Introduction: As research on the role of the Th17/IL-23 pathway gains importance, the relationship between atopic dermatitis (AD) and psoriasis is becoming elucidated. Objective: The objective of this study wasto evaluate whether AD and its severity affect the risk for psoriasis. Methods: This retrospective population-based study used the database from the 2009 National Health Insurance Services-Health Screening Cohort in Korea. A total of 3,957,922 adult subjects were included and observed until 2018. The primary outcome was newly diagnosed psoriasis. Results: After adjusting for possible confounding factors, the moderate-to-severe AD group had the highest hazard ratio (HR) for psoriasis (HR = 2.50; 95% confidence interval (CI), 2.40–2.61), followed by the mild AD group (HR = 2.31; 95% CI: 2.19–2.44) compared with the non-AD group during a median 8.11 ± 1.19 years of follow-up. Limitations: It is difficult to define AD, which is not standardized, using a claims database and exclude patients who were misdiagnosed with AD. Conclusion: Patients with severe AD showed an increased risk for psoriasis compared to controls, and the risk for psoriasis was increased according to AD severity. This suggests that psoriasis and AD could share inflammatory, immune, and genetic features.

Psoriasis is a chronic inflammatory skin disorder that affects 1–3% of the worldwide population [1] and is known to be associated with an increased risk for various comorbidities, such as cardiometabolic diseases, psychological illness, inflammatory bowel diseases, psoriatic arthritis, and malignancies [2, 3]. Atopic dermatitis (AD) is also a notable chronic inflammatory skin disease that affects up to 25% of children and 2–3% of adults worldwide [4]. AD is associated with an increased risk for comorbidities, such as atopic comorbidities, allergic contact dermatitis, cutaneous/extracutaneous infections, cardiometabolic diseases, and psychological illness [5].

AD and psoriasis have been frequently compared as opposite immune-mediated diseases. AD is driven by the Th2 inflammatory mechanism, whereas psoriasis is a polar Th1/Th17 disease. However, psoriasis and AD may have overlapping clinical phenotypes [6]. Furthermore, these two diseases can co-exist, with their concomitant prevalence ranging from 0.2% to 16.5% [7]. Recently, our research team reported that “atopic clusters,” especially AD, increase the risk for psoriasis. Still, there are a limited number of comparative studies focusing on the risk for psoriasis and AD. Therefore, we aimed to investigate whether AD and its severity affect the risk for psoriasis using a Korean National Health Insurance Service (KNHIS) database.

Data Source

This study is a retrospective population-based cohort study using a database from the 2009 National Health Insurance Services (NHIS) – Health Screening Cohort in Korea. The NHIS, managed by the government, provides complete medical care coverage to 97.1% of the Korean population. We used the NHIS-National Health Screening Cohort (NHIS-HEALS) database which was constructed by the Korean government in 2015 to provide an extensive database for health research. It consists of a complete set of eligibility, claims for medical expenses, and health screening data. Other details are described in an article by Sung et al. [8].

This study was conducted according to the Declaration of Helsinki and approved by the Institutional Review Board of the Catholic University of Korea (KBSMC 2019-09-015). This research used data (NHIS REQ202104382-001) made available by the NHIS. Because of the anonymous and de-identified characteristics of the NHIS-HEALS database, the requirement for informed consent was waived.

Study Design and Population

We extracted a sample cohort from the 2009 NIHS-HEALS. Among the 4,238,822 participants who underwent health screening in 2009, 4,481 participants under the age of 20 years, and 147,657 participants with missing data were excluded. To include only newly diagnosed psoriasis, we excluded subjects who had already been diagnosed with psoriasis before 2009 (n = 104,463). Those who developed psoriasis within a year after AD diagnosis (n = 24,229) were also excluded for the washout periods. A total of 3,957,922 subjects were included in the final analysis and were observed until December 31, 2018. The total follow-up duration was 8.11 ± 1.19 years (mean ± SD). This process is depicted in Figure 1.

Fig. 1.

Study enrollment flow.

Fig. 1.

Study enrollment flow.

Close modal

Definition of AD and Its Severity Classification

All types of claims and diagnoses in the NIHS-HEALS are based on the International Statistical Classification of Disease and Related Health Conditions, 10th revision (ICD-10) code. All participants satisfying the inclusion criteria were divided into three groups: the group without AD, the mild AD group that was treated with one or more topical treatments and had 1 visit coded as L20 (AD in ICD-10) or 3 or more visits coded with L20 in 2009, and a moderate-to-severe AD group that received systemic treatment or phototherapy 12 times or more per year (L20) in 2009.

Primary Outcome and Follow-Up

The primary outcome of this study was newly diagnosed psoriasis, defined as ICD-10 code L40 diagnosis. The diagnosis of psoriasis was defined as more than two visits with designated as L40 or one or more records of hospitalization with the L40 code. The cohort was observed up to the occurrence of the primary outcome, loss of NHIS eligibility (death or migration), or end of the study period (December 31, 2017), whichever came first. At the end of the study, death or migration prior to the primary endpoint was censored (right censored).

Comorbidities and Measurements Performed during Health Screening

The history of common comorbidities (hypertension, diabetes mellitus, and dyslipidemia) was defined using ICD-10-CM codes and medical prescription data. Detailed information about smoking status, alcohol consumption, and physical activity (including the amount and frequency) was obtained from health screening questionnaires. We classified participants based on smoking status as nonsmokers, ex-smokers, and current smokers. We classified alcohol consumption status into three categories: abstinence (no alcoholic drinks within the past year), moderate alcohol consumption (<30 g pure alcohol/day), and heavy alcohol consumption (≥30 g pure alcohol/day). We classified participants’ physical activity as vigorous-moderate (≥1 day per week of moderate- or vigorous-intensity exercise) or absent. At each health screening, venous blood samples after an overnight fast were obtained for the measurement of fasting glucose, lipid profiles, and liver enzyme levels. Information about body mass index (BMI), systolic and diastolic blood pressures, and waist circumference was also obtained during the health screening [9].

