**4. Discussion**

In this study, we investigated the relationship between the severity of dry eye disease and systemic comorbidities. Aging is the most clearly documented risk factor for dry eye disease [6], and the elderly tend to be a ffected by multiple comorbidities. As expected, the average age of subjects in our study was relatively high (62.6 years on average), and approximately half of the patients had systemic comorbidities. Prevalent systemic comorbidities were hypertension, insomnia and depression. We found that patients with dry eye disease and any systemic comorbidity exhibited significantly worse utility, as estimated by HUI-3, than those without. In addition, those with systemic comorbidities had a higher severity of dry eye disease, as demonstrated by objective parameters, such as the TBUT and Schirmer value. These correlations remained statistically significant after adjustment for age. Schein et al. reported that the presence of many systemic comorbidities is strongly associated with reported symptoms of dry eye disease [5], although no association was seen between systemic comorbidity and Schirmer or Rose Bengal test scores. Our findings sugges<sup>t</sup> that the presence of systemic comorbidities may compromise the ocular surface, as well as the QOL of patients with dry eye disease. However, systemic comorbidities include numerous conditions with di fferent pathophysiology, which a ffect various parts of the body. Our findings are based on a statistical and epidemiological approach and therefore must be re-examined cautiously in further studies.

In this study, we found that patients with dry eye disease and depression showed a significantly shorter TBUT after adjustment for age. Interestingly, those taking antidepressants showed a significantly lower Schirmer value, in addition to a shorter TBUT, after such adjustments. These results sugges<sup>t</sup> that depression and/or anti-depressant use is associated with poor tear film stability, and that antidepressant use may worsen tear secretion in addition to tear stability, which is an indicator of dry eye severity. Previous studies, mostly involving epidemiological questionnaires, have also reported a relationship between depression and dry eye disease [23–26]. In a comparison of depressive and control groups, one case-control study revealed significantly lower Schirmer scores and shorter TBUT, as well as a consistently higher Oxford scores in individuals with depression [27]. Similarly, we demonstrated an association between depression/anti-depressant use and dry eye severity by undertaking objective measurements, although the mechanisms involved in this association remain unknown.

Another interesting finding in our study is that patients with dry eye disease and insomnia showed a significantly shorter TBUT after adjustment for age. Several investigators have recently reported the association of sleep disturbances, like insomnia, and dry eye disease [28–30]. Our results also support such an association. Insomnia is a common problem in the elderly and sedative drugs, including benzodiazepines, are frequently prescribed for this condition in Japan. The mechanism involved in the association between insomnia and dry eye disease may be derived from insomnia itself. Alternatively, the intake of sedative drugs may have some influence. Further studies are needed to clarify this issue.

In addition, we found that FRD was significantly associated with dry eye disease and was a comorbidity with insomnia or depression, even after adjusting for age. The presence of FRD, including conjunctivochalasis and lid wiper epitheliopathy, is a predisposing factor for dry eye disease and its severity. To our knowledge, an association between FRD and depression or insomnia has not ye<sup>t</sup> been reported. We also noted that antidepressant users had decreased tear secretions. Since FRDs are associated with lower tear volume and quality (e.g., secreted mucin), dry eye disease is more likely to be associated with FRD when depression and insomnia are also present than when they are not. Another possible explanation is that anti-depressant and sedative drugs may induce blepharospasm [31,32], as blinking problems induced by drugs may result in the development of FRDs. Although the cause is not known, our study suggests that anti-depressant use may reduce tear secretion and worsen ocular surface conditions.

In this study, we demonstrated that depression and insomnia comorbidity with dry eye disease increased dry eye severity and that, in turn, ocular symptoms may worsen depression/insomnia, thus creating a vicious circle. Therefore, ophthalmologists should be aware of patients' systemic comorbidities in the diagnosis and treatment of dry eye disease. Similarly, psychologists should be aware of patients' eye conditions when considering mental health management. We feel that it is especially important for psychiatrists to take this into account, particularly when prescribing antidepressants, which may aggravate dry eye signs.

Our findings must be considered along with several limitations. First, detailed patient medical information could not be collected regarding the severity of depression, details of medications, duration of systemic comorbidities and duration of oral medicine use. This information is important for establishing the underlying mechanisms of dry eye severity and systemic comorbidities. Dry eye disease is common in patients with depression, especially those of older age who have been experiencing depression and taking antidepressant medication for longer periods. Therefore, future studies should focus on patients with depression and dry eye disease. Second, our study was based on a cross-sectional design; therefore, it is di fficult to ascertain whether the observed associations are due to causal e ffects. Third, we included patients with dry eye disease who had been receiving various treatments. Therefore, there may have been di fferences in the clinical presentation of these patients.

In conclusion, in the clinical setting ophthalmologists should pay careful attention to the systemic comorbidities of patients with dry eye disease, and especially of older adults who have a higher prevalence of such comorbidities. Comorbidities including depression, insomnia and the use of antidepressants are associated with dry eye disease severity.

**Author Contributions:** Conceptualization, M.K., M.Y. and Y.H.; data curation, C.S.; formal analysis, Y.H.; funding acquisition, M.Y. and K.S.; investigation, M.K. and C.S.; methodology, M.Y.; project administration, M.Y.; supervision, N.Y. and K.T.; writing—original draft, M.K.; writing—review & editing, M.K., M.Y., C.S., K.S., M.U., Y.H., N.Y. and K.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Japan Dry Eye Society, Tokyo, Japan, and Santen Pharmaceutical Co., Ltd., Osaka, Japan.

**Conflicts of Interest:** K.S. is an employee of Santen Pharmaceutical Co., Ltd.; M.Y. is a consultant for Otsuka Pharmaceutical Co., Ltd. (Tokyo, Japan) and Johnson & Johnson Vision Care Co. (Paranaque City, Philippines); N.Y. is a speaker's bureau member of Santen Pharmaceutical Co., Ltd. and Otsuka Pharmaceutical Co., Ltd., and is a consultant of Rohto Co., Ltd. (Osaka, Japan) and Alcon Japan Co., Ltd. (Tokyo, Japan), and has patents for ophthalmologic apparatus with Kowa Co., Ltd. (Nagoya, Japan), and is a consultant for Kissei Co., Ltd. (Matsumoto City, Japan) and Rohto Co., Ltd.; K.T. is a consultant, speaker's bureau member, and gran<sup>t</sup> recipient of Santen Pharmaceutical Co., Ltd., as well as a speaker's bureau member and gran<sup>t</sup> recipient of Otsuka Pharmaceutical Co., Ltd. The authors report no other conflicts of interest for this work. This research was conducted by a joint study organization from the Dry Eye Society and Santen Pharmaceutical Co., Ltd. Both organizations contributed to the creation of the documents, including the study protocol; managemen<sup>t</sup> of the study's progress; both provided information and support to the study sites; disclosed the outcome of the study; and registered and updated the present study in the public registration system.
