NOTICE: We are experiencing technical issues with Academy members trying to log into the JAND site using Academy member login credentials. We are working to resolve the issue as soon as possible. Alternatively, if you are an Academy member, you can access the JAND site by registering for an Elsevier account and claiming access using the links at the top of the JAND site. Email us at [email protected] for assistance. Thanks for your patience!

A Vegetarian Diet Is Associated with a Lower Risk of Cataract, Particularly Among Individuals with Overweight: A Prospective Study

Open AccessPublished:December 11, 2020DOI:https://doi.org/10.1016/j.jand.2020.11.003

      Abstract

      Background

      Cataracts are caused by oxidative stress in the lens of the eyes and plant-based dietary patterns can contain a wide variety of protective antioxidants. However, strict vegetarians with inadequate vitamin B-12 intakes can have elevated homocysteine levels, which could increase the risk of cortical cataracts. Whether the benefits of a vegetarian diet outweigh its risks in the development of cataracts warrants investigation.

      Objective

      Our aim was to study the prospective association between a Taiwanese vegetarian dietary pattern and cataract risk.

      Design

      This was a prospective cohort study.

      Participants/setting

      The Tzu Chi Health Study recruited 6,002 participants from 2007 to 2009 at Dalin Tzu Chi Hospital. Diet was assessed through a validated food frequency questionnaire. Participants 40 years and older and without cataracts at recruitment (3,095 nonvegetarians and 1,341 vegetarians) were followed until the end of 2014, death, or occurrence of cataracts.

      Main outcome measures

      Cataract incident cases ((International Classification of Diseases, 9th Revision, Clinical Modification code 366) were identified by linkage to the National Health Insurance Research Database in Taiwan.

      Statistical analysis

      Cox proportional hazard regression with age as the underlying scale was used to estimate the association between dietary patterns and cataract risk while adjusting for potential confounders.

      Results

      Compared with nonvegetarians, vegetarians had higher intakes of soy, vegetables, nuts, whole grains, dietary fiber, vitamin C, folate, and vitamin A equivalent. In the 25,103 person-years of follow-up, 476 incident cases of cataracts were identified. A vegetarian diet was associated with a 20% reduced risk of cataracts (hazard ratio 0.80, 95% CI 0.65 to 0.99; P = 0.04) after adjusting for sex, education, smoking, alcohol drinking habits, physical activities, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, corticosteroid prescription, and body mass index (calculated as kg/m2). This association was more pronounced among individuals with overweight (defined as body mass index ≥24 in Taiwan [hazard ratio 0.70, 95% CI 0.50 to 0.99; P = 0.04]).

      Conclusions

      A vegetarian diet was associated with a lower risk of cataracts, particularly in study participants with overweight.

