Lifestyle Exposures and Eye Diseases in Adults

Abstract

Purpose

To provide a perspective of observed associations between lifestyle exposures and eye diseases in adults.

Design

Summary of recent epidemiologic studies.

Methods

Non-systematic review of the literature.

Results

Data suggest that some lifestyle exposures (e.g., smoking) are significantly associated with risk of several different diseases, suggesting that altering behaviors may decrease risk. Where risk is inherent to an occupational or recreational activity, eye protection is likely to decrease risk.

Conclusions

Although data are imperfect, it seems that some lifestyle alterations may decrease the risk of eye diseases at little or not risk to the individual.

Lifestyle exposures or behaviors have been linked to many human illnesses. Examples are wide ranging from the observation of the association of lung cancer with smoking to food-borne parasitic infection due to ingesting raw or partially cooked shellfish containing parasites (e.g., paragonimus metacercariae). Other examples are common in chronic and infectious disease epidemiology. The best way to establish the efficacy of a preventive (or causative) exposure for a disease outcome, whether a drug, device, or behavior (lifestyle), is to evaluate the effect in a well-designed randomized, masked, controlled clinical trial. However this is often not feasible when examining the relationship of lifestyle exposures with disease, where observational studies may be the only methodology available. Even though observational studies may add important information on the effectiveness of interventions in the community, investigators involved in such studies must be aware of self-selection and confounding inherent in such studies and should attempt to investigate and adjust for these (e.g., by calculating a propensity score).

Assessing relationships from observational data is complicated by varying exposure ‘doses’ over the previous life course and by interacting lifestyle exposures (omitting the ever complicating genetic factors). In addition, documentation of the exposures usually depends almost entirely on self report. This may be complicated by inaccurate memory, by deliberate over or under estimation of the exposure under investigation, and by non-random exposure. Nevertheless, important insights may be gained by examining the sometimes imperfect epidemiological data. The principles described by Hill in 1965 regarding evaluating a risk factor (e.g., smoking) to an outcome (e.g., lung cancer) in epidemiologic study (its consistency, specificity, temporality, biological gradient, plausibility), are of special importance in examining the relation of lifestyles to disease.¹ Application of results from observational studies to the population or to individual patients may be considered if the lifestyle alteration is not harmful even if the benefit is not confirmed by stringent tests of study design and statistical significance. With these considerations in mind, we present epidemiological observations on the role of some potentially modifiable lifestyle exposures and eye diseases of adults (Table 1). We confine our discussion to information largely obtained in studies in the U.S. because of constraints on the length of this review.

Table.

Lifestyle Exposures by Eye Disease.

Age-related cataract

smoking

alcohol

diet and supplements

light exposure

Age-related macular degeneration

smoking

alcohol

diet and supplements

light exposure

physical activity

Diabetic retinopathy

smoking

alcohol

diet and supplements

physical activity

Open angle glaucoma

smoking

alcohol

diet and supplements

physical activity

Trauma

sex

occupation

motor vehicle accidents

combat

sports, gardening

Dry eye

sex

smoking

alcohol

diet and supplements

climate

Refractive error

near work

Age‐Related Cataract

Smoking

The three most common cataracts considered here (nuclear, cortical, and posterior subcapsular) are not consistently related to the same risk factors.² Smoking is consistently associated with nuclear cataract. The estimated increased risk of nuclear cataract to smokers was about 9% for every ten pack-years smoked in the Beaver Dam Eye Study.³ In addition, there was about a 5% increased risk of posterior subcapsular cataract for every ten pack-years smoked. Data from the Physician’s Health Study⁴ showed that stopping smoking decreased the risk of (any type of) cataract and that, although risk did not revert to ‘normal’ with cessation of smoking, limiting exposure appeared to limit risk so that stopping at any time appeared to be relatively beneficial.

Alcohol

Alcohol consumption has also been associated with cataract, although the level of intake is usually not specified.⁵ An increased cross-sectional risk of all three forms of age-related cataract (with estimates of increasing severity varying from 34% for nuclear sclerosis, to 37% for cortical opacity, to 57% for posterior subcapsular opacity) in past heavy drinkers (defined as more than 4 drinks per day) compared to never drinkers was reported in Beaver Dam.⁶ However, they found a modest protective effect (a reduction in the odds of approximately 20%) of current moderate drinking for any cataract. These data should not be translated into a recommendation for drinking even moderate amounts of alcohol because of the possible adverse effects of alcohol. However, the data do suggest a possible benefit to the lens of avoidance of heavy drinking.

