Original Article

 

Frequency and Risk Factors of Dry Eye Disease in Pakistani Population, A Hospital Based Study

 

Abdullah Ayub, Faryal Muhammad Akhtar, Najeeha Saleem, Muhammad Hassaan Ali, Muhammad Hammad Ayub, Nadeem Hafeez Butt

 

Pak J Ophthalmol 2017, Vol. 33, No. 4

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . .

See end of article for

authors affiliations

 

…..………………………..

 

Correspondence to:

Muhammad Hassaan Ali

Department of Ophthalmology,

Jinnah Hospital, Lahore

Email: mhassaanali@hotmail.com

 

 

 

 

 

 

 

 

 

 

 

 

Purpose: To determine prevalence and risk factors of dry eye in hospital based Pakistani population.

Study Design: Cross-sectional study.

Place and Duration of Study: Department of Ophthalmology, Jinnah Hospital, Lahore. From April 2016 to June 2016.

Materials and Methods: Three hundred cases above 18 years of age who presented to out-patient department with various ophthalmic complaints were enrolled in the study. Patients with acute inflammatory conditions, acute infections, gross corneal or conjunctival diseases, contact lens wearers and those who had undergone any form ocular surgery within the last 6 months were excluded from the study. After taking detailed history, a pre-designed questionnaire was administered to the patients that inquired about various symptoms of dry eyes followed by a detailed ophthalmic assessment and measurement of tear film breakup time (TBUT) using fluorescein dye. Patients exhibiting TBUT < 10 seconds were labeled as dry eye sufferers.

Results: There were 300 participants enrolled in the study with mean age 46.8 ± 8.3 years. 54.3% were female patients in the study. The prevalence of dry eye was found to be 18.7%. Patients aged more than 70 years showed significantly higher prevalence of dry eyes (p = 0.006). There were 18.9% hypermetropes, 16.2% myopes and 15.2% emmetropes suffering from dry eyes. Multivariate regression analysis showed that outdoor workers, people working in air conditioners, housewives, diabetics, smokers, people exposed to excessive sunlight, wind, temperature, and patients suffering from meibomian gland dysfunction were at higher risk of developing dry eye.

Conclusion: Dry eye is associated with increasing age, female gender, outdoor occupations, smoking, diabetes, meibomian gland dysfunction and refractive errors.

Key Words: Dry eye, Frequency, Diabetes, Risk Factors.

 


Dry eye is caused by abnormalities in the tear film which consists of lipid, aqueous and mucous layers from anterior to posteriorly1. The definition of dry eye disease has been under continuous revisions in the recent past. The current diagnosis and definition of dry eye disease is based on 3 criteria as given in International Dry Eye Workshop (DEWS) guidelines 2007:  1) decreased tear production or increased tear evaporation, 2) damage to the ocular surface, and 3) associated ocular discomfort or visual disturbance2. Various researchers have labeled dry eye as the most prevalent condition seen in ophthalmology clinics3. Due to tear film instability, dry eye patients report varying degree of stinging, burning, irritation, foreign body sensation, watering, fatigue, redness and photophobia etc. in the affected eye. The symptomatology of dry eye may even be seen in the absence of full triad of features mentioned above that define dry eye disease.

Various studies have reported prevalence of dry eyes from 10% to 70% depending on the type of patients/subjects, diagnostic criteria and objective tear film tests used in the study3. Many factors have been reported in literature that are associated with dry eyes that include hot weather, windy air conditions, excessive sun exposure, pollution, smoking, advancing age in females, menopause etc4. Recently, a large prevalence of dry eye disease has been reported in computer users due to decreased blink rate during performance of visual tasks with deep concentration5,6.

We conducted a literature on Google Scholar, PubMed, EMBASE and Cochrane library to find out earlier researches on the subject from Pakistan. Only one study was found in which Jehangir et al had shown presence of dry eye disease with various other ocular pathologies7. We believe that climatic and environmental conditions of Pakistan mandate similar studies. We conducted this study with the objectives to determine prevalence and risk factors of dry eye in hospital based Pakistani population.

 

MATERIALS AND METHODS

The study was conducted in Department of Ophthalmology of Jinnah Hospital, Lahore, a tertiary care hospital, from April 2016 to June 2016, after taking approval from Ethical Review Board/ Institutional Review Committee of Allama Iqbal Medical College/Jinnah Hospital, Lahore. The study was conducted following the principles of good clinical practice as laid down in Declaration of Helsinki. Informed written consent was taken from all the study participants.

