BJD

E P I DE M I O L O G Y British Journal of Dermatology

Prevalence and determinants of seborrhoeic dermatitis in a

middle-aged and elderly population: the Rotterdam Study*
M.G.H. Sanders iD ,1 L.M. Pardo,1 O.H. Franco,2 R.S. Ginger3 and T. Nijsten iD 1
Departments of 1Dermatology and 2Epidemiology, Erasmus Medical Center, Rotterdam, the Netherlands
3
Unilever Research and Development, Colworth Science Park, Sharnbrook, U.K.

Linked Comment: Drucker and Doiron. Br J Dermatol 2018; 178:22–23.

Summary

Correspondence Background Seborrhoeic dermatitis is a chronic relapsing inflammatory skin disease

Tamar Nijsten. with unclear pathophysiological mechanisms.
E-mail: t.nijsten@erasmusmc.nl Objectives To establish which lifestyle and physiological determinants are associated
with seborrhoeic dermatitis.
Accepted for publication
Methods Seborrhoeic dermatitis was diagnosed by a trained physician during a
23 September 2017
full-body skin examination within the Rotterdam Study, a prospective popula-
Funding sources tion-based cohort study in middle-aged and elderly people. The current design is
This study is funded by Unilever. The Rotterdam a comparative cross-sectional study embedded in the Rotterdam Study. Potential
Study is funded by Erasmus Medical Center and factors were identified from the literature and analysed in a multivariable logistic
Erasmus University Rotterdam; the Netherlands
regression, including: age, sex, obesity, skin colour, stress, depression, education
Organization for Health Research and Development
(ZonMw); the Research Institute for Diseases in
level, hypertension, climate, xerosis cutis, alcohol and tobacco use.
the Elderly (RIDE); the Ministry of Education, Results Of the 5498 participants, 788 participants were diagnosed with sebor-
Culture and Science, the Ministry for Health, rhoeic dermatitis (14
3%). We found associations between seborrhoeic dermatitis
Welfare and Sports, the European Commission and male sex [adjusted odds ratio (OR) 2
09, 95% confidence interval (CI)
(DG XII) and the Municipality of Rotterdam. 1
77–2
47], darker skin (adjusted OR 0
39, 95% CI 0
22–0
69), season (summer
vs. winter: adjusted OR 0
63, 95% CI 0
48–0
82) and generalized xerosis cutis
Conflicts of interest
M.G.H.S. is supported by Unilever and R.S.G. is
(adjusted OR 1
41, 95% CI 1
11–1
80).
a Unilever employee. Conclusions Seborrhoeic dermatitis is one of the most common inflammatory
dermatoses in middle-aged and elderly individuals, especially during winter. Men,
*Plain language summary available online and people with a light and dry skin were most likely to have seborrhoeic dermatitis.

DOI 10.1111/bjd.15908
What’s already known about this topic?
• Seborrhoeic dermatitis is a common chronic relapsing inflammatory skin disease
with unclear pathophysiological mechanisms.
• Numerous host and environmental factors have been implicated in the pathogenesis
of seborrhoeic dermatitis, but most have not been validated.

What does this study add?

• Male sex, light skin colour, the season winter and generalized dry skin increased
the likelihood of having seborrhoeic dermatitis.
• Interventions aiming to improve skin-barrier function could become a target in the
treatment of patients with seborrhoeic dermatitis.

Seborrhoeic dermatitis is a common relapsing inflammatory severe form of erythema and scaling on the scalp, face and
skin disease, characterized by poorly demarcated erythematous chest. Cross-sectional and retrospective studies using medical
patches with greasy scaling.1 The diagnosis is mainly clinical records show a prevalence between 2% and 8%,3 but the
and the disease spectrum is heterogeneous, ranging from a prevalence in prospective studies was found to be as high as
mild form of scaling limited to the scalp (dandruff),2 to a 23% in selected populations.4,5 Despite the high frequency of

