Impact of signs and symptoms of dry eye disease on health-related quality of life: a cross-sectional population study among older adults

A healthy ocular surface is essential for clear vision, comfort, and prevention of various eye disorders. The ocular surface is covered by a thin fluid layer, the tear film, which lubricates, protects, and supplies nutrients and optical clarity for the ocular surface [1]. Imbalances in tear film homeostasis can lead to ocular surface disease, such as dry eye disease (DED) caused by reduced tear volume and/or increased tear evaporation. DED is characterized by dryness, burning, redness, excessive tearing, blurry vision, light sensitivity, and overall discomfort of the eye. Consequently, recent large questionnaire-based studies from diverse regions of the world indicate that DED is associated with worsened general health-related QoL (HRQoL) and that QoL scores worsen with more severe forms of the disease [2‐9]. The prevalence of DED varies from 5 to 50%, depending on geographic region, study population, as well as definitions and diagnostic criteria employed [10]. DED is more common in women and older individuals and is associated with, e.g., certain autoimmune diseases, depression, and anxiety [10‐13]. The economic burden of DED is also significant, mostly due to loss of work productivity [14].

Ocular surface health, including the presence of DED, can be evaluated by clinical examination and self-reported outcome questionnaires. Traditional clinical tests for DED include slit lamp examination, different ocular surface staining methods to detect defects on the corneal or conjunctival epithelium (Fluorescein, Lissamine green, and Rose Bengal), and tear film examinations to test tear stability and production (tear break-up time [TBUT], Schirmer test). It is also recommended to evaluate lid margin and meibomian gland health and test tear osmolarity. Multiple questionnaires have been developed for assessing the subjective symptoms of DED [14‐16], the most prevalent being the Ocular Surface Disease Index (OSDI) questionnaire. Due to the multifactorial nature of DED, harmonized diagnostic criteria for DED are still missing; different continentals have different preferences, and the availability of diagnostic tools, and practices varies. These inconsistencies may hinder data pooling and meta-analyses, affecting the cohesive understanding of DED’s impact on QoL and the development of best practices. Recognizing QoL impacts in DED patients enables clinicians to tailor treatments to individual needs. For example, patients with work-related eye strain might benefit more from short breaks, and environment modifications, while those with significant inflammation may need anti-inflammatory medication. This patient-centered approach enhances treatment adherence and overall QoL.

Recently, we performed the first analysis of ocular surface health in Finland and showed that in elderly people, the prevalence of DED ranged from 10 to 33% based on the diagnostic criteria employed [17]. Furthermore, we observed distinct disparities in signs and symptoms of DED between men and women. The objective of the present study was to assess the association between QoL and ocular surface health in the same Finnish study cohort using both self-reported symptoms and clinical signs of DED and accounting for varying diagnostic criteria for DED. We hypothesized that sex differences influence the relationship between DED and QoL. This is the first population-based study combining general HRQoL data with objective clinical measures of DED.

Ophthalmological examination results were obtained from > 99% of the 601 participants, with only single values missing from different tests. The QoL mean scores, BCVA, and subjective ocular surface health measures of the study population are described in Table 1. QoL mean scores revealed significant sex differences in mental health (SF-36 MCS and BDI-II). In addition to SF-36 MCS and BDI-II scores, SF-36 dimensions measuring energy/fatigue, emotional well-being, social functioning, and pain, indicated better QoL scores in men than women. Men had slightly better vision (BCVA) and experienced fewer DED symptoms (subjective eye dryness and OSDI) compared with women. Detailed results of the ophthalmologic examination in this study population and the data on the use of the medication for DED have been published previously [17].

Large population-based studies on DED often depend on data reported by the participants [2‐9]. The uniqueness of our population study lies in its comprehensive methodology, i.e., both subjective and objective measures of DED were compared with HRQoL data obtained from two different generic questionnaires and a separate depression questionnaire. We also used three different diagnostic criteria for DED in the QoL analyses; OSDI diagnostic criteria were based on self-reported DED symptoms, whereas OSD diagnosis considered symptoms or clinical signs, and DE required both symptoms and signs for positive DED diagnosis. Consistent with previously published findings, women exhibited a higher prevalence of DED compared with men and this trend was observed across all diagnostic categories. Regardless of the diagnostic criteria used DED was consistently associated with reduced mean QoL scores, although the findings displayed some degree of variability. Overall, OSDI, the symptom-based diagnosis generated the largest decrease in QoL mean scores among individuals with or without DED. OSD, and DE diagnoses indicated worsening of QoL among women but not in men. The different diagnosis criteria for DED vary in their sensitivity and specificity, leading to differences in classification of individuals and, also impacting the QoL data interpretation. Diagnostic criteria also guide treatment, so inconsistent thresholds may result in disparities in the management of DED. Hence, the effects of different diagnosis criteria for DED and sex should be acknowledged when comparing findings from studies on DED-related QoL.