Diet and Supplements

Vitamin and mineral supplements and nutrients from food have been evaluated in many observational studies and in randomized controlled clinical trials. In a small subgroup study from the cohort participating in the Nurses’ Health Study, small but significant inverse associations between intakes of riboflavin, thiamin, duration of use of vitamin E, and change in nuclear density (an alternative measure of nuclear sclerosis) were found and it was suggested that these supplements may reduce the progression of age-related ‘lens opacification’ (although the data relates to nuclear opacification).⁷ By using combined food and supplement sources of vitamins and carotenoids for the entire Nurses’ Health Study cohort, a significant decrease in risk of cortical cataract or posterior subcapsular with higher intakes of vitamin C or carotenoids was not found; in subgroup analyses protective associations were found in non-smokers.⁸ Data from a prospective randomized placebo controlled clinical trial in 1193 healthy volunteers aged between 55 and 80 years showed no benefit of Vitamin E 500 IU in the incidence of early AMD compared to placebo daily over a 4 year period of follow-up.⁹

In cross-sectional population-based analyses of Blue Mountains Eye Study data, dietary protein, vitamin A, niacin, thiamin, and riboflavin appeared to be protective against nuclear cataract.¹⁰ In the same population, long-term use of multivitamins, B vitamins, and vitamin A supplements was associated with a decreased prevalence of nuclear and cortical cataracts. In cross-sectional analysis of data from the Beaver Dam Eye Study, use of vitamin supplements in the past was associated with decreased risk of prevalent nuclear opacities.¹¹ The findings for nuclear cataract were supported by information in that study with dietary data.

Data from clinical trials on the effect of supplements on incidence of cataract are not consistent. The Linxian cataract studies found protective effects of riboflavin and niacin against nuclear cataracts,¹² while a clinical trial including subjects from the U.S. and the U.K.¹³ found a protective effect of a mixture of oral anti-oxidant supplements for retroilluminated opacities (likely to be cortical and posterior subcapsular opacities, but not specified) in the U.S. but not in the U.K. group. Gritz et al. reported on a clinical trial of antioxidants to slow progression of nuclear opalescence over time.¹⁴ There was no significant effect of the trial combination of anti-oxidants. A randomized control trial of vitamin E prevention of cataract failed to find a protective effect on incidence or progression of nuclear, cortical, or posterior subcapsular cataract.¹⁵ The Age-Related Eye Disease Study (AREDS) was a clinical trial to test the efficacy of zinc and antioxidant nutrients on the progression of age-related macular degeneration. Cataract was evaluated in that study, although the trial was not originally designed to test the effects of the supplements on cataract. No effect of the trial preparation was found for any of the three cataract types examined.¹⁶ All study participants in AREDS who wished to stay on a multivitamin during the trial were offered Centrum™, a widely available broad spectrum multivitamin/mineral supplement throughout the study. The investigators evaluated a potential effect of Centrum on the development and progression of age-related cataracts using a propensity score approach to adjust for the bias that may have occurred with electing to use this preparation.¹⁷ Using this approach, there appeared to be a protective effect of Centrum for nuclear cataract.

Use of antioxidant supplements may not be without risk. A recent meta analysis showed antioxidant supplements to be related to increased mortality.¹⁸ Doses of antioxidant nutrients were high in many of the studies contributing data for these meta analyses. Also, some additional supplements not necessarily classified as antioxidants were included in the combinations taken. In addition, potential toxicities of vitamin E and beta-carotene, components of many antioxidant supplements currently suggested to prevent age-related ocular diseases such as cataract have been described.¹⁹