148 British Journal of Dermatology (2018) 178, pp148–153 © 2017 British Association of Dermatologists
 Prevalence and determinants of seborrhoeic dermatitis, M.G.H. Sanders et al. 149

seborrhoeic dermatitis in the population and the impaired alcohol misuse and smoking. Xerosis cutis was chosen as a
quality of life associated with seborrhoeic dermatitis,6 studies proxy for barrier dysfunction.
investigating the aetiology of seborrhoeic dermatitis are scarce. Age, sex, weight, height, blood pressure, presence of xero-
Both host and environmental determinants have been sis cutis (no, extensor-side extremities or generalized) and
implicated in the pathogenesis of seborrhoeic dermatitis. It skin colour24 were registered at the FBSE. Body mass index
has been postulated that overgrowth of commensal Malasse- (BMI) < 25 kg m 2 was defined as normal weight, 25–30 as
zia species, in a predisposed host (e.g. those with a overweight and > 30 as obese.25 Diastolic blood pressure of
decreased barrier function), is causal in the development of > 90 mmHg or systolic blood pressure of > 140 mmHg was
seborrhoeic dermatitis.7 Several other determinants have been defined as hypertension.26 The date of FBSE was used to deter-
studied as well, including older age,3,8,9 male sex,3,8–11 obe- mine the season. Other determinants were assessed at a home
sity,4,12 skin colour,4 stress,13–15 depression,13,16 winter cli- interview, including: education level (low: primary education;
mate,17,18 alcohol consumption,12,19 tobacco exposure, medium: lower-intermediate vocational education; high: gen-
hypertension,12 HIV20 and Parkinson disease.21,22 However, eral secondary education and higher), alcohol (glasses/week),
some of these studies were small (less than 100 partici- tobacco consumption (never and former vs. current), depres-
pants), or conducted in a selected population (e.g. dermatol- sion and anxiety. The home interviews were conducted before
ogy outpatients). Knowledge about determinants could the FBSE. The time between the home interview and the FBSE
provide new insights into the pathophysiology of seborrhoeic was less than 6 months for more than 95% of the participants.
dermatitis, which might lead to preventative strategies and/ To assess depression, the Center for Epidemiologic Studies
or new treatment regimens. Depression Scale was used and a score of ≥ 22 was taken as the
The aim of this study was to investigate the reported deter- cut-off point for depression.27 Anxiety disorders were used as a
minants of seborrhoeic dermatitis in a population-based study proxy of stress and were assessed by a slightly adapted version
of middle-aged and elderly people in the Netherlands (Rotter- of the Munich Composite International Diagnostic Interview.28
dam Study), in which people were diagnosed with sebor- Participants without any anxiety disorder were compared with
rhoeic dermatitis by a physician during skin examination. participants with one or more anxiety disorders.

Patients and methods Statistical analysis

Patient characteristics were compared for those with or with-
Study population
out the disease using a v2-test for categorical variables and a t-
The current design is a comparative cross-sectional study test for continuous variables. Univariable logistic regression
embedded in the Rotterdam Study, an ongoing prospective analysis was then used to calculate the crude odds ratio (OR)
population-based cohort study of chronic diseases in a mid- with 95% confidence interval (CI). This was done to assess
dle-aged and elderly population in the Ommoord district of the strength of the association between the disease and the
Rotterdam, the Netherlands.23 The study started in 1990 and factors selected: age, sex, BMI, skin colour, season, smoking,
now comprises 14 926 participants aged 45 years or older alcohol, education level, hypertension, depression, stress and
and since 2010, 5498 participants have had a skin examina- xerosis cutis. Subsequently, these variables were all included
tion. in a multivariable logistic regression to adjust for possible
confounders. The fully conditional specification imputation
method with 20 imputations was used to replace missing val-
Case definition
ues on the determinants mentioned above.29 The outcome
A full-body skin examination (FBSE) was conducted by der- measurement (seborrhoeic dermatitis) was not imputed. A
matology-trained physicians. The diagnosis of seborrhoeic der- sensitivity analysis was carried out using only complete cases
matitis was based on greasy scaling, erythema and a to verify the imputation method in the multivariable model.
characteristic distribution in areas rich in sebaceous glands. All Statistical analyses were performed in IBM SPSS Statistics for
three criteria had to be present for the diagnosis. Windows version 21.0 (IBM, Armonk, NY, U.S.A.).

