Association between dry eye symptoms and suicidal ideation in a Korean adult population

Purpose This study assessed the association of dry eyes with depression and suicidal ideation in a Korean adult population. Methods Data from 16408 participants (6972 men and 9436 women) aged ≥ 19 years included in the fifth Korea National Health and Nutrition Examination Survey, conducted from 2010–2012, were analyzed. For dry eyes, surveys of previous diagnosis of dry eye disease (DED) by an ophthalmologist and experience of subjective dry eye symptoms were separately used. Diagnosis of depression and suicidal ideation were obtained via responses to an interviewer-assisted questionnaire, and questions were asked in a closed-ended response format. Logistic regression was used to examine the associations between dry eyes, depression, and suicidal ideation. Results DED diagnosis exhibited an odds ratio (OR) of 1.32 (95% confidence interval [CI] 1.11–1.57) for depression and 1.24 (95% CI 1.05–1.48) for suicidal ideation compared to those without DED, after adjusting for sex, age, education, occupation, household income, body mass index, smoking behavior, alcohol consumption, physical activity, hypertension, diabetes, dyslipidemia, thyroid diseases, major cardiovascular disease, and cancer. Similarly, the adjusted OR (95% CI) of dry eye symptoms was 1.50 (95% CI 1.30–1.73) for depression and 1.47 (95% CI 1.27–1.70) for suicidal ideation. Conclusion Our findings suggest that dry eyes (either DED diagnosis or dry eye symptoms) may be associated with the prevalence of depression and suicidal ideation in the Korean adult population.


Conclusion
Our findings suggest that dry eyes (either DED diagnosis or dry eye symptoms) may be associated with the prevalence of depression and suicidal ideation in the Korean adult population. PLOS

Introduction
Worldwide, more than one million people die by suicide every year [1], and suicide rates are expected to increase from 1.8% in 1998 to 2.4% in 2020 [2]. The suicide rate of South Korea is highest among the Organization for Economic Cooperation and Development (OECD) countries recently. As the burden due to suicide increased, suicide became a major health issue in South Korea [3]. While suicide may be associated with biological, behavioral, physical, sociocultural, and environmental factors, mental illness (including depression, bipolar disease, schizophrenia, and others) is the main cause of suicide [4][5][6]. Nevertheless, while the link between suicide and mental illness is well established, studies have yet to thoroughly investigate whether ocular diseases are linked with suicide. Dry eye disease (DED) is one of the most frequently encountered ocular morbidities worldwide. The prevalence of DED is estimated to range from 4.3% to 73.5% for either clinically diagnosed DED or cases that experienced symptoms and is comparably higher in Asian populations than in Western populations [7][8][9][10][11][12][13]. Vision plays an important role in almost every task that humans perform, at all stages of life, regardless of age. The health of the human eye depends on the flow of tears, which provide constant moisture and lubrication to maintain vision and comfort. However, once tear flow is impaired, the human eye can experience symptoms of redness, stinging, burning, or a scratchy sensation, which lead to eye fatigue, visual disturbance, and even impaired quality of life (QOL) [13][14][15][16][17][18]. Additionally, patients with DED frequently report significant disturbances in their psychiatric state, showing symptoms of anxiety and depression [19][20][21].
Many suicidal attempts happen impulsively in moments of crisis, along with a breakdown in one's ability to deal with life stresses, such as chronic pain and illness [22]. However, beyond depression, relatively few studies investigated whether prolonged symptoms of DED may cause suicidal ideation. Therefore, we aimed to examine the association between DED, depression, and suicidal ideation using nationally representative data targeting the entire Korean population. We hypothesized that equivalent results would be observed as reported by previous studies concerning the association between DED and depression, and reports of suicidal ideation would be higher in participants with DED.

