona. edn; 2015.
66. Sicras- Mainar A, Ruíz- Beato E, Navarro- Ar tieda R, Maurino J.
Impact on healthcare resource utilization of multiple sclero-
sis in Spain. BMC Health Serv Res. 2017;17(1):854. htt ps://doi.
o r g / 1 0 . 1 1 8 6 / s 1 2 9 1 3 - 0 1 7 - 2 8 0 7 - x
67. Roux J, Guilleux A, Lefort M, Leray E. Use of healthcare services
by patients with multiple sclerosis in France over 2010– 2015: a na-
tionwide population- based study using health administrative data.
Mult Scler J Exp Transl Clin. 2019;5(4):2010- 2015.
68. Ramos R, Balló E, Marrugat J, et al. Validez del Sistema de
Información para el Desarrollo de la Investigación en Atención
Primaria (SIDIAP) en el estudio de enfermedades vasculares: estu-
dio EMMA . Rev Esp Cardiol. 2012;65(1):29- 37.
69. Henderson T, Shepheard J, Sundararajan V. Quality of diagnosis
and procedure coding in ICD - 10 administrative data. Med Care.
2006;4 4(11):1011- 1019.
70. Januel J- M, Luthi J- C, Quan H, et al. Improved accuracy of co-
mor bidit y coding over tim e af ter the introd uc ti on of ICD - 10 admin-
istrative data. BMC Health Serv Re s. 20 11;11:194.
71. Smith JR, Jones FJS, Fureman BE, et al. Accuracy of ICD- 10- CM
claims- based definitions for epilepsy and seizure type. Epilepsy Res.
2020;166:106414.
72. Moreno- Iribas C, Sayon- Orea C, Delfrade J, et al. Validity of type
2 diabetes diagnosis in a population- based electronic health record
database. BMC Med Inform Decis Mak. 2017;17(1):3 4.
73. Davis K AS, Bashford O, Jewell A , et al. Using data linkage to ele c-
tronic patient records to assess the validity of selected mental
health diagnoses in English Hospital Episode Statistics (HES). PLoS
One. 2018;13(3):e0195002.
74. A gència de Salut Pública de Barcelona. La salut a Barcelona 2015:
Agència de Salut Pública de Barcelona; 2016.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
Supporting Information section.
How to cite this article: Cárdenas- Robledo S, Otero- Romero
S, Passarell- Bacardit MA, et al. Multiple sclerosis is
associated with higher comorbidity and health care resource
use: A population- based, case– control study in a western
Mediterranean region. Eur J Neurol. 2021;00:1– 11. ht t p s ://
doi .org /10.1111/ene.15030