Associated factors Subgroup analysis

A literature search for potential determinants of seborrhoeic Subgroup analysis was done to test whether previously
dermatitis was carried out using PubMed, Embase, Web of reported sex differences in seborrhoeic dermatitis were the
Science and Google Scholar and it was limited to observational result of hormonal differences. Hormone levels were available
studies on humans, published in English and updated until for a previous Rotterdam Study visit. On average, hormones
January 2016. The following determinants of seborrhoeic der- were measured 7
5 years before the FBSE. We tested the asso-
matitis were identified: elderly age, male sex, obesity, white ciation between seborrhoeic dermatitis and total testosterone
skin colour, stress, depression, hypertension, HIV infection, in men, using a logistic regression adjusting for age, skin col-
Parkinson disease, winter climate, high educational level, our, xerosis cutis and season. We used the same model in

© 2017 British Association of Dermatologists British Journal of Dermatology (2018) 178, pp148–153
150 Prevalence and determinants of seborrhoeic dermatitis, M.G.H. Sanders et al.

women, but replaced total testosterone with the Free Andro- Table 1 Characteristics of participants with and without seborrhoeic
gen Index (FAI). Also, we determined the P-value of the inter- dermatitis
action terms between sex and the other determinants in the
full model. Seborrhoeic No seborrhoeic
Coding dermatitis dermatitis P-valuea
Participants 788 (14
3) 4710 (85
7)
Results Sex
Male 466 (59
1) 1922 (40
8) < 0
001
A total of 5498 participants from the Rotterdam Study under-
Age (years), 70
1 (63
1–77
4) 67
7 (61
7–76
3) 0
027
went a FBSE (Table 1). The median age of participants at the
median (IQR)
date of the FBSE was 67
9 years [interquartile range (IQR) Body mass index (kg m 2)
61
9–76
4] and the proportion of women was 57%. Of the < 25 215 (27
3) 1418 (30
1) 0
241
5498 participants, 788 participants were diagnosed with seb- 25–30 383 (48
6) 2165 (46
0)
orrhoeic dermatitis (point prevalence of 14
3%). The percent- > 30 189 (24
0) 1115 (23
7)
age of missing data per variable was low (< 5%), except for Missing 1 (0
1) 12 (0
3)
Skin colour
alcohol use (17
3%; Table 1).
Very white– 694 (88
1) 3858 (81
9) < 0
001
In the univariable logistic regression we found significant
white
associations between an increased prevalence of seborrhoeic White–olive 81 (10
3) 650 (13
8)
dermatitis and age, male sex, a high education level, general- Light brown– 13 (1
6) 202 (4
3)
ized xerosis cutis and alcohol consumption, whereas darker black
skin colour and summer season were associated with a Education
decreased prevalence (Table 2). In the multivariable model, Low 74 (9
4) 477 (10
1) 0
038
Average 454 (57
6) 2884 (61
2)
male sex (adjusted OR 2
09, 95% CI 1
77–2
47), a darker
High 250 (31
7) 1286 (27
3)
skin colour (white–olive and brown skin vs. white skin:
Missing 10 (1
3) 63 (1
3)
adjusted OR 0
39, 95% CI 0
22–0
69), season (summer vs. Depression
winter: adjusted OR 0
63, 95% CI 0
48–0
82) and generalized CESD ≥ 22 31 (3
9) 199 (4
2) 0
687
xerosis cutis (adjusted OR 1
41, 95% CI 1
11–1
80) remained Missing 7 (0
9) 52 (1
1)
significant predictors for seborrhoeic dermatitis (Table 2). All Anxiety disorder(s)
significant variables remained significant in the model without Yes 43 (5
5) 234 (5
0) 0
623
Missing 27 (3
4) 227 (4
8)
imputed data and the effect sizes were comparable (data not
Xerosis cutis
shown).
No 271 (34
4) 1815 (38
5) 0
017
In the logistic regression analyses between seborrhoeic der- Extensor-side 390 (49
5) 2293 (48
7)
matitis and total testosterone in men (adjusted OR 1
00, 95% extremities
CI 0
98–1
02) and seborrhoeic dermatitis and the FAI in Generalized 121 (15
4) 557 (11
8)
women (adjusted OR 1
01, 95% CI 0
92–1
10) no significant Other 6 (0
8) 43 (0
9)
associations were found. Also, none of the interaction terms Missing 0 2 (0
04)
Hypertension
between sex and the other determinants were found to be sig-
Yes 418 (53
6) 2413 (51
2) 0
329
nificant (data not shown).
Missing 8 (1
0) 43 (0
9)
Table 3 presents an overview of the determinants for sebor- Season
rhoeic dermatitis from previously published papers in compar- Winter 224 (28
4) 1091 (23
2) 0
001
ison with the data in the current study. We replicated Spring 172 (21
8) 1056 (22
4)
previous associations between seborrhoeic dermatitis and sex, Summer 92 (11
7) 756 (16
1)
skin colour and season; however, we did not replicate several Autumn 300 (38
1) 1807 (38
4)
Smoking
other associations such as obesity, depression, hypertension,
Current 115 (14
6) 682 (14
5) 0
937
alcohol consumption and tobacco use.
Missing 1 (0
1) 8 (0
2)
Alcohol (glasses/week)
Median (IQR) 5
8 (0
8–14
6) 4
4 (0
6–11
8) 0
004
Discussion
Missing 119 (15
1) 831 (17
6)
In this middle-aged and elderly population, the point preva-
Data are n (%) unless otherwise specified. IQR, interquartile
lence of seborrhoeic dermatitis was 14
3%, which is higher
range; CESD, Center for Epidemiologic Studies Depression Scale.
than the 2–8% in previous studies.1,3,30 This difference may v -test for categorical and t-test for continuous variables.
a 2