Light Exposure

Light exposure, especially in the UV-B light range, has been examined and found to be a risk factor for cataract in some studies.^20,21 Effects of such exposure seem to be significant in men. Taylor et al. found that Maryland Watermen with cortical cataract had 21% higher average annual exposure to UV-B than those without cortical cataract.²¹ In Beaver Dam, men but not women wearing glasses for refractive error at younger ages appeared to have been protected against nuclear cataract (22% reduction) and cortical opacities (33% reduction) and men who had greater estimated UVB exposure were 36% more likely to have cortical opacities.²² In that study, using a hat or sunglasses was associated with decreased risks of nuclear and cortical opacities in women and men but the associations were modest and not statistically significant. Nevertheless, if West et al.²³ are correct in their assumptions, the effect of ozone depletion and resultant increased exposure to UV-B may result in a sizable increase in the number of persons with cortical cataract. In the face of these data and predictive models, even if imperfect, the use of hats with brims and protective (against UV-B) lenses may be warranted as an approach to reduce the risk of cataract.

Summary

Posterior subcapsular cataract is infrequent, so sample sizes for this endpoint tend to be small, and so except in the case of very strong effects (as is likely true for steroids and diabetes, neither of which are ‘lifestyle’ factors), it will likely remain difficult to definitively identify associations with lifestyle exposures. Cortical and nuclear cataracts are more frequent, but it seems that the effects of nutrients and diet on these types of cataract are of small magnitude. While the benefits of multivitamin supplements may not be substantial, it is likely that there is little risk associated with their use at the recommended dose. However, available data at this time are not sufficient to support a recommendation for taking high concentrations of vitamins, minerals, and/or supplements to prevent cataract.

Age‐related Macular Degeneration

Smoking

Observations in many studies have indicated a role for smoking in the incidence and progression of AMD.²⁴ In a study in Europe, it has been estimated that based on the prevalence of smoking and the risk of progression to late AMD the population attributable risk associated with smoking for geographic atrophy is about 33% and for neovascular AMD it is about 25%.²⁵ Based on cross-sectional data from an Australian cohort, it was estimated that up to 20% of blindness among Australians aged 50 years or older was attributable to smoking.²⁶ These data suggest that a large number of cases of incident vision impairing AMD might be avoided if people did not smoke.

The effects of smoking are influenced by the presence of specific gene polymorphisms.²⁷ One is the complement factor H (CFH) gene Y402H and the other is the LOC387715 A69S gene. The former is involved in inflammation and innate immunity while the function of the latter gene is not known. In the Nurses’ Health Study, persons with 1 or 2 copies of the Y402H of CFH were approximately 2 and 4 times more likely to develop AMD, respectively, while those with the 1 and 2 copies of the LOC387715 A69S variants were approximately 2 ½ and 5 ½ times as likely to develop AMD, respectively, as persons without this variant.²⁸ Cigarette smoking multiplied these risks. Similarly, in a sample from the AREDS, smoking increased the odds of progression to late AMD in persons with either CFH or LOC387715 A69S variants.²⁹ In Beaver Dam, smokers with the wild type CFH Y402H allele had a similar risk of developing incident early AMD regardless of their smoking status (Klein R et al., unpublished data). Only those who were either heterozygous or homozygous for the CFH Y402H variant were at higher risk of developing incident early AMD if they were currently smoking compared to if they had never smoked. It is likely that there are other gene/smoking interactions that will become apparent as more candidate genes for AMD are discovered. Because smoking increases the risk of systemic disease, it is reasonable to consider advising all patients to stop smoking and also to inform those who persist of the potential hazard of developing AMD.

Alcohol

There are inconsistent epidemiological data on the association of prevalent or incident AMD with alcohol intake. In the Beaver Dam Eye Study, while adjusting for age, sex, and smoking history, there was a statistically significant 40% increase in the odds of prevalent neovascular AMD with beer drinking (per 7 drinks per week).³⁰ In the Los Angeles Latino Eye Study (LALES), a cross-sectional association of beer drinking was found with early AMD.³¹ In the Blue Mountains Eye Study, there was a cross-sectional association of alcohol consumption and early AMD (RR =60%).³² However, few other studies have shown relations between beer drinking or amount of alcohol consumed and prevalent AMD.