We evaluated 300 cases above 18 years of age who presented to our out-patient department with various ophthalmic complaints. The patients were selected following consecutive purposive sampling.  Only those patients were selected who consented to take part in the study. Patients with acute inflammatory conditions, acute infections, gross corneal or conjunctival diseases, contact lens wearers and those who had undergone any form of ocular surgery within the last 6 months were excluded from the study.

Three researchers (AB, FA, NS) recorded detailed ophthalmic and general history of the patients focusing on main causations associated with dry eyes. Sun exposure raised outdoor temperatures, smoking, exposure to air pollutants and drugs were especially asked about. After detailed history, another researcher (MHnA) administered a pre-designed questionnaire about symptoms of dry eye. The questionnaire was administered in language of the patients and asked for socio-demographic data including age, gender, occupation, residence; and symptoms of dry eyes including recurrent watering, foreign body sensation (grittiness), itching, burning, stinging, dryness, soreness, heaviness etc.

Subsequently, a consultant ophthalmologist (MHdA) performed detailed ophthalmic examination of all the patients. During ocular examination, special attention was paid to ocular surface abnormalities, diseases of the eyelids, meibomian gland dysfunction, presence of any strands or filaments etc. Lastly, status of the precorneal tear film was assessed using tear film break-up time (TBUT). The test was performed at room temperature, keeping fans turned off after application of fluorescein dye in the inferior fornix. The patient was asked to blink 4 – 5 times to allow even distribution of fluorescein dye over the ocular surface. Finally, the patient was asked not to blink any further and was examined on slit lamp biomicroscope with cobalt blue filter. Time interval between last blink and appearance of first area of discontinuation in the precorneal tear film was TBUT. Patients exhibiting this interval to be less than ten seconds were labelled to be suffering from dry eyes. If any patient reported usage of ocular lubricants, his TBUT was measured after discontinuation of that medicine for at least 24 hours.

SPSS version 20.0 (SPSS Inc., Chicago, Illinois, USA) was used to do data analyses using 95% confidence intervals (CI). The likelihood ratio was used to calculate the p-values using Pearson Chi-square tests with a p-value of less than 0.05 considered as statistically significant. Associations of environmental risk factors with dry eye were assessed using regression analysis. The strength of association of environmental factors with dry eyes was found out by calculating odd ratio.The difference in the mean number of dry eye symptoms between dry eyed and normal individuals was assessed using analysis of variance test (ANOVA). A p-value < 0.05 was considered statistically significant.

 

RESULTS

Three hundred patients participated in the study with mean age 46.8 ± 8.3 years. There were 163 (54.3%) females and 137 (45.3%) males in the study. The detailed socio-demographic characteristics of the patients are shown in Table 1. The most common presenting complaint of the patients with dry eyes was blurring of vision in 29 (51.8%) of the cases followed by recurrent watering and itching in 39.3% and 19.6% of the cases respectively (Table 2). The frequency of dry eye was found to be 18.7% (56 cases) in our study population. Patients aged more than 70 years showed significantly higher prevalence as compared with rest of the groups in 40.0% of the cases (p = 0.006; 95% CI 1.346 – 5.780) (Table 3). A relative peak of dry eye prevalence was noted in age group 31-40 years (20.3%). As compared to males (16.1%), females showed significantly higher prevalence of dry eyes (26.4%) (p = 0.021; 95% CI 1.080 – 2.631). Menopausal women showed higher prevalence than non-menopausal women 30.1% versus 21.5% respectively. Though the prevalence of dry eye came out to be more in rural population (20.6%) as compared with urban population (17.6%) (p = 0.611; 95% CI 0.552 – 1.392), this result was not statistically significant. A detailed analysis of prevalence of dry eye according to age, gender and place of residence is presented in Table 3.

The most commonly affected people with dry eye were farmers and laborers 13 (27.1%) followed by high exposure individuals (4, 25.0%) that included professional computer users, professional drivers, field salesmen, field workers, outdoor painters, mechanics and cooks (Table 4). Odd’s ratios to show strength of associations of various occupations, environmental factors, systemic illnesses and drugs with dry eye as measured through multivariate regression analysis are shown in Table 5. Briefly, outdoor workers, people working in air conditioners, housewives, diabetics, smokers, people exposed to excessive sunlight, wind, temperature, and patients suffering from meibomian gland dysfunction were at higher risk of developing dry eye.