be because most of these studies were based on patient

records, and therefore most likely included patients with mod-
erate-to-severe seborrhoeic dermatitis. In contrast, the Rotter- dermatitis prevalence increases with age.5,31 Indeed, several
dam Study is a population-based study in which all grades of physiopathological alterations occur in the skin of elderly peo-
seborrhoeic dermatitis were considered. Furthermore, the Rot- ple (i.e. a decreased amount of lipids in the stratum corneum
terdam Study focuses on elderly people, and seborrhoeic and thinning of the epidermis and dermis) resulting in a

British Journal of Dermatology (2018) 178, pp148–153 © 2017 British Association of Dermatologists
 Prevalence and determinants of seborrhoeic dermatitis, M.G.H. Sanders et al. 151

Table 2 Associations between seborrhoeic dermatitis and reported men and seborrhoeic dermatitis and the FAI in women sug-
determinants gested no direct association between previously measured
hormone levels and presence of seborrhoeic dermatitis among
Crude odds Adjusted odds the participants of the Rotterdam Study.33 As the hormone
Variable, coding ratio (95% CI) ratio (95% CI)
measurements were taken several years before the FBSE, repli-
Sex, male 2
10 (1
8–2
45)** 2
09 (1
77–2
47)** cation of our findings with more recent data is recom-
Age at FBSE (years), 1
01 (1
00–1
02)* 1
01 (0
99–1
02) mended. Sex differences could also be explained in part by
continuous
cutaneous differences in pH and/or use of skin products,
BMI (kg m 2)
which might influence microbial colonization and barrier
< 25 Reference Reference
25–30 1
17 (0
97–1
40) 1
05 (0
87–1
27) integrity.34,35 Unfortunately, these variables were not avail-
> 30 1
12 (0
91–1
38) 1
12 (0
90–1
39) able in the Rotterdam Study and therefore were not investi-
Skin colour gated. Other studies will be needed to assess the behavioural
Very white–white Reference Reference or biological differences underlying the statistical associations
White–olive 0
69 (0
54–0
88)* 0
70 (0
53–0
86)* between sex and seborrhoeic dermatitis.
Light brown–black 0
36 (0
20–0
63)** 0
39 (0
22–0
69)*
In line with previous studies, the prevalence of seborrhoeic
Education
dermatitis was lowest during summer, which could be
Low 0
98 (0
76–1
28) 0
99 (0
75–1
29)
Average Reference Reference explained by increased ultraviolet (UV) radiation exposure.
High 1
23 (1
04–1
45)* 1
14 (0
85–1
53) UV-induced cutaneous immunosuppressive effects might
Depression, 1
03 (0
64–1
65) 1
06 (0
70–1
59) reduce the inflammatory response to environmental stimuli
CESD ≥ 22 and UV radiation has a direct effect on the microbiome,
Anxiety 1
24 (0
83–1
84) 1
33 (0
93–1
89) including on Malassezia yeast of the skin.36,37 The decreased
disorder(s), yes
risk of seborrhoeic dermatitis in people with coloured skin is
Xerosis cutis
again consistent with a previous study.4 Although this may be
No Reference Reference
Extensor-side 1
14 (0
97–1
34) 1
14 (0
96–1
35) because of a diagnostic bias in dark-skinned individuals,
extremities because erythema is less striking, it might also be the result of
Generalized 1
46 (1
17–1
81)* 1
41 (1
11–1
80)* a superior barrier function in more pigmented skin.38
Other 0
94 (0
60–1
46) 0
79 (0
33–1