Study population
This study used data from the fifth Korea National Health and Nutrition Examination Survey (KNHANES-V), which was conducted from 2010-2012. The KNHANES is a nationally representative cross-sectional survey of the non-institutionalized civilian population. The survey is repeated annually by the Korea Centers for Disease Control and Prevention (KCDC). A complex, stratified, multistage probability sampling design based on region and households was applied in this survey to represent the national population of South Korea. The survey consisted of three parts: a health interview survey, a health examination survey, and a nutrition survey. Survey questionnaires were self-administered or conducted by trained staff members depending on the section, and examinations were performed by highly trained medical personnel. Further details of the survey have been described elsewhere [23]. The survey procedures were carried out in accordance with the Declaration of Helsinki, and participants who were willing to participate in the survey were required to sign informed consent forms approved by the Institutional Review Board of the KCDC. Data are available from the KCDC (https://knhanes.cdc.go.kr/ knhanes/eng/index.do). The design of this study was approved by the Institutional review board of Yonsei University Graduate School of Public Health in Seoul, Korea.
A total of 18571 participants aged 19 years or older were initially included over the three years for analyses. We further excluded 1921 participants that were lacking information in dry eyes, depression diagnosis, suicidal ideation, and other covariates. Since previous studies indicated that patients taking antidepressants had a higher chance of developing DED [10,24,25], we additionally excluded 242 participants who were receiving treatment for depression. Lastly, after excluding non-responses on dry eye questionnaires, 6972 men and 9436 women were analyzed in relation to their DED diagnosis, while 6718 men and 9146 women were analyzed according to their dry eye symptoms. All analyses were performed separately according to the presence of DED diagnosis and dry eye symptoms (not mutually exclusive).

Assessment of dry eyes
Examination for eye diseases was conducted by ophthalmologists designated by the Korean Ophthalmological Society (KOS). In order to ensure accurate examinations, ophthalmologists were educated and trained twice a year by the KCDC and the KOS. In 2010, the KNHANES introduced dry eye questionnaires to evaluate the prevalence of dry eyes. Due to the survey design, presence of DED and dry eye symptoms were observed by closed-ended response questionnaires. Ophthalmologists interviewed each participant on whether they had been diagnosed with DED by an ophthalmologist before. To increase accuracy of the data collected, participants were also asked whether they experienced frequent symptoms of dryness or irritation of the eye. Participants who responded "yes" to the above questionnaires were either assigned to DED diagnosis or dry eye symptoms. Those who answered "no" were classified as a control.

Assessment of depression and suicidal ideation
Diagnosis of depression was assessed by interviewing each participant on whether they had previously been diagnosed with depression by a psychiatrist. To increase the accuracy of the data collected, participants were first asked whether they had been diagnosed with depression so far which was followed up "by a psychiatrist." Suicidal ideation was assessed by asking participants whether they had thought about dying within the last year.

Covariates
Age, sex, socioeconomic status including education, occupation, and household income, body mass index (BMI), smoking behavior, alcohol consumption, physical activity, diagnosis of hypertension, diabetes, dyslipidemia, thyroid diseases, major CVD, and cancer were considered as covariates in the present analyses. Age and BMI were analyzed as continuous variables. Education was classified according to the highest graduate level of school (middle school/high school/college), occupation was classified according to the standard occupation classification, and household income was classified in quartiles. Smoking behavior was categorized as nonsmoker (never smoked for their entire life) versus former or current smoker. Alcohol consumption was categorized as non-drinker (never consumed alcohol for their entire life and/or within the past year) versus drinker (consumption of alcohol once per week or less or twice per week or more). Physical activity was categorized according to frequency of exercise: none (no physical activity within the past week) versus at least once per week or more. Diagnosis of other diseases were obtained by interviewer-assisted questionnaire.

Statistical Analysis
All statistical analyses in this study were conducted using sampling weights assigned to each participant provided by the KNHANES. Student's t-test and chi-square test were used to analyze the general characteristics of the study population according to presence of dry eyes (DED diagnosis and dry eye symptoms). Serial multiple logistic regression models were used to examine the independent associations between dry eye, depression, and suicidal ideation. All analyses were separated by sex. A P-value < 0.05 was considered statistically significant. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Inc. Cary, NC).