As part of detailed ophthalmic examination, all patients underwent assessment of their refractive status as well. 15.2% of emmetropes suffered from dry eyes while 16.2% (16/105) myopes and 18.9% (22/116) hypermetropes were affected by this condition. The mean number of ocular symptoms in dry eye patients was significantly higher as compared to non-dry eye group: 6.8 ± 2.1 versus 3.4 ± 2.3 (p = 0.001; 95% CI 1.69- 2.86).


 

Table 1: Socio-demographic characteristics of participants in the Study (n = 300).

 

 

Number of Subjects

Percentage (%)

Age Groups (years)

21 – 30

70

23.4

31 – 40

59

19.7

41 – 50

62

20.6

51 – 60

55

18.3

61 – 70

34

11.3

Above 70

20

6.7

Gender Distribution

Male

137

45.7

Female

163

54.3

Residence

Urban

193

64.3

Rural

107

35.7

Occupation

Farmers / Labourers

48

16.0

Others with High Exposurea

16

5.3

Indoor Office Workers/Shopkeepers

53

17.7

Others with Low Exposureb

55

18.3

Housewives/Students

123

41.0

Factory Workers

5

1.7

a Professional computer users, professional drivers, field salesmen, field workers, outdoor painters, mechanics, cooks etc.

b Doctors, teachers, scientists, priests etc.

 

Table 2: Presenting symptoms of dry eye patients (n = 56).

 

Symptoms

Number of Participants

Percentage (%)

Blurred Vision

29

51.8

Watering

22

39.3

Itching

11

19.6

Heaviness

   8

14.3

Burning

  7

12.5

Stickiness

   6

10.7

Dryness

   5

  8.9

Grittiness

   4

  7.1

Excessive Mucoid discharge

   3

  1.7

 

Table 3: Prevalence of dry eyes according to age, gender and residence.

 

 

Number of Subjects

Dry Eye Subjects

Prevalence (%)

p-value

95% CI

Age Groups (years)

21 – 30

  70

10

14.3

0.110

0.381 – 1.118

31 – 40

  59

12

20.3

0.502

0.671 – 1.898

41 – 50

  62

10

16.1

0.182

0.471 – 1.394

51 – 60

  55

  9

16.4

0.990

0.485 – 1.620

61 – 70

  34

  7

20.5

0.054

0.551 – 2.121

Above 70

  20

   8

40.0

0.006

1.346 – 5.780

Gender Distribution

Male

137

22

16.1

0.021

1.080 – 2.631

Female

163

43

26.4

Residence

Urban

193

34

17.6

0.611

0.552 – 1.392

Rural

107

22

20.6

Table 4: Prevalence of dry eye in various occupational groups.

 

Occupation

Number of Subjects

Dry eye subjects

Prevalence (%)

p-value

95% CI

Farmers/ labourers

48

13

27.1

0.059

0.980 – 2.863

Others with high exposure

16

4

25.0

0.781

0.459 – 2.897

Housewives / students

123

25

20.3

0.341

0.783 – 1.979

Factory workers

    5

  1

20.0

0.714

0.088 – 6.183

Indoor office workers/ shopkeepers

  53

  7

13.2

0.132

0.310 – 1.192

Others with low exposure

  55

  7

12.7

0.079

0.287 – 1.098

 

Table 5:   Multivariate logistic regression analysis showing strength of associations of various variables with dry eyes.

