Importance of Tear Film Instability in Dry Eye Disease in Office Workers Using Visual Display Terminals: The Osaka Study


To evaluate the relationship between subjective symptoms and clinical signs in dry eye disease (DED) in office workers using visual display terminals (VDTs).


Cross-sectional study.


This study involved 672 Japanese young and middle-aged office workers who use VDTs. The subjects completed questionnaires designed to detect subjective symptoms and risk factors for DED. Dry eye tests, including tear film break-up time (TBUT), corneal-conjunctival staining with fluorescein and lissamine green, and the Schirmer test, were performed. Based on the Japanese diagnostic criteria for DED, the subjects were classified into 3 groups: definite DED, probable DED, and non-DED. Between each group, subjective symptoms and clinical signs were compared.


Of the 672 subjects, 561 (374 male, 187 female) completed the questionnaire (response rate: 83.5%). Definite DED was diagnosed in 65 subjects (11.6%), probable DED in 303 subjects (54.0%), and non-DED in 193 subjects (34.4%). The mean subjective symptom score was significantly less in subjects with probable DED (2.05 ± 0.42) and non-DED (1.63 ± 0.38) than in those with definite DED (2.19 ± 0.40) ( P < .05 and P < .01, respectively). In the subjects with probable DED, a subgroup with positive subjective symptoms and abnormal TBUT (≤5 seconds) was categorized as short TBUT-type DED, and it was found that they had a higher subjective symptom score (2.09 ± 0.40), equivalent to that of those with definite DED ( P = .269).


Despite no or minor epithelial damage, the severity of subjective symptoms was greater in short TBUT-type DED, most likely attributable to tear film instability. Thus, it might prove important to evaluate TBUT to successfully treat those patients.

The prevalence of dry eye disease (DED) has been increasing owing to changes in lifestyle driven by information technology (IT). In a previous epidemiologic study involving Japanese office workers using visual display terminals (VDTs), the prevalence of clinically diagnosed DED was 10.1% in male subjects and 21.5% in female subjects, and severe symptoms of DED were observed in 26.9% and 18.7% of the male and female subjects, respectively. In 2009, the personal computer ownership rate reached 87.2% in Japan, and the rapid spread of IT devices has promoted the daily use of VDTs for the visualization of information. Hence, the number of VDT users and the amount of time they spend using their respective terminals have both dramatically increased.

Continuous use of VDT displays results in decreased blinking and a subsequent increase in ocular surface dryness, which is known to be a causative factor in DED cases. Excess aqueous evaporation from the ocular surface tear film easily occurs during VDT use because the tear film break-up time (TBUT) is shorter than the interval between blinks. It has been reported that using a VDT for more than 4 hours a day (compared to less than 2 hours daily) increases the risk of DED (odds ratio: 1.83). Work-related prolonged VDT use is considered to increase the risk of ocular disorders including eye strain and accommodative spasm, as well as DED. Thus, collecting the diagnostic and therapeutic data about VDT use–related ocular disorders is vital.

To date, there are limited data on the prevalence and the risk factors of DED in VDT users. The relationship between VDT use–associated DED and shorter TBUT has been described previously; however, there is insufficient epidemiologic evidence for the prevalence and risk factors of DED subtypes in VDT users.

Under the supervision of the Japanese Dry Eye Society, we conducted a large-scale epidemiologic study in office workers using VDTs in Japan (The Osaka Study). The Osaka Study was a questionnaire-based assessment of DED symptoms combined with dry eye examinations carried out by dry eye specialists investigating the epidemiology, risk factors, and pathology of DED. In this present study, we evaluated the relationship between DED subjective symptoms and clinical signs based on the Japanese dry eye diagnostic criteria established in 2006. Furthermore, we investigated the characteristics of probable DED with TBUT ≤5 seconds and normal tear secretion (short TBUT-type DED) in order to study the independent association of definite DED with reduced TBUT in VDT office workers using VDTs.


This study followed the tenets of the Declaration of Helsinki, the Ethical Guidelines for Clinical Research (issued on December 28, 2004 by the Ministry of Health, Labor and Welfare [MHLW] of Japan), and the Ethical Guidelines for Epidemiological Research (issued on June 17, 2002 by the MHLW). The protocol of this institutional and cross-sectional study was approved prospectively by the Institutional Review Board of Ryogoku Eye Clinic, Tokyo, Japan.