Characteristics of the study population according to the presence of dry eyes
DED diagnosed men had a higher household income level, higher frequency of white-collar and lower frequency in pink, green, and grey-collar occupations. They showed a lower frequency of alcohol consumption and higher frequency of physical activity compared to men without DED diagnosis (Table 1). DED diagnosed men had a higher frequency of dyslipidemia, and for mental health components, perceived stress, depression diagnosis, and suicidal ideation did not differ between the two groups. The comparison of men with dry eye symptoms and without, showed a similar distribution with DED diagnosis. Interestingly, men with dry eye symptoms noted a significantly higher frequency in the mental health components. Men experiencing dry eye symptoms reported significantly higher levels of stress perception, (28.7% versus 24.0%, p = 0.03), prevalence of depression (9.3% versus 6.1%, p = 0.003) and suicidal ideation (12.0% versus 9.0%, p = 0.04) than men without symptoms.
DED diagnosed women were more frequent in agricultural workers, and showed significant differences in smoking behavior, perceived stress, and depression diagnosis compared to women without DED diagnosis, although no significant difference was observed in suicidal ideation ( Table 2). Regarding dry eye symptoms, significant differences were observed in mental health components, similarly to men. Women with either DED diagnosis or dry eye symptoms had a higher frequency of dyslipidemia and thyroid disesases. Women with dry eye symptoms were likely to perceive more stress (35.7% versus 28.1%, p<0.001), have depression (23.2% versus 17.5%, p<0.001), and suicidal ideation (22.0% versus 16.6%, p<0.001) than women without symptoms. Table 3 shows the odds ratio (OR) and 95% confidence interval (CI) of depression when unadjusted and adjusted for age, socioeconomic status (education, occupation, household income), BMI, smoking behavior, alcohol consumption, physical activity, history of hypertension, diabetes, dyslipidemia, thyroid diseases, major CVD, and cancer. Regardless of sex, participants diagnosed with DED were associated with a higher prevalence of depression (OR 1.32; 95% CI 1.11-1.57), compared to non-DED participants. After stratifying for sex, the association remained significant for women (OR 1.31; 95% CI 1.08-1.57), though was no longer significant in men (OR 1.32; 95% CI 0.84-2.09).

Association between dry eyes and depression
Similar patterns were observed for participants experiencing dry eye symptoms. Overall, participants with dry eye symptoms showed an OR of 1.50 (95% CI 1.30-1.73) for depression than those without dry eye symptoms. The ORs for depression with dry eye symptoms were 1.55 (95% CI 1.13-2.13) for men and 1.47 (95% CI 1.25-1.72) for women, respectively. Table 4 shows the OR and 95% CI of suicidal ideation when unadjusted and adjusted for potential confounders. Overall, participants diagnosed with DED had an OR of 1.24 (95% CI