Statistical Analysis

The data are presented as the mean ± standard deviation for continuous variables, numbers and percentages for dichotomous variables, and geometric means and 95% confidence intervals (CIs) for continuous variables with skewed distributions. Regarding baseline characteristics, we analyzed continuous variables using analysis of variance and dichotomous variables using the χ2 test. For the primary outcome evaluation, we analyzed the hazard ratio (HR) and 95% CI using the Cox proportional hazards model. The following multivariable-adjusted Cox proportional hazards models were applied: model 1 was non-adjusted; model 2 was adjusted for age and sex; and model 3 was adjusted for age, sex, history of smoking, alcohol consumption status, physical activity, and comorbidities, such as diabetes mellitus, hypertension, and dyslipidemia. We also conducted subgroup analyses according to the following confounding factors: age (20–39, 40–64, or ≥65 years), sex, BMI (<25 kg/m2 or ≥25 kg/m2), smoking status, alcohol consumption status, physical activity, and the presence of comorbidities. Bonferroni correction was used for multiple comparisons among subgroups. All statistical analyses were performed using the SAS software (version 9.4, SAS Institute, Cary, NC, USA) and the R program (version 3.2.4, R Core Team, Vienna, Austria, 2017). p values <0.05 were considered statistically significant.

Baseline Characteristics

The baseline characteristics of the study population are shown according to the presence of AD and its severity in Table 1. Of the total 3,957,922 members of the study population, 53,734 had AD (1.39%). Among them, 21,347 had mild AD (39.98%), and 32,387 had moderate-to-severe AD (60.02%). The mean ages of the non-AD group, mild AD group, and moderate-to-severe AD group were 47 ± 14.07, 48.64 ± 15.33, and 46.54 ± 15.24 years, respectively. The proportions of men in the non-AD group, mild AD group, and moderate-to-severe AD group were 55.09%, 44.63%, and 65.64%, respectively. The proportions of current smokers in the non-AD, mild AD, and moderate-to-severe AD groups were 26.33%, 17.99%, and 21.07%, respectively, the highest being the non-AD group. Conversely, comorbidities, such as hypertension, diabetes mellitus, and dyslipidemia, were highest in the mild AD group.

Table 1.

Baseline characteristics of study population

Subjects, n
Non-AD group (n = 3,904,188)Mild AD group (n = 21,347)Moderate-to-severe AD group (n = 32,387)p value
Age, years 47±14.07 48.64±15.33 46.54±15.24 <0.0001 
 ≥20 and <40 years, n (%) 1,236,850 (31.68) 6,433 (30.14) 11,361 (35.08)  
 ≥40 and <65 years, n (%) 1,855,567 (47.53) 9,012 (42.22) 13,625 (42.07)  
 ≥65 years, n (%) 811,771 (20.79) 5,902 (27.65) 7,401 (22.85)  
Male sex, n (%) 2,150,871 (55.09) 9,528 (44.63) 14,706 (45.41)  
Smoking, n (%)    <0.0001 
 Never 2,317,254 (59.35) 14,403 (67.47) 21,259 (65.64)  
 Ex-smoker 558,987 (14.32) 3,104 (14.54) 4,304 (13.29)  
 Current smoker 1,027,947 (26.33) 3,840 (17.99) 6,824 (21.07)  
Drinking, n (%)    <0.0001 
 No 2,005,115 (51.36) 12,621 (59.12) 18,051 (55.74)  
 <30 g/day 1,587,931 (40.67) 7,533 (35.29) 12,288 (37.94)  
 ≥30 g/day 311,142 (7.97) 1,193 (5.59) 2,048 (6.32)  
Regular physical activity, n (%) 707,172 (18.11) 4,051 (18.98) 5,804 (17.92) 0.0032 
BMI    <0.0001 
 <18.5 kg/m2, n (%) 143,921 (3.69) 898 (4.21) 1,435 (4.43)  
 ≥18.5 and <23 kg/m2, n (%) 1,521,181 (38.96) 8,505 (39.84) 13,377 (41.3)  
 ≥23 and <25 kg/m2, n (%) 962,572 (24.65) 5,057 (23.69) 7,645 (23.61)  
 ≥25 and <30 kg/m2, n (%) 1,137,773 (29.14) 6,140 (28.76) 8,890 (27.45)  
 ≥30 kg/m2, n (%) 138,741 (3.55) 747 (3.5) 1,040 (3.21)  
Comorbidities, n (%) 
 Diabetes mellitus 338,614 (8.67) 2,410 (11.29) 2,549 (7.87) <0.0001 
 Hypertension 1,046,609 (26.81) 6,772 (31.72) 8,522 (26.31) <0.0001 
 Dyslipidemia 705,632 (18.07) 4,557 (21.35) 6,312 (19.49) <0.0001 
Subjects, n
Non-AD group (n = 3,904,188)Mild AD group (n = 21,347)Moderate-to-severe AD group (n = 32,387)p value
Age, years 47±14.07 48.64±15.33 46.54±15.24 <0.0001 
 ≥20 and <40 years, n (%) 1,236,850 (31.68) 6,433 (30.14) 11,361 (35.08)  
 ≥40 and <65 years, n (%) 1,855,567 (47.53) 9,012 (42.22) 13,625 (42.07)  
 ≥65 years, n (%) 811,771 (20.79) 5,902 (27.65) 7,401 (22.85)  
Male sex, n (%) 2,150,871 (55.09) 9,528 (44.63) 14,706 (45.41)  
Smoking, n (%)    <0.0001 
 Never 2,317,254 (59.35) 14,403 (67.47) 21,259 (65.64)  
 Ex-smoker 558,987 (14.32) 3,104 (14.54) 4,304 (13.29)  
 Current smoker 1,027,947 (26.33) 3,840 (17.99) 6,824 (21.07)  
Drinking, n (%)    <0.0001 
 No 2,005,115 (51.36) 12,621 (59.12) 18,051 (55.74)  
 <30 g/day 1,587,931 (40.67) 7,533 (35.29) 12,288 (37.94)  
 ≥30 g/day 311,142 (7.97) 1,193 (5.59) 2,048 (6.32)  
Regular physical activity, n (%) 707,172 (18.11) 4,051 (18.98) 5,804 (17.92) 0.0032 
BMI    <0.0001 
 <18.5 kg/m2, n (%) 143,921 (3.69) 898 (4.21) 1,435 (4.43)  
 ≥18.5 and <23 kg/m2, n (%) 1,521,181 (38.96) 8,505 (39.84) 13,377 (41.3)  
 ≥23 and <25 kg/m2, n (%) 962,572 (24.65) 5,057 (23.69) 7,645 (23.61)  
 ≥25 and <30 kg/m2, n (%) 1,137,773 (29.14) 6,140 (28.76) 8,890 (27.45)  
 ≥30 kg/m2, n (%) 138,741 (3.55) 747 (3.5) 1,040 (3.21)  
Comorbidities, n (%) 
 Diabetes mellitus 338,614 (8.67) 2,410 (11.29) 2,549 (7.87) <0.0001 
 Hypertension 1,046,609 (26.81) 6,772 (31.72) 8,522 (26.31) <0.0001 
 Dyslipidemia 705,632 (18.07) 4,557 (21.35) 6,312 (19.49) <0.0001 