There are fewer data regarding alcohol consumption and incidence of AMD. In Beaver Dam, beer drinking was associated with the 5-year incidence of soft indistinct retinal drusen and increase in drusen area in men.³³ However, there were no statistically significant effects in women and no evidence of a trend in the amount drunk with increasing risk nor was there evidence of an association with late stages of AMD. In the Reykjavik Eye Study, current alcohol consumption at baseline was associated with a decrease in the 5-year risk of drusen formation and to an increase the risk of pigmentary abnormalities.³⁴ At the 10-year follow-up of the Beaver Dam cohort, a history of heavy alcohol drinking at baseline was found to be associated with an increase in the odds of developing incident exudative AMD over 10 years of follow-up.³⁵ Contrary to baseline findings, consumption of beer was not associated with the incidence of exudative macular degeneration in Beaver Dam.

In the National Health and Nutrition Examination Survey I, wine consumption was reported to be protective (a reduction in the relative risk of 34%) for AMD.³⁶ The authors of this report speculated that antioxidant phenolic compounds found in high concentrations in red wine may explain their finding. In Beaver Dam, wine consumption was associated with a reduction in the 15-year incidence of early AMD; however, the association did not reach statistical significance. Consumption of wine has not been found to be protective for AMD in other studies.

Thus, there are few consistent data supporting an association of alcohol and incident AMD and none showing that altering alcohol intake would alter the risk of AMD.

Diet and Supplements

Data concerning the relationship of diet to AMD are inconsistent and suggest the possibility of small effects at best. Green leafy vegetables and antioxidants (from diet) have been inconsistently associated with AMD.^37,38 In Beaver Dam, dietary intake of saturated fat was found to be associated with an 80% increased risk of AMD.³⁹ Caution must be exercised because it is difficult to control for other lifestyle choices throughout life (aside from diet) that might influence or account for these associations and obtaining accurate dietary data is problematic.

The possible protective role of antioxidants and zinc supplements has been an active area of research. Because the retinal photoreceptors and retinal pigment epithelium exist in a highly oxidizing environment, oxidative stress has been hypothesized to be involved in the pathogenesis of AMD. Free radical scavengers (e.g., vitamin E, antioxidant molecules [e.g., glutathione, lutein, ascorbate, vitamin E]) and metal co-factors involved in enzymatic processes (e.g., selenium and glutathione peroxidases and zinc with superoxide dismutases) have been hypothesized to play a role in prevention of AMD associated with oxidative stress. However, data from most observational studies have not found high dose antioxidant supplements and/or use of zinc to have a protective effect on AMD.

Data from the AREDS showed a protective effect against late AMD for a high dose antioxidant/zinc supplementation in those with moderate to severe early AMD. The effect was about a 25% reduction (an absolute reduction of 7%).¹⁶ Currently, it is recommended that in the presence of late stage AMD in one eye and moderate to severe AMD lesions in the other (AMD severity Levels 3 and 4 of AREDS), that the benefits outweigh the risks associated with use of AREDS formulation supplement. It has been shown that vitamin therapy for AMD improves quality of life at a reasonable cost.⁴⁰ However, a recent Cochrane review suggested that further trials of such supplements are necessary as the strongest evidence for such use is derived from one randomized clinical trial.⁴¹ In addition, there is a paucity of data from ongoing observational studies that use of supplements contained in the AREDS formulation is protective against AMD.

The use of lutein and omega 3 fatty acid supplements to prevent loss of vision from AMD has also received attention because data from epidemiological studies suggested a protective effect of use of such supplements for AMD.³⁸ Cross-sectional epidemiological data from the AREDS showed a statistically significant 30% reduction in odds of neovascular AMD and a 50% reduction in geographic atrophy when comparing the highest to lowest quintile dietary intakes of lutein/zeaxanthin. Few epidemiological studies, with the exception of the AREDS, have shown a statistically significant protective effect of dietary omega 3 fatty acids, docosahexaenoic acids, and fish intake with AMD.^38,42 Based on the review of six studies, the University of Ottawa Evidence-Based Practice Center group wrote: “…though there is some evidence for omega-3 fatty acids playing a role in primary prevention of AMD, this prevention is neither clearly supported nor refuted by the present world literature…needs further study with prospective cohort designs or randomized controlled trials targeting high-risk patients with the objective of preventing advanced dry AMD or neovascular AMD”.⁴² The Age-Related Eye Disease Study 2, a randomized parallel clinical trial will examine the efficacy of dietary supplements (Lutein 10 mg/Zeaxanthin 2 mg) and omega-3 long chain polyunsaturated fatty acids (LCPUFA) 350 mg docosahexaenoic acid (DHA) and 650 mg eicosapentaenoic acid (EPA) along with various formulations of the AREDS supplement including lower zinc concentrations and no beta-carotene. The main outcomes will be progression to AMD in eyes with moderate to severe early AMD.

