- Email: [email protected]
Low prevalence of psoriasis among children and adolescents in a large multiethnic cohort in southern California
Low prevalence of psoriasis among children and adolescents in a large multiethnic cohort in southern California Jashin J. Wu, MD,a,b Mary Helen Black, MS, PhD,a Ning Smith, PhD,a Amy H. Porter, MD,c Steven J. Jacobsen, MD, PhD,a and Corinna Koebnick, MS, PhDa Pasadena, Los Angeles, and Baldwin Park, California Background: Little information is available on the prevalence of psoriasis in children and adolescents. Objective: We sought to estimate the prevalence of pediatric psoriasis in southern California and to investigate the validity of psoriasis diagnosis by a dermatologist compared with a nondermatologist. Methods: In a southern California population of 710,949 children who were enrolled in an integrated prepaid health plan in 2007 through 2008, cases of psoriasis were identified from electronic medical records and validated by medical chart review. Positive predictive values for valid diagnosis were reported for dermatologists and nondermatologists. Results: The prevalence of pediatric psoriasis confirmed by medical chart review was 19/10,000 patients. The prevalence of psoriasis diagnosis (confirmed and unconfirmed) was 30/10,000 patients. The age at onset of psoriasis was slightly earlier in boys than in girls. The positive predictive value for a valid diagnosis of psoriasis was 63.7% when the diagnosis was made by any health care provider, 90.0% by a dermatologist, and 26.6% by a nondermatologist. The prevalence of psoriasis was higher in girls than in boys. Psoriasis affected 29 (95% confidence interval [CI] 27-32) non-Hispanic whites, 20 (95% CI 16-24) Asian/Pacific Islanders, 16 (95% CI 15-18) Hispanic whites, and 6 (95% CI 4-9) blacks per 10,000 patients. Limitations: Information on the age at onset was estimated based on the first documented diagnosis of psoriasis. Conclusion: The overall prevalence of pediatric psoriasis was lower compared with other published studies. This could be in part a result of underdiagnosis because of greater sunlight exposure in southern California and a lower proportion of non-Hispanic whites in the population. ( J Am Acad Dermatol 2011;65:957-64.) Key words: adolescence; childhood; epidemiology; prevalence; psoriasis; treatment; validation.
cant information is available on the prevalence of psoriasis in children. Although a recent study suggested that childhood onset of psoriasis is not associated with disease severity,1 early
S
onset may result in longer exposure to a chronic inflammatory condition and, thus, may affect the morbidity and mortality risk. Previous studies found prevalence estimates of pediatric psoriasis ranging
From the Department of Research and Evaluation, Kaiser Permanente Southern California, Pasadenaa; Department of Dermatology, Kaiser Permanente Los Angeles Medical Centerb; and Department of Pediatrics, Baldwin Park Medical Center, Southern California Permanente Medical Group.c Supported by Kaiser Permanente Direct Community Benefit Funds. Disclosure: Dr Wu received research funding from Abbott Laboratories and Amgen. Dr Jacobsen received research grants from Merck and Beckman-coulter. He also served as consultant for
Merck without compensation. Drs Black, Smith, Porter, and Koebnick have no conflicts of interest to declare. Accepted for publication September 7, 2010. Reprint requests: Jashin J. Wu, MD, Department of Dermatology, Kaiser Permanente Los Angeles Medical Center, 1515 N Vermont Ave, 5th Floor, Los Angeles, CA 90027. E-mail: [email protected]. Published online February 9, 2011. 0190-9622/$36.00 ª 2010 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2010.09.005
957
J AM ACAD DERMATOL
958 Wu et al
NOVEMBER 2011
from 0.5% to 1.4%.2,3 Onset during the first 2 decades extracting ICD-9 code 696.1 (psoriasis) from elecof life is reported by 31% to 45% of affected tronic medical records for all inpatient and outpaadults.1,4-6 tient encounters from enrollment in the health plan Large epidemiologic studies often rely on diagnoto enrollment in the current studyethese cases were sis codes3,7 or self-reporting8 for the identification of subdivided into whether or not the diagnosis of cases. A recent report in adult psoriasis suggests that psoriasis was made by a dermatologist; (2) by case identification based solely on the International extracting ICD-9 code 696.0 (psoriatic arthritis) Classification of Diseases, from electronic medical recNinth Revision (ICD-9) codes ords; and (3) by performing a CAPSULE SUMMARY may lead to misclassificasearch for psoriasis-specific tion.9 Only 73.5% of psoriasis medications and treatment. The prevalence of psoriasis in children cases identified in this study Medications were limited to and adolescents in southern California were confirmed as prevalent those that are very specific was found to be low compared with or incident psoriasis after for psoriasis treatment or other countries. manual review of patient exclusively approved for psocharts by a dermatologist. The prevalence of psoriasis was higher in riasis treatment. PsoriasisNo information is available girls than in boys and varied by specific medications were on the validity of diagnosis race/ethnicity with the highest defined as evidence of a precodes for psoriasis in chilprevalence in non-Hispanic whites and scription for one of the foldren. This information is the lowest in blacks. lowing topical therapies needed to provide valid prev(anthralin, liquor carbonis Based on diagnostic codes, more than alence estimates for psoriasis. distillate, tar, and 2 or more 90% of psoriasis cases diagnosed by a We investigated the prevaprescriptions of calcipotriene dermatologist could be confirmed after lence, disparities in the or calcitriol) and systemic an independent medical chart review prevalence, and treatment of therapies (acitretin, alefacept, compared with 27% diagnosed by psoriasis in a multiethnic efalizumab, or ustekinumab). nondermatologists. population-based cohort of For most medications, pamore than 710,000 children tients with at least one preaged 2 to 19 years10 in southern California who were scription were considered as potential cases of enrolled in an integrated prepaid health plan in 2007 psoriasis for manual chart review; for calcipotriene through 2008. We also examined the validity of psoand calcitriol, at least two prescriptions were needed. riasis diagnosis and devised a comprehensive search To provide information on treatment of psoriasis, algorithm to detect patients with uncoded psoriasis. we identified patients who were prescribed topical treatments (steroid and nonsteroid), phototherapy, METHODS traditional systemic medications (acitretin, cycloStudy design and subjects sporine, methotrexate), other less commonly used For the Kaiser Permanente Southern California systemic agents (azathioprine, mycophenolate mo(KPSC) Children’s Health Study, a cohort of children fetil, hydroxyurea), biologics (tumor necrosis factoraged 2 to 19 years (n = 710,949) was established at alfa inhibitors, T-cell inhibitors, or interleukin-12/23 KPSC, which provided health care to over 3.2 million inhibitors), or phototherapy (ultraviolet B or psoramembers throughout the 7-county region in the len plus ultraviolet A).12 11 years 2007 through 2008. The KPSC population is ethnically diverse and largely representative of Case validation southern California, and includes approximately Potential psoriasis cases identified by ICD-9 code or 14% of the population of the state of California.11 treatment search were adjudicated by chart review. Inclusion criteria were at least one body weight and Patients who were given a diagnosis by a primary care height measurement in the electronic medical record provider, were younger than 5 years of age, had a during 2007 through 2008. Pregnant girls and young diagnosis of psoriatic arthritis, or were prescribed women were excluded from the study. The study psoriasis-specific medications, were reviewed by a protocol was reviewed and approved by the institudermatologist (J. J. W., n = 1082). Potential patients tional review board of KPSC before implementation. with psoriasis identified by ICD-9 code older than 5 years of age and with a diagnosis made at least once by Outcome ascertainment and treatment a dermatologist (n = 1037) were reviewed by trained Prevalent cases of psoriasis were identified by 3 research staff for the following terms: ‘‘silvery,’’ different search strategies (n = 2287) (Fig 1): (1) by ‘‘flaky,’’ ‘‘papulosquamous,’’ ‘‘red,’’ ‘‘papules,’’ and d