Continuous variables are presented as mean ± SD and categorical variables as number (percentage). For baseline comparisons, Student’s t test or Mann-Whitney U test was used for continuous variables and χ2 test or Fisher’s exact test for categorical variables.

Primary Outcome: Risk for Psoriasis

During the mean follow-up period of 8.11 ± 1.19 years, new-onset psoriasis occurred in 104,801 (3.31/1,000 person-years), 1,307 (7.73/1,000 person-years), and 2,075 (8.10/1,000 person-years) members of the non-AD group, mild AD group, and moderate-to-severe AD group, respectively. After adjusting for possible confounding factors (age, sex, history of smoking, alcohol consumption statuses, physical activity, and comorbidities), the moderate-to-severe AD group had the highest HR for psoriasis (HR = 2.50; 95% CI: 2.40–2.61), followed by the mild AD group (HR = 2.31; 95% CI: 2.19–2.44) compared to the non-AD group (Table 2).

Table 2.

Risk for psoriasis according to presence of AD and its severity

SubgroupsTotal, nPsoriasisFollow-up durationaIncidence ratebModel 1, HR (95% CI)Model 2, HR (95% CI)Model 3, HR (95% CI)
Non-AD group 3,904,188 104,801 31,673,516 3.31 1 (reference) 1 (reference) 1 (reference) 
Mild AD group 21,347 1,307 169,161 7.73 2.34 (2.21–2.47) 2.31 (2.19–2.45) 2.31 (2.19–2.44) 
Moderate-to-severe AD group 32,387 2,075 256,290 8.10 2.45 (2.34–2.56) 2.50 (2.40–2.61) 2.50 (2.40–2.61) 
p value  <0.01 <0.01 <0.01 
SubgroupsTotal, nPsoriasisFollow-up durationaIncidence ratebModel 1, HR (95% CI)Model 2, HR (95% CI)Model 3, HR (95% CI)
Non-AD group 3,904,188 104,801 31,673,516 3.31 1 (reference) 1 (reference) 1 (reference) 
Mild AD group 21,347 1,307 169,161 7.73 2.34 (2.21–2.47) 2.31 (2.19–2.45) 2.31 (2.19–2.44) 
Moderate-to-severe AD group 32,387 2,075 256,290 8.10 2.45 (2.34–2.56) 2.50 (2.40–2.61) 2.50 (2.40–2.61) 
p value  <0.01 <0.01 <0.01 

CI, confidence interval; HR, hazard ratio.

Model 1 is non-adjusted. Model 2 is adjusted for age and sex. Model 3 is adjusted for age, sex, history of smoking, alcohol consumption statuses, physical activity, and comorbidities such as diabetes mellitus, hypertension, and dyslipidemia.

aFollow-up duration is in person-years.

bIncidence rate is presented as 1,000 person-years.

Subgroup Analysis of the Risk for Psoriasis according to the Presence of AD

Among the possible confounding factors, age, sex, and smoking history were found to contribute to the development of psoriasis in patients with AD (Table 3). In the moderate-to-severe group, the risk for psoriasis was significantly higher in the age group between 20 and 40 years (HR = 3.28; 95% CI: 3.05–3.52) compared to the rest of the group (HR = 2.13; 95% CI: 1.98–2.28 compared those between 40 and 65 years old and HR = 2.30; 95% CI: 2.12–2.50 compared to those more than 65 years old). The overall trend between age groups was also observed in the mild AD group. In terms of gender, the risk for psoriasis was higher in men with AD (moderate-to-severe group, HR = 2.88; 95% CI: 2.72–3.06 for males and HR = 2.14; 95% CI: 2–2.28 for females). Regarding smoking history, the risk for psoriasis was higher in current smokers with AD (moderate-to-severe group, HR = 2.91; 95% CI: 2.68–3.17 for current smokers and HR = 2.39; 95% CI: 2.27–2.51 for nonsmokers and ex-smokers). There was no significant interaction in the BMIs, alcohol consumption statuses, regular physical activities, and presence of comorbidities in the subgroups.