89) Several studies underline the importance of skin-barrier
Hypertension, yes 0
97 (0
82–1
16) 0
99 (0
85–1
16) function in the pathogenesis of seborrhoeic dermatitis.2,7,32 In
Season
this study, xerosis cutis was taken as a proxy for skin-barrier
Winter Reference Reference
dysfunction and an association between generalized xerosis
Spring 0
79 (0
64–0
98)* 0
84 (0
67–1
05)
Summer 0
59 (0
46–0
77)** 0
63 (0
48–0
82)** cutis and seborrhoeic dermatitis was observed. Although we
Autumn 0
81 (0
67–0
98)* 0
81 (0
67–0
98)* cannot rule out overdiagnosis of seborrhoeic dermatitis
Smoker, 1
01 (0
81–1
25) 0
98 (0
78–1
22) because of excessive squamae that may accompany xerosis
yes (current) cutis of the scalp and face, the association implies that partici-
Alcohol 1
01 (1
01–1
02)** 1
01 (0
99–1
01) pants with seborrhoeic dermatitis more often have a skin-bar-
(glasses/week),
rier dysfunction and that this is not limited to the areas
continuous
affected by seborrhoeic dermatitis. Therefore, interventions
CI, confidence interval; FBSE, full-body skin examination; CESD, aiming to improve skin-barrier function may become a target
Center for Epidemiologic Studies Depression Scale. *P < 0
05, in the treatment of patients with seborrhoeic dermatitis.
**P < 0
001. In contrast to previous studies,12,19 we did not find associa-
tions for other determinants, including smoking, alcohol and
education (Table 3). None of these previous studies adjusted
higher vulnerability to external stimuli in this age group.31,32 for the important seborrhoeic dermatitis predictors such as
In contrast to other studies, age was not significant in the sex, age or skin colour, leading to residual confounding and
multivariable analysis, most likely because of the advanced age spurious associations, especially because differences in smok-
of the entire study population. ing and alcohol consumption between men and women are
Of the reported determinants of seborrhoeic dermatitis, well known.39 In addition, the definition of determinants
sex, skin colour and season were seen in the Rotterdam between studies differed and might explain the lack of replica-
Study, whereas obesity, depression, hypertension, alcohol tion of other determinants such as obesity, stress and depres-
consumption and tobacco use were not (Table 3). We found sion. A large study with 9255 seborrhoeic dermatitis cases and
that men were twice as likely to develop seborrhoeic der- 9246 controls suggested that seborrhoeic dermatitis and
matitis than women. Hormonal differences between men and hypertension were associated (adjusted OR 1
23, 95% CI
women may explain this association, as seborrhoeic dermati- 1
12–1
35), but we did not replicate this finding.12
tis has an incidence peak during puberty, where the levels of We did not investigate the association between seborrhoeic
androgens in men are high.1,30 However, logistic regression dermatitis and Parkinson disease and HIV, because the data
analyses on seborrhoeic dermatitis and total testosterone in were incomplete and/or not available. Another limitation lies

© 2017 British Association of Dermatologists British Journal of Dermatology (2018) 178, pp148–153
152 Prevalence and determinants of seborrhoeic dermatitis, M.G.H. Sanders et al.