d
d
J AM ACAD DERMATOL
Wu et al 959
VOLUME 65, NUMBER 5
Abbreviations used: CI: ICD-9: KPSC: PPV:
confidence interval International Classification of Diseases, Ninth Revision Kaiser Permanente Southern California positive predictive value
‘‘plaques’’ in the following body locations: ‘‘knees’’ and ‘‘elbows.’’ Any unclear or questionable records were referred to a dermatologist for further review. Race/ethnicity Race and ethnicity information were obtained from administrative records supplemented by birth certificate information. We categorized race/ethnicity as non-Hispanic white, Hispanic white, black (regardless of ethnicity), Asian or Pacific Islander, other or multiple race/ethnicity, and unknown because of missing information (52.5%). For unknown race and ethnicity information, administrative records were supplemented by an imputation algorithm based on surname lists and address information derived from the US Census Bureau.13,14 Hispanic ethnicity and Asian race were assigned based on surnames. For blacks and non-Hispanic whites, additional geocoding information based on member address was linked to racial/ethnic information of the US Census Bureau data. With a hierarchical assignment order, race/ethnicity was assigned using probability cut-offs more than 50% for Asian surname, more than 50% for Hispanic surname, more than 75% for black race from geocoding if probability for Asian surname was less than 50% (Hispanic blacks were assigned to black race for this study), and white race more than 45% from geocoding if no other assignment could be made before. The specificity and positive predictive values (PPV) were more than 98% for all races/ethnicities.11 Statistical analysis Prevalence of psoriasis was calculated per 10,000 patients with corresponding 95% confidence intervals (CIs) assuming that the occurrence of psoriasis follows a Poisson distribution. Sensitivity for psoriasis diagnosis was defined as the number of confirmed psoriasis cases identified by a particular search strategy divided by the total number of confirmed psoriasis cases. Sensitivities were estimated using cases that were identified by ICD-9 code during a period of 24 months before the study. PPV were defined as the proportion of confirmed cases to potential cases identified. We also examined prevalence per 10,000 patients and calculated odds ratios for psoriasis with
95% CI stratified by sex, age group (2-5, 6-11, and 1219 years), and race/ethnicity. Analyses were conducted using PASW 17.0 (SPSS Inc, Chicago, IL) and SAS 9.1 (SAS Institute Inc, Cary, NY).
RESULTS The study population was racially and ethnically diverse with Hispanic whites representing the largest group. Almost half the population was aged 12 to 19 years (Table I). Boys and girls were comparable with regard to age and racial/ethnic distribution along with membership duration and number of medical encounters during enrollment in the health plan. Case validation study We identified a total of 2287 patients with potential psoriasis, of which 2110 patients were identified by a diagnosis of psoriasis in the electronic medical record (Fig 1). Sensitivity for detecting psoriasis based on ICD-9 codes was 99.6% for diagnoses made by any care provider. PPV for psoriasis diagnosed by any care provider was 63.7%; PPV was 90.0% for psoriasis diagnosed by a dermatologist compared with 26.6% for psoriasis diagnosed by a nondermatologist (Table II). PPV increased with increasing number of encounters consistent with a diagnosis of psoriasis. PPV was approximately 85% for at least two diagnoses by any provider and improved to 90% if at least one diagnosis was made by a dermatologist. Strategies to search for uncoded psoriasis using ICD-9 codes for psoriatic arthritis and a psoriasisspecific medication search added few additional valid cases (6 of 177 reviewed cases). Sensitivity and PPV for using the ICD-9 code for psoriatic arthritis to identify cases of psoriasis was low (Table II). Using a psoriasis-specific medication search, we identified 154 cases that did not have a recorded diagnosis of psoriasis in the electronic medical record. PPV to identify additional patients with psoriasis using psoriasis-specific medication was low (3.2%). Prevalence of psoriasis The prevalence of psoriasis confirmed by electronic medical chart review among children aged 2 to 19 years was 19/10,000 patients. The prevalence of confirmed and unconfirmed psoriasis based on having a diagnosis code indicating psoriasis was 30/10,000 patients. The prevalence increased consistently after 5 years of age (Fig 2) and was higher in boys than in girls (odds ratio 1.20, 95% CI 1.08-1.33). The prevalence was highest in non-Hispanic whites and lowest in blacks (Table III) with the highest prevalence in
J AM ACAD DERMATOL
960 Wu et al
NOVEMBER 2011
Fig 1. Flow chart illustrating ascertainment of psoriasis cases among children enrolled in large integrated health plan in 2007 through 2008. Treatment consistent with psoriasis was defined as at least one prescription for the following topical therapies (anthralin, liquor carbonis distillate, tar, and 2 or more prescriptions of calcipotriene or calcitriol) and systemic therapies (acitretin, alefacept, efalizumab, or ustekinumab). ICD-9, International Classification of Diseases, Ninth Revision; KPSC, Kaiser Permanente Southern California.
non-Hispanic white adolescents (42/10,000 patients, data not shown). The age at first diagnosis of psoriasis in the electronic medical record was slightly lower in boys than in girls (11.8 6 4.4 in boys vs 12.5 6 4.4 in girls, P = .003). When first given a diagnosis of psoriasis, 1.6% of children with confirmed psoriasis were younger than 2 years, 7.7% were 2 to 5 years, 36.4% were 6 to 11 years, 25.9% were 12 to 14 years, and 28.4% were 15 to 19 years. Most patients (92.5%) with confirmed psoriasis received a prescription treatment consistent with psoriasis (Table IV). The most frequently prescribed medications were topical steroids (91.4% of patients with psoriasis). Overall, 3.9% of patients with psoriasis were administered systemic therapy or phototherapy.
DISCUSSION The findings of our study show that the overall prevalence of psoriasis among children aged 2 to 19 years was low in southern California and varied significantly by sex, age, and racial/ethnic group. The prevalence of psoriasis confirmed by electronic medical chart review was 19/10,000 patients; the prevalence increased to 30/10,000 patients if unconfirmed cases were included. Groups who were most affected by psoriasis were white and Asian children and adolescents. The validity of psoriasis diagnosis codes was significantly higher when the diagnosis was made or confirmed by dermatologists as compared with nondermatologists. Although psoriasis is a chronic health condition associated with a significantly increased health risk,15
J AM ACAD DERMATOL
Wu et al 961
VOLUME 65, NUMBER 5
Table I. Characteristics of patients aged 2 through 19 years in southern California Patients Variable
Total
Boys
Girls
N 710,949 357,205 353,744 % of total population 100 50.2 49.8 Age group, % 2-5 y 24.0 24.6 23.4 6-11 y 29.2 29.9 28.4 12-19 y 46.9 45.5 48.2 Race/ethnicity, %* Non-Hispanic white 28.6 28.5 28.6 Hispanic white 50.8 50.5 51.1 Black 8.8 8.8 8.7 Asian or Pacific Islander 7.9 8.1 7.7 Others 4.0 4.1 3.9 Medi-Cal support 11.9 12.1 11.7 receivers, % Membership duration, yy 7.0 6 5.3 7.0 6 5.3 7.1 6 5.4 *Among patients with known race/ethnicity. Race/ethnicity was missing in 52.5% of patients before and in 12.4% of patients after imputation. y Data are mean and SD.
