Increased risk of lichen simplex chronicus in
people with
anxiety disorder: a nationwide
population-based
retrospective cohort study*
Y.-H. Liao,
1,2C.-C. Lin,
3,4P.-P. Tsai,
1,2W.-C. Shen,
1,5F.-C. Sung
3,4and C.-H. Kao
2,6The brain and the skin originate from the same embryonic neuroectoderm in the last differentiation of embryological development. Pruritus is the major bridging symptom between these two organs. The cingulate cortex is the area in the brain mainly involved in pruritus processing and is deactivated after scratching. Furthermore, the anterior cingulate cortex is involved in emotional and cognitive activity modulation, such as reward anticipation. Mood and motivation affect pruritus perception and processing, and this can possibly be explained by the physiology of the cingulate cortex.1,2 Lichen simplex
chronicus (LSC), or circumscribed neurodermatitis, is a common skin disorder characterized by skin lichenification following excessive scratching.3 LSC affects up to 12% of the total
population, and affects the female and adult populations more frequently than the male and young populations. LSC typically involves the neck, elbow, ankles, vulva, face and eyelids. It is not a life-threatening disease, but it can result in psychosocial problems, and it can impair quality of life through sleep disturbance and sexual dysfunction.4
Pruritus is a predominant symptom of LSC and causes a strong desire to scratch. Long-term scratching results in skin lesions that are thick, lichenified plaques, which further result in pruritus. LSC is a chronic skin disease caused by the pruritus scratching cycle. Psychological factors may contribute to
5 A retrospective study of 30 participants indicated that
patients with LSC suffer disproportionately from depressive, dissociative and anxiety disorders compared with the general population.5 Another study of 60 participants showed that
patients suffering from LSC had a greater tendency to avoid pain, and were more dependent on the desires of other people, more conformist and more dutiful than the general population.
6 To estimate the impacts of psychological factors, such
as anxiety disorders, on the incidence of LSC in a large general population, we conducted a nationwide population-based cohort study.
Patients and methods
Data source
The National Health Insurance Research Database (NHIRD) contains reimbursement claim data from the Taiwan National Health Insurance programme, which was established as a nationwide single-payer health insurance programme in 1996 and has covered > 99% of the residents of Taiwan since 1998. The database is maintained and managed by the National Health Research Institutes (NHRI).
This study was conducted using the Longitudinal Health Insurance Database (LHID), which is a subset of the NHIRD. The LHID consists of one million insured people who were selected using random sampling between 1996 and 2000. Before releasing the database to researchers, the NHRI created a scrambled and anonymous identification number to identify each insured person. Each record in the database includes sex, birth date, occupation and medical services records. This study was exempted by the Institutional Review Board of China Medical University in central Taiwan (CMU-REC-101-012). We recorded the disease history in inpatient and outpatient files and defined the disease according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).
Study population
This study was a population-based retrospective cohort study. The anxiety cohort consisted of patients newly diagnosed with anxiety (ICD-9-CM 300.0, 300.2, 300.3, 308.3 and 309.81)
between 2000 and 2009; the index date was set as the date on which anxiety was diagnosed. The anxiety cohort was also separated into two subcohorts: (i) individuals with general anxiety (ICD-9-CM 300.0, 300.2, 308.3 and 309.81) and (ii) individuals with obsessive–compulsive disorder (OCD; ICD-9-CM 300.3). The comparison cohort contained people in the LHID who had never been diagnosed with anxiety, and were twofold frequency matched according to age (within 5-year intervals) and sex. The index dates of the comparison cohort were randomly assigned a day and month in the same index year as the anxiety cohort. People who had received an LSC diagnosis before the index date were excluded. An interesting event in this study was the development of LSC (ICD-9-CM 698.3). Follow-up was terminated when a person withdrew from the insurance plan, an event occurred, or on 31 December 2013.
Comorbidity history was considered a potential confounding factor. The comorbidities included hypertension
(ICD-9-CM 401–405), diabetes (ICD-9-CM 250), depression (ICD-9-CM 296.2, 296.3, 300.4 and 311) and schizophrenia (ICD-9-CM 295).
Statistical analysis
We calculated the number and proportion of categorical variables, and the mean and SD of age to determine the distributions of the comparison and anxiety cohorts. To assess the difference between the two cohorts, we conducted a t-test for age and v2-tests for sex and comorbidities. The incidence of LSC in the
two cohorts was calculated by dividing the total number of LSC occurrences by the total sum of follow-up years in each cohort by 10 000 person-years. The cumulative LSC incidence curves were measured using the Kaplan–Meier method, and the difference between the two curves was tested using the log-rank test.
Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using the Cox proportional hazards model to determine the influence of anxiety on the risk of LSC.
All data management and statistical analyses were performed using SAS 9.3 software (SAS Institute Inc., Cary, NC, U.S.A.). The cumulative incidence curves were plotted using R software
(http://www.r-project.org/). The significance level was defined using a two-sided P-value < 0_05.
Results
This study included a cohort of 69 386 patients with anxiety who were an average of 49_5 years old and predominantly female (62_3%). The comparison cohort had the same mean
age and sex ratio (P > 0_05, Table 1). The proportion of LSCassociated comorbidity in the anxiety cohort was much higher
than in the comparison cohort (P < 0_0001).
Table 2 shows an analysis of LSC risk stratified according to the presence or absence of OCD in the general anxiety study participants. The incidence of LSC in the anxiety cohort was 25_68 per 10 000 person-years, which was 1_45-fold greater than in the comparison cohort (17_77 per 10 000 personyears) (Fig. 1). After adjusting for age, sex and LSC-associated
comorbidities, people with anxiety were shown to have a 1_41-fold greater risk of developing LSC than people in the comparison cohort (HR 1_41, 95% CI 1_30–1_52). When the participants with general anxiety disorders were further
divided into the OCD and non-OCD subgroups, the OCD subgroup had a 1_72-fold higher risk of developing LSC than the
comparison cohort, and the non-OCD subgroup had a 1_37-fold higher risk of developing LSC.