Sex and sex hormones play a significant role in DED, as emphasized in comprehensive reviews by Sullivan [27] and Matossian [28]. In addition to DED symptoms, more attention has recently been paid to sex-specific clinical signs among people with DED [29‐32]. There is some disparity in findings, but overall, in DED studies women tended to have higher corneal staining [29, 32], shorter tear break-up time (TBUT) [29, 30, 32], and increased tear osmolarity [29, 30], while men experienced more conjunctival redness [31]. The first stages of our analysis revealed that several individual DED symptoms and signs correlated with QoL scores. In all participants, higher OSDI and corneal and conjunctival staining grades as well as short NIBUT were associated with worse QoL scores. At the population level, increased DED symptoms have been associated with worse general health- and vision-related QoL [9‐18, 33, 34], but very little data are available on how clinical signs may affect these outcomes. A population-based cross-sectional study of 229 participants [33] concluded that none of the clinical variables of DED correlated with the vision-related QoL score. Similarly, two HRQoL studies concentrating only on DED patients (less than 160 individuals), did not find a correlation with clinical signs [35, 36]. We identified sex-differences in ocular surface signs affecting QoL scores. In women, high OSDI, conjunctival staining, and short NIBUT correlated with worse QoL scores. Among men, increased blepharitis correlated with higher depression scores. These QoL-associated symptoms and signs of DED were somewhat similar to our initial findings [17] indicating that women of this study population had higher OSDI, and corneal staining scores compared with men, while men more frequently showed signs of meibomian gland disease (MGD), blepharitis and conjunctival redness. In addition to recognizing sex differences in these associations, it is evident that a sufficiently large study population is necessary to observe the impact of clinical ocular variables on QoL measures. DED is a complex disease and understanding sex differences in DED can aid in selecting treatment options and thereby improving the QoL of patients. On basis of our previous and current results, for men in this study population, the priority for better ocular surface health-related QoL might be initiating and encouraging using the tear substitutes and taking care of lid hygiene, while women could benefit more from gaining insight into the pathogenesis and management of DED symptoms.

When comorbidities of all study participants were controlled for in HLR analysis, subjective symptoms (OSDI) still significantly affected QoL (15D, SF-36 PCS, and BDI-II) scores. Sex-specific analyses indicated that in women, OSDI was associated with 15D and SF-36 PCS, while among men, subjective eye dryness was associated with SF-36 MCS. Increased conjunctival staining score was associated with worse BDI-II among women, whereas MGD was associated with lower SF-36 MCS among men. By including comorbidities in regression analysis, we observed a significantly reduced association between DED and QoL, which aligns with the understanding that DED can be linked to various systemic diseases and medications used for their treatment [37‐41]. The statistical approaches used in the analysis are sensitive to changes in the sample size, which can also play a role in the reduced associations observed. Furthermore, it is well-established that clinical findings and subjective symptoms in DED show poor correlation [42]. We found a significant association of DED symptoms with QoL, but inconsistent results with clinical signs. This suggests a similar paradigm than observed with DED symptoms and signs: clinical signs poorly predict the subjective wellbeing of people with DED.

In a large population-based study (n = 78 165) performed with self-reported data in the Netherlands, the OR of worse QoL in people with DED was 1.54 according to SF-36 PCS and 1.39 according to SF-36 MCS when comorbidities were not included in the statistical analysis [5]. The odds decreased to 1.22 after adjusting for comorbidities. Regardless of diagnostic criteria used, in our study population, DED diagnosis was associated with approximately two-fold higher odds of worse QoL after adjusting for comorbidities when compared with people without DED. This indicates that DED itself has a serious impact on the QoL of older individuals, whether detected by subjective symptoms or objective clinical signs. The results also imply that both symptoms and clinical signs of DED impact QoL. The Beaver Dam Offspring Study in the USA [3] and the DED prevalence survey in Germany [7] found lower physical and mental SF-36 scores in people with DED than in controls. Contrary to these population-based studies, in our study DED neither influenced the SF-36 PCS nor increased the odds of worse physical health (SF-36 PCS). The study populations in the Netherlands, USA, and Germany comprised on average of younger participants and the number of participants was higher than in our study, but DED was only diagnosed through a self-administered questionnaire. These factors may have contributed to the differences in QoL outcomes among different study populations.