only 3 epidemiologic studies provide estimates on the prevalence of psoriasis in children.2,3,16 In the United Kingdom, prevalence of psoriasis was 55/10,000 children aged 0 to 9 years and 137/10,000 children and adolescents aged 10 to 19 years, with a higher prevalence in girls than in boys.3 In Germany, 71/10,000 children were affected by psoriasis.2 Results from a survey among dermatologists and general practitioners in the Netherlands suggested a prevalence of 37/10,000 children aged 0 to 10 years and 109/10,000 children aged 11 to 19 years.16 The prevalence of psoriasis found in our study was considerably lower, even when unconfirmed cases were included. This may be explained by differences in ethnicity (eg, a higher percentage of non-Hispanic whites in Europe), and by improvement as a result of sunlight exposure in southern California, and, consequently, potentially fewer patients seeking medical advice. Our data confirm a relatively early onset of psoriasis, as reported in previous studies.1,3-6,17 In Australia, 16% of pediatric patients with psoriasis were younger than 1 year and 27% younger than 2 years of age.16 Although only 2% of the children in our study were younger than 2 years at first documented diagnosis, 7% were younger than 5 years, and 45% were younger than 12 years of age. Moreover, our data suggest that the prevalence of pediatric psoriasis consistently increases with increasing age. In addition to age, race and ethnicity are important risk factors. Non-Hispanic whites and Asians were
most affected by psoriasis, whereas blacks were the least affected, with rates ranging from 6/10,000 in blacks to 30/10,000 in non-Hispanic whites. These racial/ethnic disparities in prevalence have been reported by others. Gelfand et al18 observed a 52% lower prevalence of psoriasis in black compared with non-Hispanic white adults. Data from the National Health and Nutrition Examination Survey also suggest that the prevalence of psoriasis is lower in black and Hispanic than white adults.19 The validity of psoriasis diagnosis codes in our study was relatively comparable with that observed in a study of adult psoriasis.9 PPV increases as the number of ICD-9 codes increases, and diagnosis made by a dermatologist has a higher PPV than a diagnosis made by a nondermatologist. Adding different search strategies based on psoriasis medications to identify additional cases of psoriasis that were not properly coded did not add a significant number of cases. The results of our study suggest that an ICD-9 codeebased search is well justified, but that if data are available, cases should be restricted to those with at least one diagnosis made by a dermatologist. Underdiagnosis is problematic for all epidemiologic investigations of psoriasis. In US adults, the prevalence of diagnosed psoriasis was 3.2% (95% CI 2.2-4.5) whereas undiagnosed psoriasis contributed an additional 0.4% by conservative estimates and 2.3% by a less conservative definition.19 Adult patients without a diagnosis were more likely to be non-white, male, less educated, and unmarried compared with those who had received a psoriasis diagnosis, suggesting that differences in the prevalence of psoriasis by race and sex may partially be explained by differential distribution of undiagnosed cases.19 This may in turn underlie the lower prevalence of psoriasis observed in boys and blacks. For example, a psoriasis plaque (especially on the elbows and knees) on a boy could be interpreted as the result of trauma from normal play and, therefore, may not result in a medical visit. It is also possible that boys spend more time playing outside in the sun compared with girls. Only 3.9% of pediatric patients were treated with systemic therapy or phototherapy. This is a small fraction in comparison with treatment trends observed in studies conducted in adults. In the absence of data on other pediatric populations, results from a survey conducted by the National Psoriasis Foundation suggest that 26% of adult patients with psoriasis are treated with systemic therapy, phototherapy, or both.20 The low number of children treated with systemic therapy or phototherapy in our study may be explained by a reluctance of care
J AM ACAD DERMATOL
962 Wu et al
NOVEMBER 2011
Table II. Sensitivity and positive predictive values by search strategy and by diagnosis made by type of provider in potential cases of psoriasis Sensitivity Search strategy
Psoriasis identified by: Any care provider $ 1 Diagnosis $ 2 Diagnoses $ 3 Diagnoses Dermatologist $ 1 Diagnosis $ 2 Diagnoses $ 3 Diagnoses Nondermatologist $ 1 Diagnosis $ 2 Diagnoses $ 3 Diagnoses Psoriasis identified by ICD-9 696.1 (psoriatic arthritis) Medication consistent with psoriasis Acitretin Anthralin Calcitriol or calcipotriene Tar or liquor carbonis distillate Psoriasis/psoriatic arthritis ICD-9 code and medication
Positive predictive value
n
% (95% CI)
n
% (95% CI)
1350 1350 1350
99.6 (98.9-99.8) 54.4 (51.7-57.0) 33.1 (30.6-35.7)
2110 864 480
63.7 (61.6-65.7) 85.0 (82.4-87.2) 93.1 (90.3-95.2)
1350 1350 1350
82.3 (80.1-84.3) 51.8 (49.0-54.5) 32.3 (29.8-34.9)
1234 736 448
90.0 (88.2-91.6) 95.0 (93.1-96.4) 97.3 (95.2-98.5)
1350 1350 1350 1350
17.3 2.6 0.8 0.1
1350 1350 1350 1350 1350 1350
0.4 0.0 0.0 0.4 0.0
(15.3-19.4) (1.8-3.6) (0.4-1.5) (0.0-0.4) (0.1-0.9) (0.0-0.3) (0.0-0.3) (0.1-0.9) (0.0-0.4) 100
876 127 32 23
26.6 27.6 34.4 4.3
(23.7-29.7) (20.2-36.3) (19.2-53.2) (0.2-23.0)
154 8 52 65 24 2287
3.2 0.0 0.0 7.7 0.0 59.0
(1.2-7.8) (0.0-4.0) (0.0-8.6) (2.9-17.8) (0.0-1.7) (57.0-61.0)
CI, Confidence interval; ICD-9, International Classification of Diseases, Ninth Revision.
providers to prescribe systemic treatment in pediatric populations. According to a survey of Dutch dermatologists, only 20.6% of dermatologists prescribed systemic medication for pediatric patients with psoriasis but 64.7% of dermatologists prescribed phototherapy.16 These numbers do not refer to patient numbers but reflect percentages of care providers. Thus, we cannot exclude undertreatment of pediatric patients with severe or widespread psoriasis. Moreover, considering the high sun exposure and low prevalence of psoriasis in southern California, the low proportion of patients treated with systemic therapy and phototherapy may also be explained by a reduced need for these treatment options in southern California compared with other locations. Our analysis has several potential limitations. Although the results are consistent with those observed by other investigators, our data do not include duration of psoriasis but first documented diagnosis of psoriasis since patient enrollment in the health plan. However, the average membership duration of our study population was relatively long in relation to the patient’s age. Consequently, age at first onset of psoriasis may be slightly younger than reported in our study. Different morphologies such as pustular, inverse, and guttate may have led to misclassification of psoriasis cases as noncases during chart review,
Fig 2. Prevalence of psoriasis in children and adolescents by age and sex. At age 12 years and older, prevalence of psoriasis in girls was significantly higher than in boys (P \ .001).