Table 3 shows an analysis of LSC risk stratified according to demographics and comorbidities. Among people younger than 40 years old, people with anxiety had a 1_54-fold increased risk of developing LSC compared with people without anxiety. Among other age groups (40–49, 50–59 and ≥ 60 years), the risk of developing LSC was approximately 1_3-fold higher among people with anxiety compared with people without anxiety. Men with anxiety had a 1_47-fold increased risk of developing LSC compared with men without anxiety (HR 1_47, 95% CI 1_30–1_66). In addition, women with anxiety had only a 1_36-fold increased risk of developing LSC compared with women without anxiety (HR 1_36, 95% CI 1_22–
1_50). However, the differences between men and women were not significant (overlapping 95% CIs). In addition, people with anxiety had a relatively higher risk of developing LSC than people without anxiety, especially in the population
without hypertension (HR 1_52, 95% CI 1_38–1_68), with diabetes (HR 1_55, 95% CI 1_24–1_93), without depression (HR 1_40, 95% CI 1_29–1_52) and without schizophrenia (HR 1_38, 95% CI 1_27–1_49).
Discussion
This is the first nationwide population-based cohort study to estimate the effects of anxiety disorder on LSC prevalence. Anxiety disorder, a major psychological disease, is both an aggravating factor and a consequence of pruritus and scratching, the major symptoms of LSC. The increased anxiety levels can increase the severity of pruritic dermatitis, and one of the main psychiatric sequelae secondary to chronic pruritus is anxiety.7
Anxiety and mood disorders are common in the general
population, and the difference in their prevalence is not substantial. The lifetime prevalence rate of anxiety disorder is 4–
7%.8 Patients with anxiety disorder often worry uncontrollably
and suffer from physiological symptoms such as sleep disturbance, muscle tension and difficulty concentrating. These people
may not be able to attain their social and occupational
potential. One study indicated that 38% of people with anxiety exhibited an inability to work and loss of role functioning for an average of 6_3 days per month because of the disorder.
9 Some patients with anxiety disorder have an increased
risk of suicide.10 Anxiety disorders may go undiagnosed
because of the physical symptoms of anxiety and the stigma of mental illness. The large sample size of our study, drawn from a nationwide population-based dataset, strengthens the statistical power of our study regarding the association
between anxiety disorder and LSC. As LSC is a type of neurodermatitis, treatment requires the
work of dermatologists, psychiatrists, psychologists and caregivers who can educate patients to address the psychological
and physical aspects of the disease. To treat the dermatological aspects of LSC, anti-inflammatory agents are useful. For widespread skin lesions, phototherapy treatment may be considered.
Moisturizers may help in repairing the skin. The
application of moisturizers distracts the patient from the sensation of pruritus and is an aspect of the behavioural replacement
for scratching. Covering local LSC lesions is crucial for
increasing the healing rate because it both enhances the effects of the topical agent and provides protection from further
trauma. The most critical aspect of LSC treatment is breaking
the pruritus scratching cycle by making it difficult to scratch the lesions and reducing the euphoria derived from scratching.
7,11 To treat the psychological aspects of LSC, education,
support and behavioural therapy can strengthen a person’s psychological ability to control the scratch process.12 Patients
with neurosis or psychosis should be treated by a psychologist or psychiatrist. Pharmacotherapeutic agents may be useful; however, most LSC-affected patients prefer to see a dermatologist, and are unwilling to see a psychiatrist. In such cases a
multidisciplinary medical unit is necessary.13
It is our expectation that OCD could result in a significantly higher risk of developing LSC. When the participants with general anxiety disorders were further divided into OCD and non-OCD subgroups (Table 2), individuals with OCD (HR 1_72, 95% CI 1_03–2_88) had a significantly higher risk of developing LSC than patients without OCD (HR 1_37, 95% CI 1_26–1_48, P = 0_0395). Therefore, we think the level of anxiety could affect the outcome or prognosis of LSC. In addition, LSC affects the female population predominantly. However, in our study, male patients (1_47-fold) with anxiety disorder had a higher (but not significantly) risk of developing LSC compared with female patients (1_36-fold) (Table 3). A possible
explanation is that men who seek psychological help may have a worse health status than women because of different personalities. Perhaps more follow-up or a different therapeutic strategy
should be given to male patients with anxiety disorders
with or without LSC. However, further studies are recommended to ascertain the reasons scientifically.
A limitation of our study is the inconsistent diagnosis of anxiety disorders. In general practice, not all anxiety disorders are diagnosed strictly by psychiatrists in accordance with the
Diagnostic and Statistical Manual of Mental Disorders, fourth edition criteria. Some patients may have been diagnosed with anxiety disorders
used in this study therefore consisted partly of people without anxiety disorder. Therefore, some of our study results could be interpreted as the increased prevalence of LSC in people with a personality tending to be anxious.
Another major limitation of our study is that the evidence
derived from a cohort study is generally of a lower methodological quality than that from randomized trials because a
cohort study design is subject to many biases related to adjustments for confounding factors. Despite our meticulous study
design with adequate controls for confounding factors, a primary limitation was that bias could still remain because of
possible unmeasured or unknown confounders.
In conclusion, our nationwide population-based retrospective cohort study provides evidence for the increased prevalence of LSC among people with anxiety disorders compared with the general population without anxiety disorders. The management of LSC should focus on its psychological aspects to facilitate favourable outcomes.