Light Exposure

Exposure to intense bright sunlight or ultraviolet (UV) radiation may cause damage to the retinal pigment epithelium similar to that seen in AMD. However, to date, only weak associations of light and no association of UV exposure with AMD have been found in epidemiological studies. This may be due, in part, to the difficulty in assessing these exposures using questionnaires. Accurate information of lifetime exposure about confounding factors such as use of hats and sunglasses, cataract presence and severity, and levels of macular pigment in the eyes that may affect the relation of UV or sunlight exposure at the retina to the incidence of AMD is hard to obtain in such studies.

Physical Activity

In Beaver Dam, while controlling for relevant confounders, persons with an active lifestyle (defined as regular activity 3 times/week) at baseline were 70% less likely to develop neovascular AMD compared to persons without an active lifestyle; increased number of blocks walked per day decreased the risk of exudative AMD by 30%. Physical activity was not related to the incidence of early AMD or pure geographic atrophy. This is consistent with findings from the Eye Disease Case Control Study in which neovascular AMD was associated with less physical activity. This protective relationship may be due to the effect of physical activity on lowering systolic blood pressure, lowering white blood cell count, and decreasing body mass index, factors previously found to be associated with incident AMD. Physical activity also reduces systemic inflammation and endothelial dysfunction; both hypothesized in the pathogenesis of AMD. At this time, it is premature to recommend specific activities that might be helpful to potential patients with AMD, but more activity seems generically to be beneficial.

Summary

Aside from data showing a relationship of smoking and use of antioxidant-zinc supplements, epidemiologic data do not indicate strong or consistent effects of alcohol consumption, light exposure, taking lutein/zeaxanthin or omega 3-fatty acids, or physical activity, so there is no rationale to strongly recommend changes in these exposures at this time.

Diabetic retinopathy

Smoking

Smoking would be expected to result in a higher incidence of diabetic retinopathy because it leads to increased platelet aggregation and adhesiveness and tissue hypoxia, factors hypothesized to be involved in the pathogenesis of diabetic retinopathy. However, data from most epidemiological studies show no relationship between cigarette smoking and the incidence or progression of diabetic retinopathy. Regardless, diabetic patients should be advised not to smoke because of an increased risk of morbidity and mortality to which persons with diabetes are already prone. In the WESDR, while controlling for other risk factors, persons with type 1 diabetes who smoked were about 2 1/2 times and persons with type 2 diabetes were about 1 ½ times as likely to die as those who did not smoke.⁴³

Alcohol

Alcohol has been found to have both beneficial, deleterious, and no effects on retinopathy in different studies. This may be due to a protective effect associated with decreased platelet aggregation and adhesiveness. Of interest in the WESDR, was an association of reduction in cardiovascular disease mortality in persons with type 1 diabetes who drank on average one drink of alcohol per day.⁴⁴

Diet and Supplements

There are few data showing that after controlling for the level of glycemia and blood pressure that dietary factors or use of multivitamin supplements protects against the progression of diabetic retinopathy. In the National Health and Nutrition Examination Survey III, serum levels of alpha-tocopherol and ascorbic acid were not associated with diabetic retinopathy.⁴⁵ In the San Luis Valley Diabetes Study, a cross-sectional study in persons with type 2 diabetes mellitus, no protective effect of dietary or supplemental intake of antioxidant nutrients (e.g., carotene and vitamin C or E) for diabetic retinopathy was found.⁴⁶ In the Atherosclerosis Risk in Communities Study, vitamins C and E intake were not related to diabetic retinopathy.⁴⁷ To our knowledge, there are no data from clinical trials that have examined the efficacy of antioxidant supplement use to prevent diabetic retinopathy.