especially in children younger than 5 years of age. This misclassification would likely bias our results toward the null. In addition, given the geographic context of an investigation in southern California, our findings should be interpreted with caution. They do not necessarily reflect the prevalence of psoriasis in the United States, but provide valuable information on its prevalence in an area with high
J AM ACAD DERMATOL
Wu et al 963
VOLUME 65, NUMBER 5
Table III. Prevalence of psoriasis (n = 1350 cases) among 710,494 children enrolled in integrated health plan in southern California in 2007 through 2008 Boys
All Age group, % 2-5 y 6-11 y 12-19 y Race/ethnicity, %y Non-Hispanic white Hispanic white Black Asian or Pacific Islander Other Unknown
n
Prevalence/10,000 (95% CI)
616
17 (16-19)
28 174 414 230 239 14 51 23 59
Girls n
Prevalence/10,000 (95% CI)
1.00 (Ref)
745
21 (19-22)
1.20 (1.08-1.33)
3 (2-5) 16 (14-19) 25 (23-28)
1.00 (Ref) 5.18 (3.47-7.72) 7.86 (5.36-11.54)
30 155 560
4 (2-5) 15 (13-18) 33 (30-36)
1.00 (Ref) 4.24 (2.87-6.28) 8.77 (6.01-12.67)
25 15 5 20
1.00 0.61 0.20 0.81
(Ref) (0.50-0.73) (0.11-0.34) (0.59-1.10)
287 279 21 48
33 17 8 20
1.00 0.58 0.22 0.66
0.97 (0.62-1.53) 0.53 (0.39-0.70)
17 93
(22-29) (13-17) (3-9) (15-26)
18 (11-27) 13 (10-17)
OR (95% CI)*
(28-36) (15-20) (5-12) (15-27)
14 (8-23) 21 (17-25)
OR (95% CI)*
(Ref) (0.49-0.69) (0.14-0.35) (0.48-0.90)
0.62 (0.37-1.04) 0.61 (0.48-0.78)
CI, Confidence interval; OR, odds ratio; Ref, reference. *Adjusted for medical center and mutually for sex, age, and race/ethnicity. y Among patients with known race/ethnicity. Race/ethnicity was missing in 52.5% of patients before and in 12.4% of patients after imputation.
Table IV. Frequency of different treatments in children with confirmed psoriasis (n = 1350) Treatment
Topicals Nonsteroid Steroid Phototherapy UVB PUVA Traditional systemics Acitretin Cyclosporine Methotrexate Other systemics Biologics TNF-alfa blocker T-cell blockers Interleukin-12/23 inhibitor
Patients, No. (%) n = 1350
12 (0.9) 1234 (91.4) 13 (1.0) 4 (0.3) 15 2 14 2
(1.1) (0.1) (1.0) (0.1)
19 (1.4) 0 (0.0) 0 (0.0)
PUVA, Psoralen plus ultraviolet A; TNF, tumor necrosis factor; UV, ultraviolet. Traditional systemics are acitretin, cyclosporine, and methotrexate. TNF-alfa blockers are etanercept, adalimumab, and infliximab. T-cell blockers are alefacept and efalizumab. Interleukin-12/23 inhibitor is ustekinumab. Other systemics are azathioprine, mycophenolate mofetil, and hydroxyurea.
sun exposure and a high proportion of non-white people. Despite the limitations, our study has several strengths. It uses a population-based approach. The large multiethnic and diverse pediatric
population has equal access to health care and lives throughout the 7-county region of southern California. Therefore, we can assume that our results are generalizable to other insured populations of a comparable geographic location with a high potential of sun exposure. Our study also involved manual review and confirmation of all psoriasis cases by dermatologists and/or trained staff to ensure the most precise prevalence estimates possible. We performed a comprehensive search to identify potentially undiagnosed cases seeking medical attention. In conclusion, the prevalence of psoriasis in children in southern California was low compared with other geographic areas but varied significantly by sex and race/ethnicity. The low prevalence may be explained by more widespread sunlight exposure in southern California. The early onset of psoriasis observed in this pediatric population and its potential link to future disease risk and disease progression requires further investigation. We thank Alexander Carruth and Monica Levitt for their excellent technical support. REFERENCES 1. de Jager ME, de Jong EM, Meeuwis KA, van de Kerkhof PC, Seyger MM. No evidence found that childhood onset of psoriasis influences disease severity, future body mass index or type of treatments used. J Eur Acad Dermatol Venereol 2010;24:1333-9. 2. Augustin M, Glaeske G, Radtke MA, Christophers E, Reich K, Schafer I. Epidemiology and comorbidity of psoriasis in children. Br J Dermatol 2010;162:633-6.
J AM ACAD DERMATOL
964 Wu et al
3. Gelfand JM, Weinstein R, Porter SB, Neimann AL, Berlin JA, Margolis DJ. Prevalence and treatment of psoriasis in the United Kingdom: a population-based study. Arch Dermatol 2005;141:1537-41. 4. Raychaudhuri SP, Gross J. A comparative study of pediatric onset psoriasis with adult onset psoriasis. Pediatr Dermatol 2000;17:174-8. 5. Farber EM, Nall ML. The natural history of psoriasis in 5,600 patients. Dermatologica 1974;148:1-18. 6. Nyfors A, Lemholt K. Psoriasis in children: a short review and a survey of 245 cases. Br J Dermatol 1975;92:437-42. 7. Mallbris L, Akre O, Granath F, Yin L, Lindelof B, Ekbom A, et al. Increased risk for cardiovascular mortality in psoriasis inpatients but not in outpatients. Eur J Epidemiol 2004;19:225-30. 8. Setty AR, Curhan G, Choi HK. Obesity, waist circumference, weight change, and the risk of psoriasis in women: nurses’ health study II. Arch Intern Med 2007;167:1670-5. 9. Icen M, Crowson CS, McEvoy MT, Gabriel SE, Maradit Kremers H. Potential misclassification of patients with psoriasis in electronic databases. J Am Acad Dermatol 2008;59:981-5. 10. Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007-2008. JAMA 2010;303:242-9. 11. Koebnick C, Smith N, Coleman KJ, Getahun D, Reynolds K, Quinn VP, et al. Prevalence of extreme obesity in a multiethnic cohort of children and adolescents. J Pediatr 2010;157: 26-31e2. 12. Menter A, Gottlieb A, Feldman SR, Van Voorhees AS, Leonardi CL, Gordon KB, et al. Guidelines of care for the management of
NOVEMBER 2011
13.
14.
15.
16.
17.
18.
19.