Discussion
In the Korean population, we found that a previous diagnosis of DED, or frequent experiences of dry eye symptoms were significantly associated with depression and suicidal ideation. However, for the association between DED diagnosis and suicidal ideation, statistical significance was observed in the overall participants and for women, when separated by sex in the analysis. According to previous studies [7,26,27], women often reported a higher prevalence of DED than men did, and our results support these findings. Suicide rates in South Korea have remained the highest among the OECD countries for 10 consecutive years since 2002, and were reported to be more than double the average of all OECD countries in 2012 (28.1 versus 12.1 for 100,000 people) [1]. Therefore, the high suicide rate in South Korea and a high prevalence of DED, which has been noted in Asian populations, makes South Korea a suitable country in which to analyze the association between dry eye and suicidal ideation.
The results on the relationship between DED and depression in this study are consistent with those in previous studies, although, relatively few studies examined the relationship between DED and suicidal ideation. One case-control study reported a higher prevalence of anxiety and depression in DED patients than the control group [19]. Additionally, the Beijing Eye study determined that depression was more prevalent in older patients with DED than those without [28]. Epidemiological studies have revealed that major depressive disorders are strong predictors of suicide attempts compared to other psychiatric disorders, including anxiety, agitation, and poor behavioral control; and 60% of those who commit suicide have a depressive disorder [29,30]. Depression can also affect an individual to think about suicide.  For example, impaired health behavior and low physical health status can cause suicidal ideation through depressive symptoms [5]. In light of the findings in our study, we suggest that depression may act as a mediator between DED and suicidal ideation. To assess the role of depressive symptoms in the association between dry eye and suicidal ideation, we conducted further analyses. Additional adjustment for depressive symptoms weakened the association between DED diagnosis and suicidal ideation, but did not affect the association between dry eye symptoms and suicide ideation (S1 and S2 Tables). These findings suggest that depressive symptoms may not be the only mechanism which links dry eye and suicidal ideation. Further studies are required to elucidate the role of depression between dry eye and suicidal ideation.
Prolonged dry eye symptoms can lead to chronic pain, which in turn may cause feelings of depression and suicidal thoughts. Patients with chronic pain often develop complications of physical dysfunction and an altered psychological state, which negatively affect a person's everyday living and QOL [31,32]. In a parallel manner, persistent dry eye symptoms may induce a depressive mood. A previous study demonstrated that the presence of ocular surface symptoms decreased the ability to perform daily activities, work capacities, and emotional well-being [33]. Other studies have shown that people who experience chronic pain are more likely to develop depression than those who do not and that suicidal ideation is highly co-morbid [22,34,35]. Additionally, patients with chronic illness and pain have an increased risk of suicidal ideation and suicide attempt, even after adjusting for mental disorders [22]. Accordingly, the chronic impact of dry eye symptoms on various aspects of life may lead to the development of depression, and as the symptoms worsen, suicide might arise as an alternative to escape unbearable pain. Unfortunately, we could not evaluate these mediations, because the KNHANES did not assessed chronic pain in this period.
Our results showed that experiencing dry eye symptoms was more strongly associated with suicidal ideation than DED diagnosis. Among those who experienced dry eye symptoms, 1,375 participants were diagnosed with DED, while 1,440 participants were not. Several participants diagnosed with DED would have been receiving ophthalmic treatments to alleviate their pain, but participants who were undiagnosed and who had dry eye symptoms may have developed depression and suicidal thoughts. Furthermore, the total number of participants with dry eye symptoms almost doubled the number of participants with DED diagnosis. Thus, the greater sample size in this study might have increased the statistical significance of the association between dry eye symptoms and suicidal ideation. Therefore, according to the results and by the nature of the data analyzed, dry eye symptoms may have revealed a higher association with suicidal ideation compared to DED diagnosis. Previous studies have demonstrated that DED patients have significantly lower QOL and vision-related QOL. In both the Women's Health Study and Physician's Health Study, DED patients demonstrated problems with reading, carrying out professional work, using a computer, watching TV, and driving during the day and night [16]. Another study that assessed general QOL in DED patients using the Short Form 36 Health Survey, which is a survey for measuring an individual's overall wellbeing not specific to dry eyes, indicated that DED negatively affects everyday life by disrupting daily activities, and even causes great deterioration in mental health [15]. This study also demonstrated that in mild DED, a patient has the potential ability to overcome the dry eye symptoms, but this becomes limited as the severity of DED symptoms increases. In a utility assessment study using a time-trade-off method, participants responded that living 10 years with severe DED is comparable to living 1.6 years without DED [18]. Other studies have indicated that the presence of DED greatly reduces QOL and that a reduced QOL is associated with suicidal ideation and attempt [36,37].