      Keywords

      The Continuing Professional Education (CPE) quiz for this article is availabe for free to Academy members through the MyCDRGo app (available for iOS and Android devices) and through www.jandonline.org (click on “CPE” in the menu and then “Academy Journal CPE Articles”). Log in with your Academy of Nutrition and Dietetics or Commission on Dietetic Registration username and password, click “Journal Article Quiz” on the next page, then click the “Additional Journal CPE quizzes” button to view a list of available quizzes. Non-members may take CPE quizzes by sending a request to [email protected] . There is a fee of $45 per quiz (includes quiz and copy of article) for non-member Journal CPE. CPE quizzes are valid for 3 years after the issue date in which the articles are published.
      Research Question: Is eating a vegetarian diet associated with the risk of developing cataracts?
      Key Findings: In this prospective cohort study, 1,341 vegetarians and 3,095 nonvegetarians without cataracts were followed for an average of 5.7 years; a vegetarian diet was associated with a 20% lower risk of cataracts compared with nonvegetarians, after controlling for age, socioeconomic factors, lifestyle factors, and cardiometabolic risk factors.
      Cataract, the opacification of the lens, is a major public health burden in aging societies, contributing to greater than 50% of blindness worldwide.
      • Pascolini D.
      • Mariotti S.P.
      Global estimates of visual impairment: 2010.
      A 10-year delay in cataract occurrence has been associated with a 50% reduction in cataract surgeries, effectively reducing the burden to the health care system.
      • Brian G.
      • Taylor H.
      Cataract blindness—Challenges for the 21st century.
      Cataractogenesis involves the misfolding and aggregation of lens crystalline protein, leading to loss of lens transparency.
      • Moreau K.L.
      • King J.A.
      Protein misfolding and aggregation in cataract disease and prospects for prevention.
      Oxidative stress damages lens protein and is thought to be a major cause of age-related cataracts, and factors contributing to oxidative stress, such as aging; cigarette smoking; exposure to ultraviolet lights; and diabetes, are well-known risk factors.
      • Vinson J.A.
      Oxidative stress in cataracts.
      α-Crystallin is the major chaperone system in mature lens fiber cells that binds to damaged protein and prevents the spread of protein aggregation.
      • Moreau K.L.
      • King J.A.
      Protein misfolding and aggregation in cataract disease and prospects for prevention.
      Modulation of α-crystallin chaperone activity has been proposed as a novel target in maintaining lens transparency and delaying cataract formation.
      • Kumar P.A.
      • Reddy G.B.
      Modulation of alpha-crystallin chaperone activity: A target to prevent or delay cataract?.
      ,
      • Makley L.N.
      • McMenimen K.A.
      • DeVree B.T.
      • et al.
      Pharmacological chaperone for alpha-crystallin partially restores transparency in cataract models.
      Dietary strategies, particularly the use of antioxidant vitamins for cataract prevention, have long been investigated, however, a Cochrane review of randomized controlled trials revealed no benefit of supplementing vitamin C, vitamin E, or β-carotene.
      • Mathew M.C.
      • Ervin A.M.
      • Tao J.
      • Davis R.M.
      Antioxidant vitamin supplementation for preventing and slowing the progression of age-related cataract.
      Conversely, most observational studies show a protective association between cataracts and dietary surrogates and biomarkers of fruits and vegetables.
      • Karppi J.
      • Laukkanen J.A.
      • Kurl S.
      Plasma lutein and zeaxanthin and the risk of age-related nuclear cataract among the elderly Finnish population.
      • Rautiainen S.
      • Lindblad B.E.
      • Morgenstern R.
      • Wolk A.
      Total antioxidant capacity of the diet and risk of age-related cataract: A population-based prospective cohort of women.
      • Theodoropoulou S.
      • Samoli E.
      • Theodossiadis P.G.
      • et al.
      Diet and cataract: A case-control study.
      Besides reducing oxidative stress, food components can influence cataractogenesis via modulation of α-crystallin chaperone activity.
      • Kumar P.A.
      • Reddy G.B.
      Modulation of alpha-crystallin chaperone activity: A target to prevent or delay cataract?.
      ,
      • Kumar P.A.
      • Reddy P.Y.
      • Srinivas P.N.
      • Reddy G.B.
      Delay of diabetic cataract in rats by the antiglycating potential of cumin through modulation of alpha-crystallin chaperone activity.
      It is possible that the synergism of multiple beneficial food components—as expressed in terms of dietary patterns—can exceed the effect of supplementing 1 or more nutrients. Prospective investigations on dietary patterns for cataract prevention are limited. Previous cross-sectional and retrospective studies have reported that dietary patterns in accordance with the Dietary Guidelines of Americans (as measured by the Healthy Eating Index) are associated with reduced cataract risk.
      • Ghanavati M.
      • Behrooz M.
      • Rashidkhani B.
      • Ashtray-Larky D.
      • Zameni S.D.
      • Alipour M.
      Healthy Eating Index in patients with cataract: A case-control study.
      • Mares J.A.
      • Voland R.
      • Adler R.
      • et al.
      Healthy diets and the subsequent prevalence of nuclear cataract in women.
      • Moeller S.M.
      • Taylor A.
      • Tucker K.L.
      • et al.
      Overall adherence to the Dietary Guidelines for Americans is associated with reduced prevalence of early age-related nuclear lens opacities in women.
      A trial found no difference in the incidence of cataract surgery between a Mediterranean diet vs a low-fat control diet designed according to the American Heart Association guidelines.
      • Garcia-Layana A.
      • Ciufo G.
      • Toledo E.
      • et al.
      The effect of a Mediterranean diet on the incidence of cataract surgery.
      Therefore, the optimal diet to prevent cataracts is yet to be determined.
      Plant-based foods contain higher total antioxidant capacities than animal-based foods,
      • Carlsen M.H.
      • Halvorsen B.L.
      • Holte K.
      • et al.
      The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
      and vegetarians have been shown to have higher plasma total antioxidant capacities than nonvegetarians.
      • Lim S.H.
      • Fan S.H.
      • Say Y.H.
      Plasma total antioxidant capacity (TAC) in obese Malaysian subjects.
      However, vegetarians living in countries with limited access to vitamin B-12–fortified foods tended to have low vitamin B-12 status and elevated homocysteine levels.
      • Elmadfa I.
      • Singer I.
      Vitamin B-12 and homocysteine status among vegetarians: A global perspective.
      Hyperhomocysteinemia has been associated with a higher risk of cortical cataract in the Blue Mountains Eye Study through both a cohort and a Mendelian randomization design.
      • Tan A.G.
      • Kifley A.
      • Mitchell P.
      • et al.
      Associations between methylenetetrahydrofolate reductase polymorphisms, serum homocysteine levels, and incident cortical cataract.
      Whether the benefits of vegetarian diets outweigh the risk in terms of cataract incidence deserves more investigations. So far, published findings on the association between vegetarian diets and cataracts are limited and inconsistent; vegetarian diets are associated with a lower incidence of cataracts in the EPIC-Oxford Cohort,
      • Appleby P.N.
      • Allen N.E.
      • Key T.J.
      Diet, vegetarianism, and cataract risk.
      but a higher prevalence of cataracts in an Asian Community in Leicester, UK.
      • Das B.N.
      • Thompson J.R.
      • Patel R.
      • Rosenthal A.R.
      The prevalence of age related cataract in the Asian community in Leicester: A community based study.
      It is unclear whether the effect of vegetarian diets on cataracts varies among different ethnic populations. This study aimed to investigate the association between a vegetarian diet and cataract risk in Asians through a prospective cohort study in Taiwan.