20.
psoriasis and psoriatic arthritis, section 1: overview of psoriasis and guidelines of care for the treatment of psoriasis with biologics. J Am Acad Dermatol 2008;58:826-50. Bureau of the Census. Census 2000 surname list. Washington (DC): 2009. Available from: URL:http://wwwcensusgov/genealogy/ www/freqnames2k.html. Accessed April 1, 2010. Word DL, Perkins RC. Building a Spanish surname list for the 1990sea new approach to an old problem; technical working paper No. 13. Washington (DC): US Bureau of the Census; 1996. Azfar RS, Gelfand JM. Psoriasis and metabolic disease: epidemiology and pathophysiology. Curr Opin Rheumatol 2008;20: 416-22. De Jager ME, Van de Kerkhof PC, De Jong EM, Seyger MM. Epidemiology and prescribed treatments in childhood psoriasis: a survey among medical professionals. J Dermatolog Treat 2009;20:254-8. Morris A, Rogers M, Fischer G, Williams K. Childhood psoriasis: a clinical review of 1262 cases. Pediatr Dermatol 2001;18: 188-98. Gelfand JM, Stern RS, Nijsten T, Feldman SR, Thomas J, Kist J, et al. The prevalence of psoriasis in African Americans: results from a population-based study. J Am Acad Dermatol 2005;52:23-6. Kurd SK, Gelfand JM. The prevalence of previously diagnosed and undiagnosed psoriasis in US adults: results from NHANES 2003-2004. J Am Acad Dermatol 2009;60:218-24. Horn EJ, Fox KM, Patel V, Chiou CF, Dann F, Lebwohl M. Are patients with psoriasis undertreated? Results of National Psoriasis Foundation survey. J Am Acad Dermatol 2007;57: 957-62.
cant information is available on the prevalence of psoriasis in children. Although a recent study suggested that childhood onset of psoriasis is not associated with disease severity,1 early
S
onset may result in longer exposure to a chronic inflammatory condition and, thus, may affect the morbidity and mortality risk. Previous studies found prevalence estimates of pediatric psoriasis ranging
From the Department of Research and Evaluation, Kaiser Permanente Southern California, Pasadenaa; Department of Dermatology, Kaiser Permanente Los Angeles Medical Centerb; and Department of Pediatrics, Baldwin Park Medical Center, Southern California Permanente Medical Group.c Supported by Kaiser Permanente Direct Community Benefit Funds. Disclosure: Dr Wu received research funding from Abbott Laboratories and Amgen. Dr Jacobsen received research grants from Merck and Beckman-coulter. He also served as consultant for
Merck without compensation. Drs Black, Smith, Porter, and Koebnick have no conflicts of interest to declare. Accepted for publication September 7, 2010. Reprint requests: Jashin J. Wu, MD, Department of Dermatology, Kaiser Permanente Los Angeles Medical Center, 1515 N Vermont Ave, 5th Floor, Los Angeles, CA 90027. E-mail: [email protected]. Published online February 9, 2011. 0190-9622/$36.00 ª 2010 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2010.09.005
957
J AM ACAD DERMATOL
958 Wu et al
NOVEMBER 2011
from 0.5% to 1.4%.2,3 Onset during the first 2 decades extracting ICD-9 code 696.1 (psoriasis) from elecof life is reported by 31% to 45% of affected tronic medical records for all inpatient and outpaadults.1,4-6 tient encounters from enrollment in the health plan Large epidemiologic studies often rely on diagnoto enrollment in the current studyethese cases were sis codes3,7 or self-reporting8 for the identification of subdivided into whether or not the diagnosis of cases. A recent report in adult psoriasis suggests that psoriasis was made by a dermatologist; (2) by case identification based solely on the International extracting ICD-9 code 696.0 (psoriatic arthritis) Classification of Diseases, from electronic medical recNinth Revision (ICD-9) codes ords; and (3) by performing a CAPSULE SUMMARY may lead to misclassificasearch for psoriasis-specific tion.9 Only 73.5% of psoriasis medications and treatment. The prevalence of psoriasis in children cases identified in this study Medications were limited to and adolescents in southern California were confirmed as prevalent those that are very specific was found to be low compared with or incident psoriasis after for psoriasis treatment or other countries. manual review of patient exclusively approved for psocharts by a dermatologist. The prevalence of psoriasis was higher in riasis treatment. PsoriasisNo information is available girls than in boys and varied by specific medications were on the validity of diagnosis race/ethnicity with the highest defined as evidence of a precodes for psoriasis in chilprevalence in non-Hispanic whites and scription for one of the foldren. This information is the lowest in blacks. lowing topical therapies needed to provide valid prev(anthralin, liquor carbonis Based on diagnostic codes, more than alence estimates for psoriasis. distillate, tar, and 2 or more 90% of psoriasis cases diagnosed by a We investigated the prevaprescriptions of calcipotriene dermatologist could be confirmed after lence, disparities in the or calcitriol) and systemic an independent medical chart review prevalence, and treatment of therapies (acitretin, alefacept, compared with 27% diagnosed by psoriasis in a multiethnic efalizumab, or ustekinumab). nondermatologists. population-based cohort of For most medications, pamore than 710,000 children tients with at least one preaged 2 to 19 years10 in southern California who were scription were considered as potential cases of enrolled in an integrated prepaid health plan in 2007 psoriasis for manual chart review; for calcipotriene through 2008. We also examined the validity of psoand calcitriol, at least two prescriptions were needed. riasis diagnosis and devised a comprehensive search To provide information on treatment of psoriasis, algorithm to detect patients with uncoded psoriasis. we identified patients who were prescribed topical treatments (steroid and nonsteroid), phototherapy, METHODS traditional systemic medications (acitretin, cycloStudy design and subjects sporine, methotrexate), other less commonly used For the Kaiser Permanente Southern California systemic agents (azathioprine, mycophenolate mo(KPSC) Children’s Health Study, a cohort of children fetil, hydroxyurea), biologics (tumor necrosis factoraged 2 to 19 years (n = 710,949) was established at alfa inhibitors, T-cell inhibitors, or interleukin-12/23 KPSC, which provided health care to over 3.2 million inhibitors), or phototherapy (ultraviolet B or psoramembers throughout the 7-county region in the len plus ultraviolet A).12 11 years 2007 through 2008. The KPSC population is ethnically diverse and largely representative of Case validation southern California, and includes approximately Potential psoriasis cases identified by ICD-9 code or 14% of the population of the state of California.11 treatment search were adjudicated by chart review. Inclusion criteria were at least one body weight and Patients who were given a diagnosis by a primary care height measurement in the electronic medical record provider, were younger than 5 years of age, had a during 2007 through 2008. Pregnant girls and young diagnosis of psoriatic arthritis, or were prescribed women were excluded from the study. The study psoriasis-specific medications, were reviewed by a protocol was reviewed and approved by the institudermatologist (J. J. W., n = 1082). Potential patients tional review board of KPSC before implementation. with psoriasis identified by ICD-9 code older than 5 years of age and with a diagnosis made at least once by Outcome ascertainment and treatment a dermatologist (n = 1037) were reviewed by trained Prevalent cases of psoriasis were identified by 3 research staff for the following terms: ‘‘silvery,’’ different search strategies (n = 2287) (Fig 1): (1) by ‘‘flaky,’’ ‘‘papulosquamous,’’ ‘‘red,’’ ‘‘papules,’’ and d
d
d
J AM ACAD DERMATOL
Wu et al 959
VOLUME 65, NUMBER 5
Abbreviations used: CI: ICD-9: KPSC: PPV:
confidence interval International Classification of Diseases, Ninth Revision Kaiser Permanente Southern California positive predictive value
‘‘plaques’’ in the following body locations: ‘‘knees’’ and ‘‘elbows.’’ Any unclear or questionable records were referred to a dermatologist for further review. Race/ethnicity Race and ethnicity information were obtained from administrative records supplemented by birth certificate information. We categorized race/ethnicity as non-Hispanic white, Hispanic white, black (regardless of ethnicity), Asian or Pacific Islander, other or multiple race/ethnicity, and unknown because of missing information (52.5%). For unknown race and ethnicity information, administrative records were supplemented by an imputation algorithm based on surname lists and address information derived from the US Census Bureau.13,14 Hispanic ethnicity and Asian race were assigned based on surnames. For blacks and non-Hispanic whites, additional geocoding information based on member address was linked to racial/ethnic information of the US Census Bureau data. With a hierarchical assignment order, race/ethnicity was assigned using probability cut-offs more than 50% for Asian surname, more than 50% for Hispanic surname, more than 75% for black race from geocoding if probability for Asian surname was less than 50% (Hispanic blacks were assigned to black race for this study), and white race more than 45% from geocoding if no other assignment could be made before. The specificity and positive predictive values (PPV) were more than 98% for all races/ethnicities.11 Statistical analysis Prevalence of psoriasis was calculated per 10,000 patients with corresponding 95% confidence intervals (CIs) assuming that the occurrence of psoriasis follows a Poisson distribution. Sensitivity for psoriasis diagnosis was defined as the number of confirmed psoriasis cases identified by a particular search strategy divided by the total number of confirmed psoriasis cases. Sensitivities were estimated using cases that were identified by ICD-9 code during a period of 24 months before the study. PPV were defined as the proportion of confirmed cases to potential cases identified. We also examined prevalence per 10,000 patients and calculated odds ratios for psoriasis with
95% CI stratified by sex, age group (2-5, 6-11, and 1219 years), and race/ethnicity. Analyses were conducted using PASW 17.0 (SPSS Inc, Chicago, IL) and SAS 9.1 (SAS Institute Inc, Cary, NY).