Study Population

Under the supervision of the Japanese Dry Eye Society, the authors randomly selected 2 large companies listed on the Japanese stock market and sent a letter to the industrial physician in charge of health management in each company; the letter explained the purpose of the study and requested their participation. The only company that responded to our letter and consented to participate was enrolled in this study. The employees working in the central office of the company were informed, via a website, of the study’s general outline and ophthalmic clinical examinations. We enrolled those employees who agreed to participate in the study after providing them with background information and receiving their consents via the website to undergo the ophthalmic examinations. Exclusion criteria included any employee who reported a history of refractive surgery or who was deemed inappropriate for this study by the ophthalmologists.

To ensure the human rights and personal security of each participant, we took strict measures to protect the privacy of participants’ personal information (ie, special care in handling of their collected information; exclusion of individually identifiable information from the questionnaire; exclusion of individually identifiable information from the data to be published; and exclusive use of the data for the purposes of the study).


We administered a web-based dry eye questionnaire to evaluate subjective DED symptoms and to obtain information on each participant’s age, sex, and usage of VDT devices.

The questionnaire used in the study was developed by Toda and associates, and included 12 questions pertaining to the diagnostic symptoms of DED. The frequency of each symptom was evaluated as never (1 point), sometimes (2 points), often (3 points), and constantly (4 points), and the mean point value of all symptoms was defined as the DED symptom score. Subjects who answered “often” or “constantly” for at least 1 item were defined as being symptom positive for DED.

To validate our symptom evaluation method, it was compared with a well-established method, the Dry Eye Severity Index Score (DESIS), which is a severity grading system based on the translated version of the Ocular Surface Disease Index (OSDI) questionnaire into Japanese. Each DESIS item was evaluated as never (0), rarely (1), occasionally (2), often (3), and always (4), and the total number of points scored on 29 questions was converted to a scale ranging from 0 to 100.

Clinical Evaluation

Seven ophthalmologists from the Japanese Dry Eye Society performed the ophthalmic examinations at the health-care institute of the participating company, located in Osaka, Japan. Subjects who used contact lenses (CLs) underwent the examination more than 1 day after removing their CLs.

Tear Function Tests and Ocular Surface Evaluation

Ophthalmic examinations included the assessment of corneal and conjunctival vital staining with fluorescein and lissamine green, respectively; measurement of TBUT; and the Schirmer test. Briefly, corneal and conjunctival epithelial damage was evaluated by the double vital staining method. Two microliters each of 1% sodium fluorescein and preservative-free 1% lissamine green were instilled into the conjunctival sac using a micropipette to evaluate epithelial damage of the cornea and bulbar conjunctiva, respectively. The ocular surface was divided into 3 sections (the nasal conjunctiva, the cornea, and the temporal conjunctiva), each of which was graded 0–3 for severity, and the overall epithelial damage was scored on a scale of 0–9.

TBUT measurement and the Schirmer test were performed to assess tear film stability and tear fluid secretion, respectively. TBUT was measured using a fluorescein dye solution. After instillation of the solution, the subjects were instructed to blink 3 times to ensure adequate mixing of the fluorescein dye with the tear film. The time interval between the third blink and appearance of the first dark spot in the cornea was consecutively measured by stopwatch, with the mean of 3 measurements regarded as the TBUT in this study.

The Schirmer test was performed without topical anesthesia after all of the other examinations had been performed. Strips of filter paper (Whatman No. 41; Showa Yakuhin Kako Co Ltd, Tokyo, Japan) were placed for 5 minutes at the outer third of the temporal lower conjunctival fornix, with the subject spontaneously blinking. The strips were then removed, and the length of the filter paper wetted by the spontaneous blinks was recorded in millimeters. To avoid the influence of stimulation by ocular surface staining, the Schirmer test was performed after a 10-minute interval.

To prevent air conditioning from influencing the examinations, no air conditioning device was used in the examination room. The room temperature and humidity were maintained at 25.0–26.5 °C and 60%–65%, respectively, during the examinations. All subjects were prohibited from using a VDT for 1 hour prior to the examinations.