Table 3 A summary of reported and tested risk factors for seborrhoeic dermatitis

Previous studies
Rotterdam Study
Determinants References Summary Measurements Summary (Table 2)
3, 8–11
Male sex Higher prevalence for males in adult Ratio, male: female Seborrhoeic dermatitis
population  1
4 : 1 has a strong association
with male sex
3, 8, 9
Age Peak between 30 and 50, which Prevalence graphs No association. Relatively
mostly persists in old age high prevalence in this
elderly cohort
4, 12
Obesity Increased triceps skinfold in male Adjusted PR 1
56, 95% No association
adolescents CI 1
12–2
18
Higher percentage of obesity in adult 16
3% vs. 15
3%,
population P = 0
014
4
Skin colour Higher prevalence in white skinned Adjusted PR (white skin) Decreased risk in white–
male adolescents 1
42, 95% CI 1
06–1
92 olive and brown skin vs.
white skin
Education 4, 12
No significant effect of socio- – No association
economic status in two studies
Higher prevalence with high socio- 25
4% vs. 19
1%,
economic status in one study P = 0
001
12
Hypertension One study found an association Adjusted OR 1
23, 95% No association
between hypertension and CI 1
12–1
35
seborrhoeic dermatitis
Depression 13, 16
Increased prevalence in depressive 36% vs. 9%, P < 0
01 No association
patients, not adjusted
Stress 13–15
Self-reported triggering factor – No association
More stressful life events in Mean  SD: 5
47  3
92
seborrhoeic dermatitis patients vs. 3
27  2
96,
P < 0
05
Higher perceived stress scale Adjusted OR 1
065 95%
CI 1
021–1
110,
P = 0
003
17, 18
Season Self-reported seasonal influences: 52% reported seasonal Lowest risk for
higher disease severity in winter influences: more skin seborrhoeic dermatitis
and lowest in summer problems in winter in summer
(50%), fewer in
summer (77%)
Visits for the diagnosis of Spring 42%, summer 7%,
seborrhoeic dermatitis to the autumn 25%, winter
outpatient clinic 25%
4, 12
Smoking One study showed no effect of Adjusted PR (smoking) No association
smoking 1
06, 95% CI 0
75–
Another study showed a slightly 1
50, P = 0
7
lower incidence of smokers in the 16
2% vs. 19
6%,
seborrhoeic dermatitis group, not P = 0
001
adjusted
12, 19
Alcohol No study with a control group Alcohol misuse 0
3% No association
shows an association (seborrhoeic dermatitis)
vs. 0
5% (control),
P = 0
13

PR, prevalence ratio; CI, confidence interval; OR, odds ratio.

in the cross-sectional design of the analysis as this means it is epidemiological determinants that allowed us to control for
not possible to make causal inferences between seborrhoeic potential confounders.
dermatitis and determinants. Further, this study covers a mid- In conclusion, one out of seven people had seborrhoeic der-
dle-aged and elderly population, which limits the generaliz- matitis, and men and those with generalized xerosis cutis were
ability to younger patients. The strength of our study is the even more likely to have seborrhoeic dermatitis. Participants
population-based setting, the large sample size, the physician- with a darker complexion and those assessed during the sum-
based diagnosis and the availability of different mer period were less likely to have the condition. The