 

Variable

P-Value

Odd’s Ratio

95% CI

Occupation

a)       Farmers / Labourers/ High Exposure Group

0.059

2.091

0.980 – 2.863

b)       Indoor office workers/ shopkeepers

0.132

1.868

0.310 – 1.192

c)       Other low exposure group

0.079

1.010

0.287 – 1.098

d)      Housewives/ Students

0.015

1.942

1.139 – 3.229

Meibomian Gland Dysfunction

0.000

21.173

4.897 – 84.854

Diabetes

0.001

2.315

1.371 – 3.810

Smoking

0.139

1.44

0.452 – 1.249

Excessive Wind

0.005

2.12

0.263 – 0.791

Sunlight/High Temperature

0.012

1.94

0.309 – 0.872

Air pollution

0.291

1.34

0.389 – 1.340

 


 

DISCUSSION


The main objectives of this study to determine the prevalence and risk factors of dry eyes were successfully met. Various studies have shown prevalence of dry eyes to vary from around 10% to 70% depending on the diagnostic criteria used and cut-off values for various tear film assessment tests8. Some studies performed tests only on those patients who had significant number of positive complaints about dry eyes resulting in higher prevalence of dry eyes9. Besides, some studies were carried out solely on patients with rheumatoid arthritis or Sjogren’s syndrome who reported a high prevalence of dry eyes10-13. Our prevalence (18.7%) results fall within the aforementioned range and are in accordance with results from various other studies.

Our results showed that the prevalence of dry eye increased as the age of the patients increased showing a direct relationship between age and dry eyes. This result was also consistent with many other studies14-17. A relative peak of dry eye prevalence was found in the age group 31-40 years as had earlier been observed by Hikichi et al18. We believe this age group suffers most from the occupational hazards due to maximum exposure to outdoor environmental risk factors leading to ocular dryness. Countries like Pakistan with plenty of sun exposure can exhibit this phenomenon to a greater extent. However, we recommend more studies in this regard to find exact cause of this spike of prevalence of dry eyes in age group 31 – 40 years.

Like many other studies19, our study also showed significantly higher prevalence of dry eyes in females as compared to males (26.4% versus 16.1%; p = 0.021). The higher prevalence of dry eyes in females implies that females seek medical attention earlier than males for their dry eye symptoms. Besides, menopause causes decreased estrogen level in females that lead to decreased tear film production in them20. We found out higher prevalence of dry eyes in post-menopausal women as compared to pre-menopausal women (30.1% versus 21.5%) which might be explained by the aforementioned normal physiological mechanism. Though our patients showed a higher prevalence of dry eyes in post-menopausal women, earlier studies have shown same prevalence of dry eyes in both pre-and post-menopausal women14. People residing in rural areas showed higher prevalence of dry eyes as compared to urban residents. But contrary to various other studies, the difference was not statistically significant. The higher prevalence in rural residents is thought to be the result of excessive exposure to sunlight and high outdoor temperatures21.

People belonging to various occupations like farmers/labourers, housewives; people with meibomian gland dysfunction, diabetics, and people with excessive exposure to sunlight, wind, cigarette smoking and air pollution were significantly related to dry eye. Khurana et al reported high prevalence of dry eyes in farmers and labourers (32% and 28% respectively) most likely due to increased exposure to hot temperatures and sunlight22,23.  This necessitates urgent need to create awareness among farmers and labourers to take safety measures during their work. Many earlier studies have shown smoking, diabetes and pollution to be risk factors for dry eyes24. Smokers who suffer from dry eyes should be counseled about quitting or limiting smoking as a potential therapy for their dry eyes by reducing direct irritant effect of smoking on ocular surface.

Moss et al showed higher prevalence of dry eyes in patients with refractive errors as compared to emmetropes25. Our study also showed this relationship with refractive errors. These patients suffered more than emmetropes from dry eyes. Comparing emmetropes with refractive error group individuals, the prevalence of dry eyes was higher in both corrected and uncorrected refractive error group. However, subjects with their refractive errors corrected suffered less in this condition as compared to those with uncorrected refractive error (16.3% versus 25.6%). This difference in patients with corrected and uncorrected refractive error was not statistically significant (p > 0.05). It has been proposed that persons with refractive errors rub their eyes frequently leading to deposition of debris and infected particles in the fornices leading to instability of the tear film. Our study also showed that prevalence of dry eye was most in hypermetropes (18.9%) followed by myopes (16.2%) and emmetropes (15.2%). Incorporating refractive errors in dry eye researches has earlier been postulated in various studies25.

The main limitation of our study was exclusion of patients with various corneal and conjunctival pathologies which led to a relative apparent underestimation of prevalence of dry eyes in our sample. Besides, participants using contact lenses were also excluded who could have been a potential source of the condition. Fluorescein dye was used to carry out tear film break up time which itself was irritating and could cause reflex tearing. So, newer studies may use non-invasive and non-contact techniques of measuring tear breakup time. Further studies may be conducted to establish a uniform criterion for diagnosis and more etiologic associations of dry eyes. This raises the need to gather more data on the subject from our country.