Diagnosis of Dry Eye Disease

DED was diagnosed based on the results of the clinical evaluation and our questionnaire, which used the dry eye diagnostic criteria defined in 2006 by the Japanese Dry Eye Society. These criteria included the presence of: (1) positive DED symptoms; (2) qualitative or quantitative abnormalities of the tear film in 1 or both eyes (TBUT of ≤5 seconds or Schirmer test results of ≤5 mm); and (3) corneo-conjunctival epithelial damage (total staining score of ≥3 out of 9 points). The presence of all 3 criteria was necessary for a diagnosis of definite DED. The presence of 2 of the 3 criteria in 1 or both eyes was necessary for a diagnosis of probable DED, while the presence of 1 or no positive criteria indicated non-DED. The eye fulfilling the most criteria was used to make the diagnosis. When both eyes met the same number of criteria, the eye with (1) the higher staining score and (2) the shorter TBUT was used for the diagnosis. If the staining score and TBUT were the same in both eyes, the right eye was used for the diagnosis. Based on the diagnostic criteria, the subjects were placed into either the definite DED, probable DED, or non-DED group, and the subjective symptom score was then compared among the groups.

Subgroup Analysis of the Short Tear Break-up Time–Type Dry Eye Disease Group

We performed a subgroup analysis of the short TBUT subtype in the probable DED group. In this study, short TBUT was defined as abnormal TBUT (≤5 seconds) with a normal Schirmer test value (>5 mm). The subjects with short TBUT in the probable DED group were assigned to the short TBUT–type DED subgroup. The subjective symptom scores of each group were then analyzed and compared.

To assess the status of tear film lipid secretion, meibomian gland function was evaluated using the meibomian gland dysfunction (MGD) grading system established by Shimazaki and associates. In brief, when meibomian gland secretion was absent after application of moderate pressure to the eyelid, meibomian gland function was defined as abnormal.

Differences in the subjective symptoms between the subjects in the short TBUT subgroup and the other subjects were evaluated. Furthermore, the differences between the short TBUT–type DED group and the definite DED group were evaluated to characterize short TBUT–type DED.

Statistical Analysis

For discrete variables, the calculated mean values were analyzed using the Student t test and the Tukey test to compare 2 groups and 3 groups, respectively. For categorical variables, the Fisher exact test was used. The P values were 2-sided and the level of significance was set at 5% ( P < .05). SAS 9.2 software for Windows (SAS Institute Inc, Cary, North Carolina, USA) was used for the data analysis.


The study period was from August 1, to September 1, 2011. The questionnaire was sent electronically to 672 eligible workers in the participating company; 561 workers (83.5%) responded and were enrolled in the study (374 male and 187 female subjects; mean age: 43.2 years). The mean (±SD) age of the male and female subjects was 45.1 (±8.9; range: 26.0–64.0) and 39.5 (±7.3; range: 24.0–60.0) years, respectively.

Based on the responses to the 12-item questionnaire, 399 subjects (71.1%) were DED positive and 162 subjects (28.9%) were DED negative. The mean DED symptom score (mean ± SD) was 1.92 ± 0.46, and the mean number of symptoms was 2.5 ± 2.5. The most frequently occurring symptoms included “eye fatigue” (48.1%), “dry sensation” (32.4%), and “blurred vision” (25.0%).

Vital staining indicated that corneo-conjunctival epithelial damage was present in 90 subjects (16.0%; score ≥3) but not in 471 subjects (84.0%; score 0–2). The corneo-conjunctival epithelial damage was generally mild in severity and the positive rate was low. TBUT was abnormal (≤5 seconds) in 441 subjects (78.6%) and normal (>5 seconds) in 120 subjects (21.4%). Of the 561 subjects who took the Schirmer test, 95 subjects (16.9%) showed abnormal values (≤5 mm) and 466 subjects (83.1%) showed normal values (>5 mm). Most subjects (66.0%, n = 370) had a TBUT of ≤5 seconds and a Schirmer test value of >5 mm (short TBUT subgroup).