Table 3.

Subgroup analysis of risk for psoriasis according to presence of AD

Atopic subgroupsTotal, nPsoriasisFollow-up durationaIncidence ratebHR (95% CI)p value
Age groups 
 ≥20 and <40 years, n (%) Non-AD group 1,236,850 26,867 10,168,335 2.64 1 (reference) <0.01 
Mild 6,433 342 51,919 6.58 2.64 (2.37–2.93) 
Moderate-to-severe 11,361 754 91,110 8.27 3.28 (3.05–3.52) 
 ≥40 and <65 years, n (%) Non 1,855,567 49,626 15,179,366 3.26 1 (reference) 
Mild 9,012 533 72,534 7.34 2.28 (2.09–2.48) 
Moderate-to-severe 13,625 747 109,793 6.80 2.13 (1.98–2.28) 
 ≥65 years, n (%) Non 811,771 28,308 6,325,815 4.47 1 (reference) 
Mild 5,902 432 44,708 9.66 2.12 (1.93–2.34) 
Moderate-to-severe 7,401 574 55,388 10.36 2.30 (2.12–2.50) 
Sex 
 Male Non 2,150,871 60,651 17,354,350 3.49 1 (reference) <0.01 
Mild 9,528 711 74,055 9.60 2.58 (2.40–2.78) 
Moderate-to-severe 14,706 1,170 114,291 10.23 2.88 (2.72–3.06) 
 Female Non 1,753,317 44,150 14,319,166 3.08 1 (reference) 
Mild 11,819 596 95,107 6.26 2.05 (1.89–2.22) 
Moderate-to-severe 17,681 905 142,000 6.37 2.14 (2–2.28) 
BMI 
 <25 kg/m2 Non 2,627,674 68,003 21,302,663 3.192 1 (reference) 0.10 
Mild 14,460 856 114,692 7.463 2.35 (2.2–2.52) 
Moderate-to-severe 22,457 1,419 177,557 7.99 2.58 (2.45–2.72) 
 ≥25 kg/m2 Non 1,276,514 36,798 10,370,854 3.54 1 (reference) 
Mild 6,887 451 54,469 8.27 2.25 (2.05–2.46) 
Moderate-to-severe 9,930 656 78,734 8.33 2.35 (2.17–2.54) 
Smoking history 
 Nonsmoker and ex-smoker Non 2,876,241 75,582 23,375,933 3.23 1 (reference) <0.01 
Mild 17,507 1,024 139,168 7.35 2.25 (2.12–2.4) 
Moderate-to-severe 25,563 1,528 202,971 7.52 2.39 (2.27–2.51) 
 Current smoker Non 1,027,947 29,219 8,297,583 3.52 1 (reference) 
Mild 3,840 283 29,994 9.43 2.58 (2.29–2.90) 
Moderate-to-severe 6,824 547 53,320 10.25 2.91 (2.68–3.17) 
Drinking 
 <30 g/day Non 3,593,046 95,782 29,163,151 3.28 1 (reference) 0.27 
Mild 20,154 1,234 159,639 7.72 2.33 (2.20–2.46) 
Moderate-to-severe 30,339 1,918 240,279 7.98 2.48 (2.37–2.6) 
 ≥30 g/day Non 311,142 9,019 2,510,365 3.59 1 (reference) 
Mild 1,193 73 9,523 7.66 2.09 (1.66–2.63) 
Moderate-to-severe 2,048 157 16,012 9.80 2.78 (2.38–3.26) 
Regular physical activity 
 Vigorous-moderate Non 3,197,016 84,995 25,920,982 3.27 1 (reference) 0.39 
Mild 17,296 1,025 137,090 7.47 2.27 (2.13–2.41) 
Moderate-to-severe 26,583 1,679 210,247 7.98 2.5 (2.38–2.62) 
 Mild-none Non 707,172 19,806 5,752,534 3.44 1 (reference) 
Mild 4,051 282 32,072 8.79 2.49 (2.21–2.8) 
Moderate-to-severe 5,804 396 46,044 8.60 2.53 (2.29–2.79) 
Diabetes mellitus 
 Yes Non 3,565,574 93,681 29,020,425 3.22 1 (reference) 0.33 
Mild 18,937 1,130 151,016 7.48 2.32 (2.19–2.46) 
Moderate-to-severe 29,838 1,890 236,981 7.97 2.53 (2.42–2.648) 
 No Non 338,614 11,120 2,653,091 4.19 1 (reference) 
Mild 2,410 177 18,146 9.75 2.27 (1.96–2.64) 
Moderate-to-severe 2,549 185 19,310 9.58 2.26 (1.95–2.61) 
Hypertension 
 Yes Non 2,857,579 72,408 23,347,082 3.10 1 (reference) 0.07 
Mild 14,575 859 116,929 7.34 2.42 (2.26–2.59) 
Moderate-to-severe 23,865 1,449 190,753 7.59 2.54 (2.41–2.67) 
 No Non 1,046,609 32,393 8,326,434 3.89 1 (reference) 
Mild 6,772 448 52,233 8.57 2.13 (1.94–2.34) 
Moderate-to-severe 8,522 626 65,537 9.55 2.43 (2.25–2.63) 
Dyslipidemia 
 Yes Non 3,198,556 83,063 25,983,230 3.19 1 (reference) 0.31 
Mild 16,790 994 133,449 7.44 2.35 (2.20–2.5 
Moderate-to-severe 26,075 1,625 206,925 7.85 2.54 (2.42–2.67) 
 No Non 705,632 21,738 5,690,286 3.82 1 (reference) 
Mild 4,557 313 35,712 8.76 2.21 (1.98–2.47) 
Moderate-to-severe 6,312 450 49,366 9.11 2.38 (2.16–2.61) 