RESULTS The study population was racially and ethnically diverse with Hispanic whites representing the largest group. Almost half the population was aged 12 to 19 years (Table I). Boys and girls were comparable with regard to age and racial/ethnic distribution along with membership duration and number of medical encounters during enrollment in the health plan. Case validation study We identified a total of 2287 patients with potential psoriasis, of which 2110 patients were identified by a diagnosis of psoriasis in the electronic medical record (Fig 1). Sensitivity for detecting psoriasis based on ICD-9 codes was 99.6% for diagnoses made by any care provider. PPV for psoriasis diagnosed by any care provider was 63.7%; PPV was 90.0% for psoriasis diagnosed by a dermatologist compared with 26.6% for psoriasis diagnosed by a nondermatologist (Table II). PPV increased with increasing number of encounters consistent with a diagnosis of psoriasis. PPV was approximately 85% for at least two diagnoses by any provider and improved to 90% if at least one diagnosis was made by a dermatologist. Strategies to search for uncoded psoriasis using ICD-9 codes for psoriatic arthritis and a psoriasisspecific medication search added few additional valid cases (6 of 177 reviewed cases). Sensitivity and PPV for using the ICD-9 code for psoriatic arthritis to identify cases of psoriasis was low (Table II). Using a psoriasis-specific medication search, we identified 154 cases that did not have a recorded diagnosis of psoriasis in the electronic medical record. PPV to identify additional patients with psoriasis using psoriasis-specific medication was low (3.2%). Prevalence of psoriasis The prevalence of psoriasis confirmed by electronic medical chart review among children aged 2 to 19 years was 19/10,000 patients. The prevalence of confirmed and unconfirmed psoriasis based on having a diagnosis code indicating psoriasis was 30/10,000 patients. The prevalence increased consistently after 5 years of age (Fig 2) and was higher in boys than in girls (odds ratio 1.20, 95% CI 1.08-1.33). The prevalence was highest in non-Hispanic whites and lowest in blacks (Table III) with the highest prevalence in
J AM ACAD DERMATOL
960 Wu et al
NOVEMBER 2011
Fig 1. Flow chart illustrating ascertainment of psoriasis cases among children enrolled in large integrated health plan in 2007 through 2008. Treatment consistent with psoriasis was defined as at least one prescription for the following topical therapies (anthralin, liquor carbonis distillate, tar, and 2 or more prescriptions of calcipotriene or calcitriol) and systemic therapies (acitretin, alefacept, efalizumab, or ustekinumab). ICD-9, International Classification of Diseases, Ninth Revision; KPSC, Kaiser Permanente Southern California.
non-Hispanic white adolescents (42/10,000 patients, data not shown). The age at first diagnosis of psoriasis in the electronic medical record was slightly lower in boys than in girls (11.8 6 4.4 in boys vs 12.5 6 4.4 in girls, P = .003). When first given a diagnosis of psoriasis, 1.6% of children with confirmed psoriasis were younger than 2 years, 7.7% were 2 to 5 years, 36.4% were 6 to 11 years, 25.9% were 12 to 14 years, and 28.4% were 15 to 19 years. Most patients (92.5%) with confirmed psoriasis received a prescription treatment consistent with psoriasis (Table IV). The most frequently prescribed medications were topical steroids (91.4% of patients with psoriasis). Overall, 3.9% of patients with psoriasis were administered systemic therapy or phototherapy.
DISCUSSION The findings of our study show that the overall prevalence of psoriasis among children aged 2 to 19 years was low in southern California and varied significantly by sex, age, and racial/ethnic group. The prevalence of psoriasis confirmed by electronic medical chart review was 19/10,000 patients; the prevalence increased to 30/10,000 patients if unconfirmed cases were included. Groups who were most affected by psoriasis were white and Asian children and adolescents. The validity of psoriasis diagnosis codes was significantly higher when the diagnosis was made or confirmed by dermatologists as compared with nondermatologists. Although psoriasis is a chronic health condition associated with a significantly increased health risk,15
J AM ACAD DERMATOL
Wu et al 961
VOLUME 65, NUMBER 5
Table I. Characteristics of patients aged 2 through 19 years in southern California Patients Variable
Total
Boys
Girls
N 710,949 357,205 353,744 % of total population 100 50.2 49.8 Age group, % 2-5 y 24.0 24.6 23.4 6-11 y 29.2 29.9 28.4 12-19 y 46.9 45.5 48.2 Race/ethnicity, %* Non-Hispanic white 28.6 28.5 28.6 Hispanic white 50.8 50.5 51.1 Black 8.8 8.8 8.7 Asian or Pacific Islander 7.9 8.1 7.7 Others 4.0 4.1 3.9 Medi-Cal support 11.9 12.1 11.7 receivers, % Membership duration, yy 7.0 6 5.3 7.0 6 5.3 7.1 6 5.4 *Among patients with known race/ethnicity. Race/ethnicity was missing in 52.5% of patients before and in 12.4% of patients after imputation. y Data are mean and SD.