Physical Activity

Physical activity would be expected to have a beneficial effect through an attendant reduction in weight, decrease in blood pressure and inflammation. However, the few epidemiologic data available do not show a strong relationship between diabetic retinopathy and physical activity.⁴⁸ In the WESDR, women diagnosed to have diabetes before 14 years of age who participated in team sports were less likely to have proliferative diabetic retinopathy than those who did not.⁴⁹ There was no association between physical activity or leisure-time energy expenditure and the presence and severity of diabetic retinopathy in men.

Summary

Good control of glycemia, blood pressure, and possibly lipid levels and early detection and treatment of proliferative retinopathy and clinically significant macular edema remain the most important approach for preventing loss of vision from retinopathy. Lifestyle behaviors and exposures beyond these do not appear to protect against development of retinopathy.

Open Angle Glaucoma

Smoking

Smoking, either current or past, did not have a significant effect on risk of self-report for open angle glaucoma in two large cohort studies.⁵⁰ In another observation study,⁵¹ Quigley et al. also found no effect of smoking. However, a meta-analysis by Bonovas et al.⁵² suggested an increased risk of current smoking on primary open angle glaucoma. Diagnostic variability compounds the difficulty in interpreting the finding across studies. In summary, at this time, there is not firm evidence of an effect of smoking on open angle glaucoma.

Alcohol

The association of alcohol consumption on self-reported open angle glaucoma was evaluated in two large cohorts.⁵³ No association was found. In a small study from Hong Kong, a protective effect of alcohol consumption on open angle glaucoma was reported.⁵⁴ These studies, in conjunction with past publications⁵⁵ and clinical observations,⁵⁶ suggest that there is unlikely to be an important effect on alcohol on open angle glaucoma.

Diet and Supplements

Kang et al.⁵⁷ reported an increased risk of open angle glaucoma with ingestion of n-3 and n-6 fatty acids. No protective effect was found for ingestion of antioxidant vitamins in that cohort.⁵⁰ Stewart⁵⁸ found no effect of diet-reduced serum lipids and intraocular pressure. Ritch⁵⁹ has reported on complementary therapies (e.g., ginkgo biloba extract, lignans, grape seed extract) for treatment of glaucoma. While there is some laboratory work to support the possible use of these preparations, at this time, there is no firm epidemiological evidence that these supplements are effective treatments for or deterrents to the development of open angle glaucoma.

Physical Activity

Small studies suggest that physical activity has a salutary effect on intraocular pressure (reduction) in those with and without intraocular pressure.^60,61 There are no recent data from epidemiologic studies to suggest that physical activity is related to open angle glaucoma.

Summary

There are currently no consistent data to support the notion that lifestyle factors have effects on open angle glaucoma.

Trauma

Trauma may cause dysfunction or disruption of any part of the eye and periorbita. Resultant injuries range in severity from minimal scratches of the lids through destruction of the orbit and globe. About 20% of persons 43–86 years of age in one study reported ocular trauma sometime in their past.⁶² Men were nearly four times as likely to report ocular trauma as women. Persons in blue collar occupations were about 50% more likely to report such trauma than persons in white collar occupations. Trauma was caused by blunt objects in 34.4% of cases, sharp objects in 54.1% of cases, and chemicals in 11.5%. Combat exposures and trauma from motor vehicle accidents (which accounted for 905 injuries in 1999–2004 in the U.S. cannot reasonably be considered lifestyle-related. However, those exposures may be the most risky with respect to numbers of eye injuries and their severity. Aside from combat situations where one usually cannot easily avoid the potential for ocular injury, there are also other environments where projectile objects or chemical exposures cannot be easily avoided or anticipated, such as recreational (e.g., sports activities, gardening) and some occupations, but wearing high impact protective goggles can diminish the number and severity of injuries and is warranted. This would imply mandates for ocular protection in the military and protective services during training and when actively deployed, in industrial venues, and in group sports activities. Teachers, public health organizations, school health offices, physicians, and the media should be enlisted to educate the public that such protection is effective and relatively inexpensive.