Proven risk factors of DED may be relevant to suicide. The Osaka study found that low physical activity and sedentary behavior were associated with DED [38]. Low physical activity was also linked with suicidal ideation. One study noted that low physical activity was correlated with suicidal ideation, and an absence of regular walking increased suicidal ideation [39]. Interestingly, diabetes [40], long working hours [41], smoking [42], air pollution [43], etc. have all been identified as risk factors of both DED and suicide.
Sleep disorders, such as insomnia or nightmare, are associated with the risk of suicide [44][45][46]. Sleep disorders are frequently observed in patients with DED, suggesting a possibility of sleep disorder as a confounder between DED and suicidal ideation. [47,48]. In the KNHA-NES-V data, there was only information of sleep duration. However, our results did not change after additionally adjusting sleep duration (S3 and S4 Tables).
Although our results showed that participants with DED were more likely to experience depression and suicidal ideation, previous studies have indicated that patients with depression are susceptible to DED. Consistent results on the relationship between DED and the use of antidepressants have been reported in the literature [10,25]. A cross-sectional study of psychiatric patients noted a higher prevalence of DED among patients with depression and/or anxiety disorders [49]. This study further observed that DED was associated with duration of psychiatric disease and use of antidepressant medications. Another study found that approximately 37% of participants who experienced symptoms of DED were taking antidepressants [7]. In our study, however, we excluded participants receiving depression treatments in order to assess the direct relationships between DED, depression, and suicidal ideation.
Our study has several limitations that need to be considered when interpreting the results. First, the association between DED and depression or suicidal ideation cannot be identified as causal due to the cross-sectional study design. Second, diagnosis of DED and dry eye symptoms were assessed by interviews rather than objective measurements. However, a survey conducted by the Korean Corneal Disease Study Group stated that Korean corneal subspecialists use either the diagnostic criteria of the International Dry Eye Workshop or the Dysfunctional Tear Syndrome Study Group guidelines to diagnose DED [38]. This statement provides evidence that participants diagnosed with DED were at least examined under internationally accepted criteria. Recent studies established the significance of using short questionnaires for determining the presence of DED in large epidemiological studies where objective examinations may not be practical [50], and demonstrated that asking about dryness and irritation was equivalent to asking a patient about up to 16 symptoms of DED [27]. Moreover, the questionnaires in the KNHANES were equivalent to one of the most frequently used short series of DED [39]. There are few evidences of the effects of the duration and severity of DED on depression or suicidal ideations. Unfortunately, KNHANES data do not have data of the duration or severity of DED, so we could not provide more informative evidences. Finally, the diagnosis of depression and suicidal ideation were obtained by interview using a single questionnaire. The validity and reliability of psychometric measures cannot be guaranteed. However, to increase the accuracy, participants were asked whether they have been diagnosed with depression "by a psychiatrist". Despite these limitations, this is one of the first studies to report an association between dry eye and suicidal ideation using a large nationwide dataset with sufficient statistical power. The results of our analyses suggest that dry eye symptoms can be an aggravating factor for depression and suicidal ideation.
In conclusion, dry eye symptoms were associated with higher risks of depression and suicidal ideation in the Korean adult population. Our results suggest that DED should be viewed not only as an eye disorder, but also as a condition that affects mental health. Ophthalmologists may provide better treatment to patients with DED by evaluating their psychiatric status. However, further study is required to verify the prospective causal effect of DED on depression and suicidal ideation in a large population-based sample.
Supporting information S1 Table. The results of multivariable logistic regression analyses for depression and suicidal ideation with DED diagnosis. (DOCX) S2 Table. The results of multivariable logistic regression analyses for depression and suicidal ideation with dry eye symptoms. (DOCX) S3 Table. The association between dry eye and suicidal ideation after adjusting depressive symptoms. (DOCX) S4 Table. The association between dry eye and suicidal ideation after adjusting sleep duration. (DOCX)