Atopic subgroupsTotal, nPsoriasisFollow-up durationaIncidence ratebHR (95% CI)p value
Age groups 
 ≥20 and <40 years, n (%) Non-AD group 1,236,850 26,867 10,168,335 2.64 1 (reference) <0.01 
Mild 6,433 342 51,919 6.58 2.64 (2.37–2.93) 
Moderate-to-severe 11,361 754 91,110 8.27 3.28 (3.05–3.52) 
 ≥40 and <65 years, n (%) Non 1,855,567 49,626 15,179,366 3.26 1 (reference) 
Mild 9,012 533 72,534 7.34 2.28 (2.09–2.48) 
Moderate-to-severe 13,625 747 109,793 6.80 2.13 (1.98–2.28) 
 ≥65 years, n (%) Non 811,771 28,308 6,325,815 4.47 1 (reference) 
Mild 5,902 432 44,708 9.66 2.12 (1.93–2.34) 
Moderate-to-severe 7,401 574 55,388 10.36 2.30 (2.12–2.50) 
Sex 
 Male Non 2,150,871 60,651 17,354,350 3.49 1 (reference) <0.01 
Mild 9,528 711 74,055 9.60 2.58 (2.40–2.78) 
Moderate-to-severe 14,706 1,170 114,291 10.23 2.88 (2.72–3.06) 
 Female Non 1,753,317 44,150 14,319,166 3.08 1 (reference) 
Mild 11,819 596 95,107 6.26 2.05 (1.89–2.22) 
Moderate-to-severe 17,681 905 142,000 6.37 2.14 (2–2.28) 
BMI 
 <25 kg/m2 Non 2,627,674 68,003 21,302,663 3.192 1 (reference) 0.10 
Mild 14,460 856 114,692 7.463 2.35 (2.2–2.52) 
Moderate-to-severe 22,457 1,419 177,557 7.99 2.58 (2.45–2.72) 
 ≥25 kg/m2 Non 1,276,514 36,798 10,370,854 3.54 1 (reference) 
Mild 6,887 451 54,469 8.27 2.25 (2.05–2.46) 
Moderate-to-severe 9,930 656 78,734 8.33 2.35 (2.17–2.54) 
Smoking history 
 Nonsmoker and ex-smoker Non 2,876,241 75,582 23,375,933 3.23 1 (reference) <0.01 
Mild 17,507 1,024 139,168 7.35 2.25 (2.12–2.4) 
Moderate-to-severe 25,563 1,528 202,971 7.52 2.39 (2.27–2.51) 
 Current smoker Non 1,027,947 29,219 8,297,583 3.52 1 (reference) 
Mild 3,840 283 29,994 9.43 2.58 (2.29–2.90) 
Moderate-to-severe 6,824 547 53,320 10.25 2.91 (2.68–3.17) 
Drinking 
 <30 g/day Non 3,593,046 95,782 29,163,151 3.28 1 (reference) 0.27 
Mild 20,154 1,234 159,639 7.72 2.33 (2.20–2.46) 
Moderate-to-severe 30,339 1,918 240,279 7.98 2.48 (2.37–2.6) 
 ≥30 g/day Non 311,142 9,019 2,510,365 3.59 1 (reference) 
Mild 1,193 73 9,523 7.66 2.09 (1.66–2.63) 
Moderate-to-severe 2,048 157 16,012 9.80 2.78 (2.38–3.26) 
Regular physical activity 
 Vigorous-moderate Non 3,197,016 84,995 25,920,982 3.27 1 (reference) 0.39 
Mild 17,296 1,025 137,090 7.47 2.27 (2.13–2.41) 
Moderate-to-severe 26,583 1,679 210,247 7.98 2.5 (2.38–2.62) 
 Mild-none Non 707,172 19,806 5,752,534 3.44 1 (reference) 
Mild 4,051 282 32,072 8.79 2.49 (2.21–2.8) 
Moderate-to-severe 5,804 396 46,044 8.60 2.53 (2.29–2.79) 
Diabetes mellitus 
 Yes Non 3,565,574 93,681 29,020,425 3.22 1 (reference) 0.33 
Mild 18,937 1,130 151,016 7.48 2.32 (2.19–2.46) 
Moderate-to-severe 29,838 1,890 236,981 7.97 2.53 (2.42–2.648) 
 No Non 338,614 11,120 2,653,091 4.19 1 (reference) 
Mild 2,410 177 18,146 9.75 2.27 (1.96–2.64) 
Moderate-to-severe 2,549 185 19,310 9.58 2.26 (1.95–2.61) 
Hypertension 
 Yes Non 2,857,579 72,408 23,347,082 3.10 1 (reference) 0.07 
Mild 14,575 859 116,929 7.34 2.42 (2.26–2.59) 
Moderate-to-severe 23,865 1,449 190,753 7.59 2.54 (2.41–2.67) 
 No Non 1,046,609 32,393 8,326,434 3.89 1 (reference) 
Mild 6,772 448 52,233 8.57 2.13 (1.94–2.34) 
Moderate-to-severe 8,522 626 65,537 9.55 2.43 (2.25–2.63) 
Dyslipidemia 
 Yes Non 3,198,556 83,063 25,983,230 3.19 1 (reference) 0.31 
Mild 16,790 994 133,449 7.44 2.35 (2.20–2.5 
Moderate-to-severe 26,075 1,625 206,925 7.85 2.54 (2.42–2.67) 
 No Non 705,632 21,738 5,690,286 3.82 1 (reference) 
Mild 4,557 313 35,712 8.76 2.21 (1.98–2.47) 
Moderate-to-severe 6,312 450 49,366 9.11 2.38 (2.16–2.61) 

CI, confidence interval; HR, hazard ratio.

aFollow-up duration is in person-years.

bIncidence rate is presented as 1,000 person-years.