only 3 epidemiologic studies provide estimates on the prevalence of psoriasis in children.2,3,16 In the United Kingdom, prevalence of psoriasis was 55/10,000 children aged 0 to 9 years and 137/10,000 children and adolescents aged 10 to 19 years, with a higher prevalence in girls than in boys.3 In Germany, 71/10,000 children were affected by psoriasis.2 Results from a survey among dermatologists and general practitioners in the Netherlands suggested a prevalence of 37/10,000 children aged 0 to 10 years and 109/10,000 children aged 11 to 19 years.16 The prevalence of psoriasis found in our study was considerably lower, even when unconfirmed cases were included. This may be explained by differences in ethnicity (eg, a higher percentage of non-Hispanic whites in Europe), and by improvement as a result of sunlight exposure in southern California, and, consequently, potentially fewer patients seeking medical advice. Our data confirm a relatively early onset of psoriasis, as reported in previous studies.1,3-6,17 In Australia, 16% of pediatric patients with psoriasis were younger than 1 year and 27% younger than 2 years of age.16 Although only 2% of the children in our study were younger than 2 years at first documented diagnosis, 7% were younger than 5 years, and 45% were younger than 12 years of age. Moreover, our data suggest that the prevalence of pediatric psoriasis consistently increases with increasing age. In addition to age, race and ethnicity are important risk factors. Non-Hispanic whites and Asians were
most affected by psoriasis, whereas blacks were the least affected, with rates ranging from 6/10,000 in blacks to 30/10,000 in non-Hispanic whites. These racial/ethnic disparities in prevalence have been reported by others. Gelfand et al18 observed a 52% lower prevalence of psoriasis in black compared with non-Hispanic white adults. Data from the National Health and Nutrition Examination Survey also suggest that the prevalence of psoriasis is lower in black and Hispanic than white adults.19 The validity of psoriasis diagnosis codes in our study was relatively comparable with that observed in a study of adult psoriasis.9 PPV increases as the number of ICD-9 codes increases, and diagnosis made by a dermatologist has a higher PPV than a diagnosis made by a nondermatologist. Adding different search strategies based on psoriasis medications to identify additional cases of psoriasis that were not properly coded did not add a significant number of cases. The results of our study suggest that an ICD-9 codeebased search is well justified, but that if data are available, cases should be restricted to those with at least one diagnosis made by a dermatologist. Underdiagnosis is problematic for all epidemiologic investigations of psoriasis. In US adults, the prevalence of diagnosed psoriasis was 3.2% (95% CI 2.2-4.5) whereas undiagnosed psoriasis contributed an additional 0.4% by conservative estimates and 2.3% by a less conservative definition.19 Adult patients without a diagnosis were more likely to be non-white, male, less educated, and unmarried compared with those who had received a psoriasis diagnosis, suggesting that differences in the prevalence of psoriasis by race and sex may partially be explained by differential distribution of undiagnosed cases.19 This may in turn underlie the lower prevalence of psoriasis observed in boys and blacks. For example, a psoriasis plaque (especially on the elbows and knees) on a boy could be interpreted as the result of trauma from normal play and, therefore, may not result in a medical visit. It is also possible that boys spend more time playing outside in the sun compared with girls. Only 3.9% of pediatric patients were treated with systemic therapy or phototherapy. This is a small fraction in comparison with treatment trends observed in studies conducted in adults. In the absence of data on other pediatric populations, results from a survey conducted by the National Psoriasis Foundation suggest that 26% of adult patients with psoriasis are treated with systemic therapy, phototherapy, or both.20 The low number of children treated with systemic therapy or phototherapy in our study may be explained by a reluctance of care
J AM ACAD DERMATOL
962 Wu et al
NOVEMBER 2011
Table II. Sensitivity and positive predictive values by search strategy and by diagnosis made by type of provider in potential cases of psoriasis Sensitivity Search strategy
Psoriasis identified by: Any care provider $ 1 Diagnosis $ 2 Diagnoses $ 3 Diagnoses Dermatologist $ 1 Diagnosis $ 2 Diagnoses $ 3 Diagnoses Nondermatologist $ 1 Diagnosis $ 2 Diagnoses $ 3 Diagnoses Psoriasis identified by ICD-9 696.1 (psoriatic arthritis) Medication consistent with psoriasis Acitretin Anthralin Calcitriol or calcipotriene Tar or liquor carbonis distillate Psoriasis/psoriatic arthritis ICD-9 code and medication
Positive predictive value
n
% (95% CI)
n
% (95% CI)
1350 1350 1350
99.6 (98.9-99.8) 54.4 (51.7-57.0) 33.1 (30.6-35.7)
2110 864 480
63.7 (61.6-65.7) 85.0 (82.4-87.2) 93.1 (90.3-95.2)
1350 1350 1350
82.3 (80.1-84.3) 51.8 (49.0-54.5) 32.3 (29.8-34.9)
1234 736 448
90.0 (88.2-91.6) 95.0 (93.1-96.4) 97.3 (95.2-98.5)
1350 1350 1350 1350
17.3 2.6 0.8 0.1
1350 1350 1350 1350 1350 1350
0.4 0.0 0.0 0.4 0.0
(15.3-19.4) (1.8-3.6) (0.4-1.5) (0.0-0.4) (0.1-0.9) (0.0-0.3) (0.0-0.3) (0.1-0.9) (0.0-0.4) 100
876 127 32 23
26.6 27.6 34.4 4.3
(23.7-29.7) (20.2-36.3) (19.2-53.2) (0.2-23.0)
154 8 52 65 24 2287
3.2 0.0 0.0 7.7 0.0 59.0
(1.2-7.8) (0.0-4.0) (0.0-8.6) (2.9-17.8) (0.0-1.7) (57.0-61.0)
CI, Confidence interval; ICD-9, International Classification of Diseases, Ninth Revision.
providers to prescribe systemic treatment in pediatric populations. According to a survey of Dutch dermatologists, only 20.6% of dermatologists prescribed systemic medication for pediatric patients with psoriasis but 64.7% of dermatologists prescribed phototherapy.16 These numbers do not refer to patient numbers but reflect percentages of care providers. Thus, we cannot exclude undertreatment of pediatric patients with severe or widespread psoriasis. Moreover, considering the high sun exposure and low prevalence of psoriasis in southern California, the low proportion of patients treated with systemic therapy and phototherapy may also be explained by a reduced need for these treatment options in southern California compared with other locations. Our analysis has several potential limitations. Although the results are consistent with those observed by other investigators, our data do not include duration of psoriasis but first documented diagnosis of psoriasis since patient enrollment in the health plan. However, the average membership duration of our study population was relatively long in relation to the patient’s age. Consequently, age at first onset of psoriasis may be slightly younger than reported in our study. Different morphologies such as pustular, inverse, and guttate may have led to misclassification of psoriasis cases as noncases during chart review,
Fig 2. Prevalence of psoriasis in children and adolescents by age and sex. At age 12 years and older, prevalence of psoriasis in girls was significantly higher than in boys (P \ .001).