British Journal of Dermatology (2018) 178, pp148–153 © 2017 British Association of Dermatologists
 Prevalence and determinants of seborrhoeic dermatitis, M.G.H. Sanders et al. 153

association between generalized xerosis cutis and seborrhoeic 20 Marino CT, McDonald E, Romano JF. Seborrheic dermatitis in
dermatitis suggest that patients might have an underlying bar- acquired immunodeficiency syndrome. Cutis 1991; 48:217–18.
rier dysfunction. Further studies investigating the genetics in 21 Binder RL, Jonelis FJ. Seborrheic dermatitis in neuroleptic-induced
parkinsonism. Arch Dermatol 1983; 119:473–5.
patients with this condition might provide additional insights.
22 Mastrolonardo M, Diaferio A, Logroscino G. Seborrheic dermatitis,
increased sebum excretion, and Parkinson’s disease: a survey of
References (im)possible links. Med Hypotheses 2003; 60:907–11.
23 Ikram MA, Brusselle GGO, Murad SD et al. The Rotterdam Study:
1 Gupta AK, Bluhm R. Seborrheic dermatitis. J Eur Acad Dermatol 2018 update on objectives, design and main results. Eur J Epidemiol
2004; 18:13–26. 2017; 32:807–50.
2 Turner GA, Hoptroff M, Harding CR. Stratum corneum dysfunc- 24 Jacobs LC, Hamer MA, Verkouteren JA et al. Perceived skin colour
tion in dandruff. Int J Cosmet Sci 2012; 34:298–306. seems a swift, valid and reliable measurement. Br J Dermatol 2015;
3 Palamaras I, Kyriakis KP, Stavrianeas NG. Seborrheic dermatitis: 173:1084–6.
lifetime detection rates. J Eur Acad Dermatol Venereol 2012; 26:524–6. 25 Flegal KM, Carroll MD, Kit BK et al. Prevalence of obesity and
4 Breunig Jde A, de Almeida HL, Jr., Duquia RP et al. Scalp sebor- trends in the distribution of body mass index among US adults,
rheic dermatitis: prevalence and associated factors in male adoles- 1999-2010. JAMA 2012; 307:491–7.
cents. Int J Dermatol 2012; 51:46–9. 26 Whitworth JA, Chalmers J. World Health Organisation-Interna-
5 Mastrolonardo M, Diaferio A, Vendemiale G et al. Seborrhoeic der- tional Society of Hypertension (WHO/ISH) hypertension guideli-
matitis in the elderly: inferences on the possible role of disability nes. Clin Exp Hypertens 2004; 26:747–52.
and loss of self-sufficiency. Acta Derm Venereol 2004; 84:285–7. 27 Radloff LS. The CES-D Scale: a self-report depression scale for
6 Szepietowski JC, Reich A, Wesolowska-Szepietowska E et al. Qual- research in general population. Appl Psychol Meas 1977; 1:385–401.
ity of life in patients suffering from seborrheic dermatitis: influ- 28 Wittchen HU, Lachner G, Wunderlich U et al. Test-retest reliability
ence of age, gender and education level. Mycoses 2009; 52:357– of the computerized DSM-IV version of the Munich-Composite
63. International Diagnostic Interview (M-CIDI). Soc Psychiatry Psychiatr
7 DeAngelis YM, Gemmer CM, Kaczvinsky JR et al. Three etiologic Epidemiol 1998; 33:568–78.
facets of dandruff and seborrheic dermatitis: Malassezia fungi, 29 Sterne JA, White IR, Carlin JB et al. Multiple imputation for miss-
sebaceous lipids, and individual sensitivity. J Investig Dermatol Symp ing data in epidemiological and clinical research: potential and pit-
Proc 2005; 10:295–7. falls. BMJ 2009; 338:b2393.
8 Johnson MT, Roberts J. Skin Conditions and Related Need for Medical 30 Dessinioti C, Katsambas A. Seborrheic dermatitis: etiology, risk fac-
Care Among Persons 1-74 years. United States, 1971-1974. Vital Health tors, and treatments: facts and controversies. Clin Dermatol 2013;