Kaplan-Meier analysis showed that the risk of psoriasis increased alongside the duration of AD compared to the control groups without AD (Fig. 2a). The slope of the Kaplan-Meier graph was steeper in the group with severe AD compared to the mild AD group (Fig. 2b) (p < 0.001 for all).

Fig. 2.

Risk of psoriasis in patients with AD. Kaplan-Meier plots of the incidence of psoriasis. a The incidence of psoriasis was significantly higher in patients with AD compared to control groups without AD. b The slope of the Kaplan-Meier graph was steeper in the group with severe AD compared to the mild AD group (p < 0.001). The log-rank test was used to analyze the differences between the study and control groups.

Fig. 2.

Risk of psoriasis in patients with AD. Kaplan-Meier plots of the incidence of psoriasis. a The incidence of psoriasis was significantly higher in patients with AD compared to control groups without AD. b The slope of the Kaplan-Meier graph was steeper in the group with severe AD compared to the mild AD group (p < 0.001). The log-rank test was used to analyze the differences between the study and control groups.

Close modal

This study used the Korean NIHS-HEALS database to investigate the subsequent risk for psoriasis in patients with AD. It investigated the risk for psoriasis according to the severity of AD to examine the relationship between the two diseases. The risk for psoriasis was increased in patients with AD to such an extent that the significant HR was more than 2. Moreover, we demonstrated that as the severity of AD increases, the risk for psoriasis increases. Recently, our study group reported on the relationship between allergic disease (AD, allergic rhinitis, and asthma) and the risk for psoriasis [10], which was the first epidemiologic study, to the best of our knowledge, to elucidate the impact of allergic disease on psoriasis. The present study was a follow-up on our previous work [10].

Psoriasis and AD are representative immunological skin disorders mediated by T helper cells as consequences of the interactions between genetic and environmental factors [11]. Conventionally, psoriasis and AD are considered opposite immune-mediated disorders. Psoriasis is driven by Th1 disease, whereas AD is considered to be a polar Th2 disease. However, the Th1/Th2 paradigm has been challenged by the discovery of additional helper T cell subsets, Th17 and Th22. Th17 cells, differentiated from Th1 or Th2 cells, mainly produce IL-17 and IL-22. IL-17, well known to play an important role in AD and psoriasis, activates macrophages to express tumor necrosis factor-α and IL-1β and fibroblasts to produce IL-6, IL-8, and matrix metalloproteinases [12]. These cytokines can induce tissue inflammation and remodeling. In psoriasis, the Th-17/IL23 axis, mainly depending on Th17 cell function, is considered a crucial pathogenesis of this disease [13, 14]. Acute AD is characterized by Th2 cells almost exclusively, while the Th1 and Th17/IL-23 pathways, as well as Th2 cells, play a major role in chronic AD [15]. Guttman-Yassky et al. [16] reported that the activation of the Th17/IL-23 pathway is a crucial pathogenesis of chronic AD and psoriasis, and systemic activation of this inflammatory pathway could explain the occurrence of non-cutaneous comorbidities associated with increased morbidity and disability that affect patients with AD and psoriasis.

Several subtypes of AD have been reported to have prominent IL-17 components and tissue patterning that overlap with distinctive psoriasis histopathology. Lesioned skin of early-onset pediatric AD patients, compared to lesioned skin of adult AD patients, shows greater epidermal hyperplasia and a significant increase in the activation of Th17-related cytokines and antimicrobials, including IL-17A, in addition to Th2 activation [17]. In a study within European-American populations, a significant increase in Th17-related cytokines was found in the intrinsic AD group compared to the extrinsic AD group, and Th17-related cytokine levels were positively correlated with disease severity [18]. Lastly, Noda et al. [19] reported that the phenotype of Asian AD showed increased hyperplasia, parakeratosis, and higher Th17 and Th22 activation, as well as Th2 activation compared to other phenotypes. The high Th17 phenotype, which is more common in South Korean AD patients, may affect the significantly elevated risk for psoriasis observed in our results. Furthermore, our study showed the young age group (those ≥20 and <40 years old) with moderate-to-severe AD had a higher risk for psoriasis compared to those without AD (HR = 3.28; 95% CI: 2.40–2.61), which agrees with the results of the above studies.

In addition to T helper cells, the pathogeneses of psoriasis and AD have several common features. Choy et al. reported that genes encoding neutrophilic chemoattractants were highly expressed in AD in their gene expression microarray analyses, comparable with those in psoriasis lesions [20]. Cookson et al. [21] reported that, in their genome screening analysis, the chromosomes 1q21, 17q25, and 20p are associated with childhood AD, and these regions correlate closely with psoriasis loci, which are associated with dermal inflammation and immunity. In addition, a genome-wide association study revealed that several SNPs, including a damaging missense variant in the CARD14 gene, inducing the keratinocyte response to inflammatory cytokines by activating the nuclear factor-kappa B pathway, are found in both diseases [22]. Cytokines produced during type 2 inflammation in AD or type 17 inflammation in psoriasis form inflammatory loops in the epidermis and dermis and further result in inflammatory skin disease [23]. Environmental factors, such as smoking, psychological stress, and dryness, also act as aggravating factors in both diseases [24]. Taken together, inflammatory, immune, genetic, and environmental factors might contribute to the elevated risk for psoriasis according to AD.

An observational study of concomitant AD and psoriasis has also been reported. Patients with concomitant AD and PS were frequently male [25], and our results also showed a significantly higher risk for psoriasis in men. Also, patients with concomitant AD and psoriasis had higher serum concentrations of IL-17, suggesting that the Th17 pathway may play an important role in the two diseases [25].