 

CONCLUSION

Dry eye is quite prevalent in our cohort of patients who were suffering from ocular disease other than gross corneal and conjunctival pathologies. The study reflects a major but underdiagnosed burden of the condition in our out-patient departments. We conclude that dry eye is associated with increasing age, female gender, outdoor occupations, smoking, diabetes, meibomian gland dysfunction and refractive errors. The results should guide ophthalmic community in developing more targeted and focused approaches towards management of this issue since dry is not only an ocular disease but also a great burden on the economics of the patient. Detailed history about symptoms of the condition and good clinical examination with tear film break up time can help us properly diagnose and manage this condition actively.

 

Author’s Affiliation

Dr. Abdullah Ayub

MBBS

 House Surgeon

Department of Ophthalmology, Jinnah Hospital Lahore.

 

Dr. Faryal Muhammad Akhter

MBBS

 House Surgeon

Department of Ophthalmology, Jinnah Hospital Lahore.

 

Dr. Najeeha Saleem

MBBS

 House Surgeon

Department of Ophthalmology, Jinnah Hospital Lahore.

 

Dr. Muhammad Hassaan Ali

BSc, MBBS

Postgraduate Resident

Department of Ophthalmology, Jinnah Hospital, Lahore

 

Dr. Muhammad Hammad Ayub

MBBS, FCPS

Associate Professor

Department of Ophthalmology, Jinnah Hospital Lahore.

 

Prof. Nadeem Hafeez Butt

MBBS, FCPS, FRSCEd,

FRCS (Glas), MHPE

Department of Ophthalmology, Jinnah Hospital Lahore.

 

Role of Authors

Dr. Abdullah Ayub

Conception, Data collection, data entry, data analysis, article write up.

 

Dr. Faryal Muhammad Akhter

Data collection, data entry, data analysis, article write up.

 

Dr. Najeeha Saleem

Data collection, data entry, data analysis, article write up.

 

Dr. Muhammad Hassaan Ali

Conception, Data collection, data analysis, article write up.

 

Dr. Muhammad Hammad Ayub

Data collection, data analysis, Final critical review.

 

Prof. Nadeem Hafeez Butt

Senior author, conception, final critical review.

REFERENCES

1.        Lienert JP, Tarko L, Uchino M, Christen WG, Schaumberg DA. Long-term Natural History of Dry Eye Disease from the Patient's Perspective. Ophthalmology, 2016 Feb. 29; 123 (2): 425-33.

2.        Lemp MA, Foulks GN. The definition and classification of dry eye disease. The Ocular Surface, 2007; 5 (2): 75-92.

3.        Ali MH, Javaid M, Jamal S, Butt NH. Femtosecond laser assisted cataract surgery, beginning of a new era in cataract surgery. Oman J Ophthalmol. 2015 Sep; 8 (3): 141.

4.        Ranjan R, Shukla SK, Singh CV, Mishra BN, Sinha S, Sharma BD. Prevalence of Dry Eye and Its Association with Various Risk Factors in Rural Setup of Western Uttar Pradesh in a Tertiary Care Hospital. Open Journal of Preventive Medicine, 2016 Jan. 15; 6 (01): 57.

5.        Rosenfield M. Computer vision syndrome (aka digital eye strain). Optometry, 2016; 17 (1): 1-0.

6.        Courtin R, Pereira B, Naughton G, Chamoux A, Chiambaretta F, Lanhers C, Dutheil F. Prevalence of dry eye disease in visual display terminal workers: a systematic review and meta-analysis. BMJ open, 2016 Jan. 1; 6 (1): e009675.

7.        Jehangir S. Dry eye syndrome in Pakistani community. J Pak Med Assoc. 1990 Mar; 40: 66-7.

8.        Bakkar MM, Shihadeh WA, Haddad MF, Khader YS. Epidemiology of symptoms of dry eye disease (DED) in Jordan: A cross-sectional non-clinical population-based study. Contact Lens and Anterior Eye, 2016 Jun. 30; 39 (3): 197-202.