especially in children younger than 5 years of age. This misclassification would likely bias our results toward the null. In addition, given the geographic context of an investigation in southern California, our findings should be interpreted with caution. They do not necessarily reflect the prevalence of psoriasis in the United States, but provide valuable information on its prevalence in an area with high
J AM ACAD DERMATOL
Wu et al 963
VOLUME 65, NUMBER 5
Table III. Prevalence of psoriasis (n = 1350 cases) among 710,494 children enrolled in integrated health plan in southern California in 2007 through 2008 Boys
All Age group, % 2-5 y 6-11 y 12-19 y Race/ethnicity, %y Non-Hispanic white Hispanic white Black Asian or Pacific Islander Other Unknown
n
Prevalence/10,000 (95% CI)
616
17 (16-19)
28 174 414 230 239 14 51 23 59
Girls n
Prevalence/10,000 (95% CI)
1.00 (Ref)
745
21 (19-22)
1.20 (1.08-1.33)
3 (2-5) 16 (14-19) 25 (23-28)
1.00 (Ref) 5.18 (3.47-7.72) 7.86 (5.36-11.54)
30 155 560
4 (2-5) 15 (13-18) 33 (30-36)
1.00 (Ref) 4.24 (2.87-6.28) 8.77 (6.01-12.67)
25 15 5 20
1.00 0.61 0.20 0.81
(Ref) (0.50-0.73) (0.11-0.34) (0.59-1.10)
287 279 21 48
33 17 8 20
1.00 0.58 0.22 0.66
0.97 (0.62-1.53) 0.53 (0.39-0.70)
17 93
(22-29) (13-17) (3-9) (15-26)
18 (11-27) 13 (10-17)
OR (95% CI)*
(28-36) (15-20) (5-12) (15-27)
14 (8-23) 21 (17-25)
OR (95% CI)*
(Ref) (0.49-0.69) (0.14-0.35) (0.48-0.90)
0.62 (0.37-1.04) 0.61 (0.48-0.78)
CI, Confidence interval; OR, odds ratio; Ref, reference. *Adjusted for medical center and mutually for sex, age, and race/ethnicity. y Among patients with known race/ethnicity. Race/ethnicity was missing in 52.5% of patients before and in 12.4% of patients after imputation.
Table IV. Frequency of different treatments in children with confirmed psoriasis (n = 1350) Treatment
Topicals Nonsteroid Steroid Phototherapy UVB PUVA Traditional systemics Acitretin Cyclosporine Methotrexate Other systemics Biologics TNF-alfa blocker T-cell blockers Interleukin-12/23 inhibitor
Patients, No. (%) n = 1350
12 (0.9) 1234 (91.4) 13 (1.0) 4 (0.3) 15 2 14 2
(1.1) (0.1) (1.0) (0.1)
19 (1.4) 0 (0.0) 0 (0.0)
PUVA, Psoralen plus ultraviolet A; TNF, tumor necrosis factor; UV, ultraviolet. Traditional systemics are acitretin, cyclosporine, and methotrexate. TNF-alfa blockers are etanercept, adalimumab, and infliximab. T-cell blockers are alefacept and efalizumab. Interleukin-12/23 inhibitor is ustekinumab. Other systemics are azathioprine, mycophenolate mofetil, and hydroxyurea.
sun exposure and a high proportion of non-white people. Despite the limitations, our study has several strengths. It uses a population-based approach. The large multiethnic and diverse pediatric
population has equal access to health care and lives throughout the 7-county region of southern California. Therefore, we can assume that our results are generalizable to other insured populations of a comparable geographic location with a high potential of sun exposure. Our study also involved manual review and confirmation of all psoriasis cases by dermatologists and/or trained staff to ensure the most precise prevalence estimates possible. We performed a comprehensive search to identify potentially undiagnosed cases seeking medical attention. In conclusion, the prevalence of psoriasis in children in southern California was low compared with other geographic areas but varied significantly by sex and race/ethnicity. The low prevalence may be explained by more widespread sunlight exposure in southern California. The early onset of psoriasis observed in this pediatric population and its potential link to future disease risk and disease progression requires further investigation. We thank Alexander Carruth and Monica Levitt for their excellent technical support. REFERENCES 1. de Jager ME, de Jong EM, Meeuwis KA, van de Kerkhof PC, Seyger MM. No evidence found that childhood onset of psoriasis influences disease severity, future body mass index or type of treatments used. J Eur Acad Dermatol Venereol 2010;24:1333-9. 2. Augustin M, Glaeske G, Radtke MA, Christophers E, Reich K, Schafer I. Epidemiology and comorbidity of psoriasis in children. Br J Dermatol 2010;162:633-6.
J AM ACAD DERMATOL
964 Wu et al
3. Gelfand JM, Weinstein R, Porter SB, Neimann AL, Berlin JA, Margolis DJ. Prevalence and treatment of psoriasis in the United Kingdom: a population-based study. Arch Dermatol 2005;141:1537-41. 4. Raychaudhuri SP, Gross J. A comparative study of pediatric onset psoriasis with adult onset psoriasis. Pediatr Dermatol 2000;17:174-8. 5. Farber EM, Nall ML. The natural history of psoriasis in 5,600 patients. Dermatologica 1974;148:1-18. 6. Nyfors A, Lemholt K. Psoriasis in children: a short review and a survey of 245 cases. Br J Dermatol 1975;92:437-42. 7. Mallbris L, Akre O, Granath F, Yin L, Lindelof B, Ekbom A, et al. Increased risk for cardiovascular mortality in psoriasis inpatients but not in outpatients. Eur J Epidemiol 2004;19:225-30. 8. Setty AR, Curhan G, Choi HK. Obesity, waist circumference, weight change, and the risk of psoriasis in women: nurses’ health study II. Arch Intern Med 2007;167:1670-5. 9. Icen M, Crowson CS, McEvoy MT, Gabriel SE, Maradit Kremers H. Potential misclassification of patients with psoriasis in electronic databases. J Am Acad Dermatol 2008;59:981-5. 10. Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007-2008. JAMA 2010;303:242-9. 11. Koebnick C, Smith N, Coleman KJ, Getahun D, Reynolds K, Quinn VP, et al. Prevalence of extreme obesity in a multiethnic cohort of children and adolescents. J Pediatr 2010;157: 26-31e2. 12. Menter A, Gottlieb A, Feldman SR, Van Voorhees AS, Leonardi CL, Gordon KB, et al. Guidelines of care for the management of
NOVEMBER 2011
13.
14.
15.
16.
17.
18.
19.
20.
psoriasis and psoriatic arthritis, section 1: overview of psoriasis and guidelines of care for the treatment of psoriasis with biologics. J Am Acad Dermatol 2008;58:826-50. Bureau of the Census. Census 2000 surname list. Washington (DC): 2009. Available from: URL:http://wwwcensusgov/genealogy/ www/freqnames2k.html. Accessed April 1, 2010. Word DL, Perkins RC. Building a Spanish surname list for the 1990sea new approach to an old problem; technical working paper No. 13. Washington (DC): US Bureau of the Census; 1996. Azfar RS, Gelfand JM. Psoriasis and metabolic disease: epidemiology and pathophysiology. Curr Opin Rheumatol 2008;20: 416-22. De Jager ME, Van de Kerkhof PC, De Jong EM, Seyger MM. Epidemiology and prescribed treatments in childhood psoriasis: a survey among medical professionals. J Dermatolog Treat 2009;20:254-8. Morris A, Rogers M, Fischer G, Williams K. Childhood psoriasis: a clinical review of 1262 cases. Pediatr Dermatol 2001;18: 188-98. Gelfand JM, Stern RS, Nijsten T, Feldman SR, Thomas J, Kist J, et al. The prevalence of psoriasis in African Americans: results from a population-based study. J Am Acad Dermatol 2005;52:23-6. Kurd SK, Gelfand JM. The prevalence of previously diagnosed and undiagnosed psoriasis in US adults: results from NHANES 2003-2004. J Am Acad Dermatol 2009;60:218-24. Horn EJ, Fox KM, Patel V, Chiou CF, Dann F, Lebwohl M. Are patients with psoriasis undertreated? Results of National Psoriasis Foundation survey. J Am Acad Dermatol 2007;57: 957-62.