There have been several reports suggesting the potential occurrence of psoriasis when using dupilumab. However, in Korea, dupilumab was recognized as a reimbursed medication in 2020, and prescription under insurance coverage is required for inclusion in the NHIS database, enabling further analysis. Our analysis was based on preexisting data before the introduction of dupilumab. Prior to the commercialization of dupilumab, various treatments for AD included topical corticosteroids, systemic corticosteroids, cyclosporin, methotrexate, and NBUVB, which are also commonly used in the treatment of psoriasis. Hence, it is challenging to definitively attribute the development of psoriasis solely to these treatments.

This study had several strengths, including the large size of the population-based sample cohort and the multiple-year follow-up. However, several limitations also exist. First, since this study was designed using a claims database, the AD definition is difficult to standardize. To increase the diagnostic accuracy, we defined the AD group as having one or more treatments with 1 visit designated as L20 or 3 or more visits designated as L20 per year, as in the previous studies [26‒28]. The prevalence of adult AD in this study was 1.36%, similar to that reported in the previous study, 0.9%, using the Korean claims database [29]. Second, although AD and psoriasis can be distinguished clinically, this study did not exclude patients who were misdiagnosed with AD and later diagnosed with psoriasis. We inadvertently overlooked the clinical differentiation between AD and psoriasis. These two skin conditions often manifest overlapping symptoms and characteristics. Additionally, the ambiguous nature of conditions like “psoriasiform dermatitis” and “eczematous psoriasis” creates a small yet significant gray area, potentially leading to misclassification within general databases. Finally, the observational nature of the study limits causal inference with a possibility of residual confounding.

In conclusion, patients with AD showed an increased risk for psoriasis compared to controls, and the risk for psoriasis was raised according to the severity of AD. Moreover, this tendency was more pronounced in younger patients and men. This suggests that psoriasis and AD might share some inflammatory, immune, and genetic features.

The authors have no ethical conflicts to disclose. This study was approved by the Institutional Review Board of the Catholic University of Korea (KBSMC 2019-09-015) and used data (NHIS REQ202104382-001) made available by the NHIS. Because of the anonymous and de-identified characteristics of the NHIS-HEALS database, the requirement for informed consent was waived.

The authors have no conflicts of interest to declare.

No funding was received for this study.

H.K. was responsible for data analysis and wrote the first draft of the manuscript. L.J.H. and Y.J.H. contributed to the design of the study and interpretation of data. H.J.L. and L.J.H. contributed to the writing of the manuscript. All co-authors provided important intellectual input and approved the final version of the manuscript.

Raw data were publicly available and were generated from NHIS-National Health Screening Cohort (NHIS-HEALS) database. Derived data supporting the findings of this study are available on request. Further inquiries can be directed to the corresponding author.