9.        Nowak MS, Smigielski J. The Prevalence and Risk Factors for Dry Eye Disease among Older Adults in the City of Lodz, Poland. Open Journal of Ophthalmology, 2016 Feb. 4; 6 (01): 1.

10.     Fostad IG, Eidet JR, Utheim TP, Ræder S, Lagali NS, Messelt EB, Dartt DA. Dry Eye Disease Patients with Xerostomia Report Higher Symptom Load and Have Poorer Meibum Expressibility. PloS one, 2016 May 5; 11 (5): e0155214.

11.     Beckman KA, Luchs J, Milner MS. Making the diagnosis of Sjögren’s syndrome in patients with dry eye. Clinical ophthalmology (Auckland, NZ), 2016; 10: 43.

12.     Ali MH, Ullah S, Javaid U, Javaid M, Jamal S, Butt NH. Comparison of characteristics of femtosecond laser-assisted anterior capsulotomy versus manual continuous curvilinear capsulorrhexis: A meta-analysis of 5-year results. J Pak Med Assoc. 2017; 67 (10): 1574–9.

13.     Atalay K, Ustun I, Cabuk KS, Kirgiz A, Karacan I, Taskapili M. How does rheumatoid arthritis affect corneal biomechanical properties? Int J Clin Exp Med. 2016 Jan. 1; 9 (2): 4665-9.

14.     Javaid U, Ali MH, Jamal S, Butt NH. Pathophysiology, diagnosis, and management of glaucoma associated with Sturge–Weber syndrome. Int Ophthalmol. Springer; 2017; 1–8.

15.     Jamal S, Ali MH, Ayub MH, Butt NH. Frequency and Grading of Diabetic Retinopathy in Diabetic End Stage Renal Disease Patients. Pak J Ophthalmol. 2016; 32: 65.

16.     Butt NH, Ayub MH, Ali MH. Challenges in the management of glaucoma in developing countries. Taiwan J Ophthalmol. Elsevier; 2016; 6 (3): 119–22.

17.     Bashir ZS, Ali MH, Anwar A, Ayub MH, Butt NH. Femto-LASIK: The recent innovation in laser assisted refractive surgery. J Pak Med Assoc. 2017; 67 (4): 609.

18.     Hikichi T, Yoshida A, Fukui Y, Hamano T, Ri M, Araki K, Horimoto K, Takamura E, Kitagawa K, Oyama M, Danjo Y. Prevalence of dry eye in Japanese eye centers. Graefe's archive for clinical and experimental ophthalmology, 1995 Sep. 1; 233 (9): 555-8.

19.     Schaumberg DA, Uchino M, Christen WG, Semba RD, Buring JE, Li JZ. Patient reported differences in dry eye disease between men and women: impact, management, and patient satisfaction. PloS one, 2013 Sep. 30; 8 (9): e76121.

20.     Agarwal R, Singh P, Rajpal T, Kumar R, Raghuwanshi S, Ramnani V. Hospital based study of prevalence of dry eye in post-menopausal women. Indian Journal of Clinical and Experimental Ophthalmology, 2016; 2 (1): 56-61.

21.     Ranjan R, Shukla SK, Singh CV, Mishra BN, Sinha S, Sharma BD. Prevalence of Dry Eye and Its Association with Various Risk Factors in Rural Setup of Western Uttar Pradesh in a Tertiary Care Hospital. Open Journal of Preventive Medicine, 2016 Jan. 15; 6 (01): 57.

22.     Khurana AK, Chaudhary R, Ahluwalia BK, Gupta S. Tear film profile in dry eye. Actaophthalmologica. 1991 Feb. 1; 69 (1): 79-86.

23.     Lee JH, Lee W, Yoon JH, Seok H, Roh J, Won JU. Relationship between symptoms of dry eye syndrome and occupational characteristics: the Korean National Health and Nutrition Examination Survey 2010–2012. BMC ophthalmology, 2015 Oct. 29; 15 (1): 1.

24.     Sayin N, Kara N, Pekel G, Altinkaynak H. Effects of chronic smoking on central corneal thickness, endothelial cell, and dry eye parameters. Cutaneous and ocular toxicology, 2014 Sep. 1; 33 (3): 201-5.

25.     Moss SE, Klein R, Klein BE. Long-term incidence of dry eye in an older population. Optometry & Vision Science, 2008 Aug. 1; 85 (8): 668-74.