The subjects were diagnostically classified into 1 of 3 categories: definite DED (n = 65; 11.6%); probable DED (n = 303; 54.0%); and non-DED (n = 193; 34.4%). Of the 303 probable DED subjects, 244 subjects (80.5%) were included in the short TBUT subgroup and 59 subjects (19.5%) in the “others” subgroup ( Table 1 ).

Table 1

Allocation of Osaka Study Subjects Into Groups Based on Diagnosis of Dry Eye Disease

Definite DED Probable DED Non-DED
N (%) 65 (11.6%) 303 (54.0%) 193 (34.4%)
Short TBUT 50 (8.9%) 244 (43.5%) 76 (13.5%)
Others 15 (2.7%) 59 (10.5%) 117 (20.9%)

DED = dry eye disease; TBUT = tear film break-up time.

The mean 12-item symptom score was significantly higher in the definite DED group (2.19 ± 0.40) than in the probable DED (2.05 ± 0.42) and non-DED (1.63 ± 0.38) groups (Tukey test, for each group P < .01) ( Table 2 ). This result suggests that the severity of dry eye symptoms is associated with the diagnostic criteria of DED. A similar relationship was observed with DESIS.

Table 2

Comparison of the 12-Item Symptom Score With the Dry Eye Severity Index in Assessing the Severity of Dry Eye Disease in Osaka Study Subjects

Subjective Symptom Diagnosis Mean ± SD (N) P Value
Symptom score by 12 items (0–4) 1. Definite DED 2.19 ± 0.40 (65) 1:2 P = .028 a
2. Probable DED 2.05 ± 0.42 (303) 1:3 P = .000 b
3. Non-DED 1.63 ± 0.38 (193) 2:3 P = .000 b
DESIS (0–100) 1. Definite DED 31.8 ± 13.6 (65) 1:2 P = .073
2. Probable DED 28.1 ± 14.0 (303) 1:3 P = .000 b
3. Non-DED 17.1 ± 9.1 (193) 2:3 P = .000 b

DED = dry eye disease; DESIS = dry eye severity index score; SD = standard deviation of the mean.

P value was calculated using the Tukey test.

a P < .05.

b P < .01.

Associations between the clinical findings and DED symptom scores are shown in Table 3 . In the subjects with abnormal (≤5 seconds) and normal TBUT (>5 seconds), the mean subjective symptom scores (mean ± SD) were 1.95 ± 0.46 and 1.82 ± 0.45, respectively, indicating a statistically significant difference (Student t test, P = .006). In the subjects with normal (<3) and abnormal (≥3) total staining scores, the mean DED symptom scores were 1.91 ± 0.45 and 2.02 ± 0.47, respectively, and were significantly different ( P = .030). There was no significant difference in DED symptom score between the subjects with normal (1.92 ± 0.46 mm) and abnormal (1.92 ± 0.46 mm) Schirmer test values ( P = .987). DESIS analysis showed a similar tendency, suggesting that subjective symptom scores were significantly associated with TBUT ( P = .009) and staining score ( P = .022) but not with the Schirmer test value ( P = .705).

Table 3

Associations Between Dry Eye Symptom Score and Ophthalmic Examination Values in Osaka Study Subjects

Subjective Symptom Score Schirmer Test (N) P Value TBUT Test (N) P Value Vital Staining Score (N) P Value
Normal (>5 mm) Abnormal (≤5 mm) Normal (>5 s) Abnormal (≤5 s) Normal (<3 Points) Abnormal (≥3 Points)
Symptom score by 12 items (0–4) 1.92 ± 0.46 (466) 1.92 ± 0.46 (95) .987 1.82 ± 0.45 (120) 1.95 ± 0.46 (441) .006 a 1.91 ± 0.45 (471) 2.02 ± 0.47 (90) .030 b
DESIS (0–100) 24.6 ± 13.5 (466) 25.2 ± 14.5 (95) .705 21.8 ± 11.9 (120) 25.5 ± 14.1 (441) .009 a 24.2 ± 13.5 (471) 27.8 ± 14.2 (90) .022 b

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 7, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Importance of Tear Film Instability in Dry Eye Disease in Office Workers Using Visual Display Terminals: The Osaka Study

Full access? Get Clinical Tree

Get Clinical Tree app for offline access