1.
Michalek
IM
,
Loring
B
,
John
SM
.
A systematic review of worldwide epidemiology of psoriasis
.
J Eur Acad Dermatol Venereol
.
2017
;
31
(
2
):
205
12
. .
2.
Takeshita
J
,
Grewal
S
,
Langan
SM
,
Mehta
NN
,
Ogdie
A
,
Van Voorhees
AS
, et al
.
Psoriasis and comorbid diseases: epidemiology
.
J Am Acad Dermatol
.
2017
;
76
(
3
):
377
90
. .
3.
Armstrong
AW
,
Read
C
.
Pathophysiology, clinical presentation, and treatment of psoriasis: a review
.
JAMA
.
2020
;
323
(
19
):
1945
60
. .
4.
Eichenfield
LF
,
Tom
WL
,
Chamlin
SL
,
Feldman
SR
,
Hanifin
JM
,
Simpson
EL
, et al
.
Guidelines of care for the management of atopic dermatitis: section 1. Diagnosis and assessment of atopic dermatitis
.
J Am Acad Dermatol
.
2014
;
70
(
2
):
338
51
. .
5.
Silverberg
JI
.
Comorbidities and the impact of atopic dermatitis
.
Ann Allergy Asthma Immunol
.
2019
;
123
(
2
):
144
51
. .
6.
Egeberg
A
,
Griffiths
CEM
,
Williams
HC
,
Andersen
YMF
,
Thyssen
JP
.
Clinical characteristics, symptoms and burden of psoriasis and atopic dermatitis in adults
.
Br J Dermatol
.
2020
;
183
(
1
):
128
38
. .
7.
Barry
K
,
Zancanaro
P
,
Casseres
R
,
Abdat
R
,
Dumont
N
,
Rosmarin
D
.
Concomitant atopic dermatitis and psoriasis - a retrospective review
.
J Dermatolog Treat
.
2021
;
32
(
7
):
716
20
. .
8.
Seong
SC
,
Kim
YY
,
Park
SK
,
Khang
YH
,
Kim
HC
,
Park
JH
, et al
.
Cohort profile: the national health insurance service-national health screening cohort (NHIS-heals) in Korea
.
BMJ Open
.
2017
;
7
(
9
):
e016640
. .
9.
Wen
CP
,
David Cheng
TY
,
Tsai
SP
,
Chan
HT
,
Hsu
HL
,
Hsu
CC
, et al
.
Are Asians at greater mortality risks for being overweight than Caucasians? Redefining obesity for Asians
.
Public Health Nutr
.
2009
;
12
(
4
):
497
506
. .
10.
Han
JH
,
Bang
CH
,
Han
K
,
Ryu
JY
,
Lee
JY
,
Park
YM
, et al
.
The risk of psoriasis in patients with allergic diseases: a nationwide population-based cohort study
.
Allergy Asthma Immunol Res
.
2021
;
13
(
4
):
638
45
. .
11.
Eyerich
S
,
Onken
AT
,
Weidinger
S
,
Franke
A
,
Nasorri
F
,
Pennino
D
, et al
.
Mutual antagonism of T cells causing psoriasis and atopic eczema
.
N Engl J Med
.
2011
;
365
(
3
):
231
8
. .
12.
Sugaya
M
.
The role of Th17-related cytokines in atopic dermatitis
.
Int J Mol Sci
.
2020
;
21
(
4
):
1314
. .
13.
Di Cesare
A
,
Di Meglio
P
,
Nestle
FO
.
The IL-23/Th17 axis in the immunopathogenesis of psoriasis
.
J Invest Dermatol
.
2009
;
129
(
6
):
1339
50
. .
14.
Leonardi
C
,
Matheson
R
,
Zachariae
C
,
Cameron
G
,
Li
L
,
Edson-Heredia
E
, et al
.
Anti-interleukin-17 monoclonal antibody ixekizumab in chronic plaque psoriasis
.
N Engl J Med
.
2012
;
366
(
13
):
1190
9
. .
15.
Langan
SM
,
Irvine
AD
,
Weidinger
S
.
Atopic dermatitis
.
Lancet
.
2020
;
396
(
10247
):
345
60
. .
16.
Guttman-Yassky
E
,
Krueger
JG
,
Lebwohl
MG
.
Systemic immune mechanisms in atopic dermatitis and psoriasis with implications for treatment
.
Exp Dermatol
.
2018
;
27
(
4
):
409
17
. .
17.
Esaki
H
,
Brunner
PM
,
Renert-Yuval
Y
,
Czarnowicki
T
,
Huynh
T
,
Tran
G
, et al
.
Early-onset pediatric atopic dermatitis is T(H)2 but also T(H)17 polarized in skin
.
J Allergy Clin Immunol
.
2016
;
138
(
6
):
1639
51
. .
18.
Suárez-Fariñas
M
,
Dhingra
N
,
Gittler
J
,
Shemer
A
,
Cardinale
I
,
de Guzman Strong
C
, et al
.
Intrinsic atopic dermatitis shows similar TH2 and higher TH17 immune activation compared with extrinsic atopic dermatitis
.
J Allergy Clin Immunol
.
2013
;
132
(
2
):
361
70
. .
19.
Noda
S
,
Suárez-Fariñas
M
,
Ungar
B
,
Kim
SJ
,
de Guzman Strong
C
,
Xu
H
, et al
.
The Asian atopic dermatitis phenotype combines features of atopic dermatitis and psoriasis with increased TH17 polarization
.
J Allergy Clin Immunol
.
2015
;
136
(
5
):
1254
64
. .
20.
Choy
DF
,
Hsu
DK
,
Seshasayee
D
,
Fung
MA
,
Modrusan
Z
,
Martin
F
, et al
.
Comparative transcriptomic analyses of atopic dermatitis and psoriasis reveal shared neutrophilic inflammation
.
J Allergy Clin Immunol
.
2012
;
130
(
6
):
1335
43.e5
. .
21.
Cookson
WO
,
Ubhi
B
,
Lawrence
R
,
Abecasis
GR
,
Walley
AJ
,
Cox
HE
, et al
.
Genetic linkage of childhood atopic dermatitis to psoriasis susceptibility loci
.
Nat Genet
.
2001
;
27
(
4
):
372
3
. .
22.
Weidinger
S
,
Willis-Owen
SA
,
Kamatani
Y
,
Baurecht
H
,
Morar
N
,
Liang
L
, et al
.
A genome-wide association study of atopic dermatitis identifies loci with overlapping effects on asthma and psoriasis
.
Hum Mol Genet
.
2013
;
22
(
23
):
4841
56
. .
23.
Dainichi
T
,
Kitoh
A
,
Otsuka
A
,
Nakajima
S
,
Nomura
T
,
Kaplan
DH
, et al
.
The epithelial immune microenvironment (EIME) in atopic dermatitis and psoriasis
.
Nat Immunol
.
2018
;
19
(
12
):
1286
98
. .
24.
Griffiths
CE
,
van de Kerkhof
P
,
Czarnecka-Operacz
M
.
Psoriasis and atopic dermatitis
.
Dermatol Ther
.
2017
;
7
(
Suppl 1
):
31
41
. .
25.
Bozek
A
,
Zajac
M
,
Krupka
M
.
Atopic dermatitis and psoriasis as overlapping syndromes
.
Mediators Inflamm
.
2020
;
2020
:
7527859
. .
26.
Rhee
TM
,
Choi
EK
,
Han
KD
,
Lee
SR
,
Oh
S
.
Impact of the combinations of allergic diseases on myocardial infarction and mortality
.
J Allergy Clin Immunol Pract
.
2021
;
9
(
2
):
872
80.e4
. .
27.
Choi
YJ
,
Choi
EK
,
Han
KD
,
Kwon
S
,
Lee
SY
,
Yang
S
, et al
.
Increased risk of atrial fibrillation in patients with atopic triad: a nationwide population-based study
.
J Allergy Clin Immunol Pract
.
2021
;
9
(
9
):
3422
30.e5
. .
28.
Paller
AS
,
Siegfried
EC
,
Vekeman
F
,
Gadkari
A
,
Kaur
M
,
Mallya
UG
, et al
.
Treatment patterns of pediatric patients with atopic dermatitis: a claims data analysis
.
J Am Acad Dermatol
.
2020
;
82
(
3
):
651
60
. .
29.
Yu
JS
,
Lee
CJ
,
Lee
HS
,
Kim
J
,
Han
Y
,
Ahn
K
, et al
.
Prevalence of atopic dermatitis in Korea: analysis by using national statistics
.
J Korean Med Sci
.
2012
;
27
(
6
):
681
5
. .