Several systematic reviews [12, 13, 43] have revealed a significant correlation between DED symptoms, depression, and anxiety. An association was found to be bidirectional, suggesting that depression can both contribute to and result from DED [4]. Wan et al. [43] indicated that the prevalence of depression and anxiety could be even three times higher in people with DED than those without. Furthermore, when compared with other common ocular disorders, the dry eye feeling had the biggest effect on mental health. In a study in Beijing, Jonas et al. [44] examined all common ocular disorders and only dry eye feeling was linked to a higher depression score. Similarly, in a study conducted in the Netherlands, individuals with a highly symptomatic DED demonstrated the highest risk of a low SF-36 MCS score compared to other eye disorders [5]. At the population level, DED symptoms have been associated with depression or depressive symptoms also in the Beaver Dam Offspring Study and Korean DED prevalence study [3, 4]. Our results also indicated that a higher occurrence of subjective symptoms of DED (dryness feeling and/or OSDI) was associated with a higher depression score and a lower SF-36 MCS. This was the first study associating DED clinical signs with mental health at the population level. The conjunctival staining in women and blepharitis in men were associated with the depression score in our study population, implying that of ocular surface signs, ocular surface tissue defects rather than aqueous deficiencies tend to influence depression score more.

Non-response bias is a potential concern in all questionnaire-based studies, as individuals who did not complete all QoL questionnaire items may differ from those who did. However, missing QoL data did not appear to significantly impact our results, as sensitivity analysis of missing data supported our original findings. Additionally, our 15D results were in line with the Finnish Health 2011 population study, where the mean 15D score was 0.9 [45] and similar ceiling effects were observed in the same age group (Petri Purola, personal communication, Aug 19, 2024). The most notable difference between responders and nonresponders was the older age of the nonresponders. Given that DED is more prevalent in older individuals, the impact of DED on QoL may be underestimated in our SF-36 and BDI-II-related results due to this missing data. However, statistically significant differences in QoL between individuals with and without DED should not be confused with the change that a person perceives as beneficial or harmful. The generic minimum important change for the 15D score is ± 0.015, suggesting that the differences in the 15D score observed between DED and non-DED participants in our study can be clinically important [46]. While a difference of five or more points in SF-36 subscale scores and two to three points in BDI-II scores have been considered clinically relevant in other conditions or treatments [47, 48], our comparisons based on DED diagnosis did reveal possible clinically meaningful changes in BDI-II scores but not in SF-36 scores. Clinically important differences can also be context-specific [49].

This study was not without limitations. In future, we recommend that QoL research should use a broad age range to understand DED across different ages. Our study population was from rural community; ideally both urban and rural areas should be included to ensure broader applicability of the findings. A longitudinal design would be valuable for capturing the progression and impact of DED on QoL over time. The selection of confounding factors could be broadened with e.g. environmental and life-style factors. On basis of this research, we recommend a careful selection of the diagnostic criteria for DED to enhance consistency and comparability across studies. For DED-related HRQoL, to minimize the complexity we recommend using the 15D instrument. Additionally, a large sample size is essential to improve reliability and facilitate subgroup analysis, as DED signs, symptoms, and subjective QoL experiences of DED differ between women and men. We believe that our current results offer valuable insights for the future research on the underlying mechanisms of sex-differences in DED.

Our analyses demonstrated a significant association between DED and a low QoL scores. Our findings indicated that DED notably increased the odds of lower mental health-related QoL (SF-36 MCS and BDI-II) and overall QoL (15D). Multiple analyses revealed that the subjective DED symptoms were statistically significantly associated with reduced QoL, particularly in women. For the first time at the population level, we showed that worse ocular surface signs correlated with worse QoL; NIBUT, corneal and conjunctival staining in all participants, NIBUT and conjunctival staining in women, and blepharitis in men. Our results also highlight the difference in how DED symptoms may affect the QoL in women and men. Therefore, a careful selection of diagnostic criteria, consideration of subjective symptoms, and an individualized approach to treating women and men with DED should be employed in clinical settings where DED is treated.