Dry Eye

Reliable epidemiological data concerning the prevalence of this condition are difficult to obtain. This is because diagnostic procedures are not uniform and results of clinical testing do not always reflect the presence and severity of signs and symptoms. Therefore, dry eye is usually diagnosed based on self-reports ranging from discomfort (i.e., scratchy or burning sensation), to difficulty with vision, especially near vision. Dry eye can lead to abrasions, vascularization, and rarely, perforation of the cornea. Symptoms increase with increasing age. Prevalent symptoms of dry eye are more common in women and are associated with some systemic diseases and/or their treatments as well as lifestyle factors. Symptoms may be first noted or exacerbated by use of contact lenses.

Current smoking is associated with symptoms of dry eye. While it might be suggested that cessation of smoking may be worth trying in smokers, there are no data to show that such abstinence decreases this symptom. One might expect that coffee drinking because of its known diuretic effects would exacerbate symptoms of dry eye, but a protective effect was found. It is possible that coffee drinkers with dry eye may have curtailed their intake before the study examination. Also, taking diuretic agents did not appear to increase risk. Alcohol consumption was not significantly associated with prevalent dry eye symptoms but there was such an effect in incidence data. Miljanovic et al.⁶³ suggest that eating foods rich in n-3 and other fatty acids found in high concentrations in fish is associated with decreased risk of dry eye symptoms. It seems unlikely, except in extremes of deprivation that vitamin A deficiency accounts for dry eye in our society. Much of the available epidemiologic data has been accumulated in persons of European ancestry and risk factors may not be the same in persons of different ethnicity. Systematic information on recreational exposures to sunlight and wind and resultant dry eye symptoms are lacking. We are unaware of data on environmental exposure due to outdoor working conditions and dry eye symptoms, although corneal changes that are related to latitude have been described. From this review, we make no recommendation for changes in exposure except for those done on an individual level that seem by trial and error to provide relief.

Refractive Error

Data from several different populations show an association of more education with myopia. The relationship may reflect reading or other near-work activities. This has become a topic of some interest because of the high prevalence and increasing incidence of myopia in Taiwan and Singapore. While education and reading are highly prized activities, high myopia is considered to be at least a mild disability, so considerable research efforts have been aimed at reducing the impact of this potentially causal relationship. It has been speculated that recreational sports activities may somewhat diminish the risk of myopia in Asian school children.

Genetic studies of refractive error are ongoing. It is hoped that the effects of genes for myopic refractive error that could be limited by lifestyles will be identified.

Sexually Transmitted or Modified Disease

Sexual activity and diseases which accompany some sexual practices may be viewed as lifestyles. While a variety of sexually transmitted diseases may influence the visual system, the topic is often viewed as ocular effects of infectious diseases. This topic is beyond the scope of this limited Perspective and will not be addressed here.

Recreational Drugs

Visual function and symptoms are related to the use of many drugs. These include, but are not limited to, drugs that may be used for performance enhancement in sports and sexual function as well as ‘recreational’ use of drugs for other reasons. While this area is of interest because of apparent widespread use of these substances and the potential for lasting harm associated with some of them, the topic is vast and the data sparse. Some of the relevant data have been reviewed elsewhere.^64–66 We will not deal with these in the current Perspective.

Conclusion

Eye diseases are common and affect the health and function of the population. Their etiologies are often viewed as the result of innate metabolic changes influenced by processes of growth and aging. However, it is clear that exposures external to the eye and visual system influence changes in function of the visual system. Some of these are lifestyle factors, many or most of which are self-selected. While it difficult to test some of the hypotheses of the strength and possible causality of such exposures, extant observational data or knowledge informed by studies in other systems may give important clues as to their role in eye diseases. When inferences appear to be reasonable and interventions based on them are apparently not harmful, advice to patients may help reduce the burden imposed by some eye disease. One example is smoking, which appears to be related to AMD, age-related cataract, and possibly dry eye. With few exceptions (e.g., use of antioxidant-zinc combinations for people at risk of developing late AMD), data are either lacking or inconsistent, so there is little basis to advocate for changes in lifestyle except for smoking cessation to prevent discomfort and loss of visual function from eye disease.

Biography

Both Drs. Barbara and Ronald Klein are Professors of Ophthalmology and Visual Sciences at the University of Wisconsin School of Medicine and Public Health. Barbara is interested in ocular epidemiology of age-related eye disease, including glaucoma, cataracts and diabetic retinopathy and Ronald is interested in ocular epidemiology of age-related eye disease and hypertensive and diabetic retinopathy.