Statistics, Vol. 11. Washington DC: U.S. Department of Health, 31:343–51.
Education, and Welfare, Public Health Service, Office of the 31 Farage MA, Miller KW, Berardesca E et al. Clinical implications of
Assistant Secretary for Health, National Center for Health Statis- aging skin: cutaneous disorders in the elderly. Am J Clin Dermatol
tics, 1978; 1–72. 2009; 10:73–86.
9 Plunkett A, Merlin K, Gill D et al. The frequency of common non- 32 Schwartz JR, Messenger AG, Tosti A et al. A comprehensive patho-
malignant skin conditions in adults in central Victoria, Australia. physiology of dandruff and seborrheic dermatitis – towards a more
Int J Dermatol 1999; 38:901–8. precise definition of scalp health. Acta Derm Venereol 2013; 93:131–7.
10 Ratzer MA. The incidence of skin diseases in the West of Scotland. 33 Rosner W, Auchus RJ, Azziz R et al. Position statement: utility, lim-
Br J Dermatol 1969; 81:456–61. itations, and pitfalls in measuring testosterone: an Endocrine Soci-
11 Baghestani S, Zare S, Mahboobi AA. Skin disease patterns in Hor- ety position statement. J Clin Endocrinol Metab 2007; 92:405–13.
mozgan, Iran. Int J Dermatol 2005; 44:641–5. 34 Chen W, Mempel M, Traidl-Hofmann C et al. Gender aspects in
12 Linder D, Dreiher J, Zampetti A et al. Seborrheic dermatitis and skin diseases. J Eur Acad Dermatol Venereol 2010; 24:1378–85.
hypertension in adults: a cross-sectional study. J Eur Acad Dermatol 35 Kim MK, Patel RA, Shinn AH et al. Evaluation of gender difference
Venereol 2014; 28:1450–5. in skin type and pH. J Dermatol Sci 2006; 41:153–6.
13 Oztas P, Calikoglu E, Cetin I. Psychiatric tests in seborrhoeic der- 36 Lee CH, Wu SB, Hong CH et al. Molecular mechanisms of UV-
matitis. Acta Derm Venereol 2005; 85:68–9. induced apoptosis and its effects on skin residential cells: the
14 Misery L, Touboul S, Vincot C et al. [Stress and seborrheic dermati- implication in UV-based phototherapy. Int J Mol Sci 2013;
tis]. Ann Dermatol Venereol 2007; 134:833–7 (in French). 14:6414–35.
15 Bae JM, Ha B, Lee H et al. Prevalence of common skin diseases and 37 Wikler JR, Janssen N, Bruynzeel DP et al. The effect of UV-light on
their associated factors among military personnel in Korea: a pityrosporum yeasts: ultrastructural changes and inhibition of
cross-sectional study. J Korean Med Sci 2012; 27:1248–54. growth. Acta Derm Venereol 1990; 70:69–71.
16 Maietta G, Fornaro P, Rongioletti F et al. Patients with mood 38 Man MQ, Lin TK, Santiago JL et al. Basis for enhanced barrier func-
depression have a high prevalence of seborrhoeic dermatitis. Acta tion of pigmented skin. J Invest Dermatol 2014; 134:2399–407.
Derm Venereol 1990; 70:432–4. 39 Brady KT, Randall CL. Gender differences in substance use disor-
17 Bergbrant IM. Seborrhoeic dermatitis: a clinical survey. J Eur Acad ders. Psychiat Clin N Am 1999; 22:241–52.
Dermatol Venereol 1992; 1:285–90.
18 Hancox JG, Sheridan SC, Feldman SR et al. Seasonal variation of
dermatologic disease in the USA: a study of office visits from Supporting Information
1990 to 1998. Int J Dermatol 2004; 43:6–11.
19 Dinis-Oliveira RJ, Magalhaes T, Moreira R et al. Clinical and foren- Additional Supporting Information may be found in the online
sic signs related to ethanol abuse: a mechanistic approach. Toxicol version of this article at the publisher’s website:
Mech Methods 2014; 24:81–110. Powerpoint S1. Journal Club Slide Set.

© 2017 British Association of Dermatologists British Journal of Dermatology (2018) 178, pp148–153