      Methods

      Study Population

      The Tzu Chi Health Study is a prospective cohort study of 6,002 individuals recruited at the Dalin Tzu Chi Hospital between the years 2007 and 2009. The majority of participants (77%) were devoted Buddhist volunteers of the Buddhists Tzu Chi Foundation (Tzu Chi volunteers), and the rest were their family members or other clientele who came to the hospital for a health examination and agreed to participate in the study. To join the study, participants needed to be at least 18 years old, willing to release their health examination data for research purposes, be interviewed for the research questionnaires, and consent to follow-up for long-term health and disease outcomes.
      Tzu Chi volunteers had at least 2 years of training before becoming “Certified Tzu Chi volunteers.” Most volunteers spent a substantial amount of time and effort on various projects of the Tzu Chi Foundation, such as community and hospital voluntary work, international relief, or environmental-related projects (eg, recycling). Volunteers are required to quit smoking and drinking alcohol (1 of the 10 precepts Tzu Chi Buddhists volunteers follow) at the time of Tzu Chi volunteer certification. Although not required, these volunteers are encouraged to consume a vegetarian diet in an effort to cultivate a sense of compassion (a teaching in Buddhism) and to lessen their environmental footprints. About one-third of the volunteers were full-time vegetarians.
      At recruitment, all participants received a health examination and were interviewed by trained research assistants on basic personal information, medical and disease conditions, family history of diseases, diet, lifestyle (smoking, alcohol drinking, and physical activities), and menstrual cycles and pregnancy history (women only). The study was conducted in accordance with the Declaration of Helsinki and the study procedures were approved by the Institutional Review Board at the Dalin Tzu Chi Hospital (Institutional Review Board numbers: B09602032 and B10403020-1). Written informed consent was obtained from all participants.
      Participants were excluded from the analysis if they provided an incorrect personal identification number or were not insured with National Health Insurance (NHI) (n = 42), as these make outcome ascertainment impossible; had a baseline age younger than 40 years (n = 509); or had baseline covariate missing (n = 13). Participants with a cataract diagnosis (n = 985) or a self-reported history of cataract surgery (n = 17) before joining the study were also excluded to enable accurate calculation of disease incidence. This left a total of 4,436 participants in the final analysis (Figure).
      Figure thumbnail gr1
      FigureFlowchart of selection of participants in the Tzu Chi Health Study for prospective analysis of the association between a vegetarian dietary pattern (vs a nonvegetarian dietary pattern) and incidence of cataracts. aTCHS = Tzu Chi Health Study. bNHIRD = National Health Insurance Research Database in Taiwan. cICD-9-CM = International Classification of Diseases, Ninth Revision, Clinical Modification.
      International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Center for Disease Control and Prevention.

      Case Ascertainment

      The baseline data were linked to the National Death Registry and the NHI database at the Health and Welfare Data Science Center,
      • Hsieh C.Y.
      • Su C.C.
      • Shao S.C.
      • et al.
      Taiwan's National Health Insurance Research Database: Past and future.
      Ministry of Health of Taiwan, by a personal identification number issued by the Taiwanese government at birth. As a government regulation to protect the privacy of study participants, all analyses were performed at the Health and Welfare Data Science Center, and only summarized results (no individual data) could be released. In order to access to these data, an application to Health and Welfare Data Science Center is required. Because Taiwan has government-sponsored universal health coverage, the NHI covers nearly 100% of the population. Participants were followed until December 31, 2014. Incident cases of cataracts were defined as having a physician-recorded diagnosis of cataracts (International Classification of Diseases, 9th Revision, Clinical Modification code 366)
      International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Center for Disease Control and Prevention.
      in at least 3 outpatient or emergency visits or 1 hospitalization.

      Assessment of Diet and Other Covariates

      At baseline, trained research assistants interviewed each participant using a structured questionnaire that included basic information, diet, and lifestyle. The dietary section consisted of a dietary habit section that included duration and type of vegetarian diet, a 64-item quantitative food frequency questionnaire reporting dietary intake during the past month, and use of supplements. The food frequency questionnaire has been validated among cohort participants, and shown to have good validity compared with dietary records (mean energy-adjusted correlation = 0.43; range, 0.38 to 0.55 for selected macronutrients and micronutrients) and biomarkers (energy-adjusted correlation of 0.41 for both vitamin B-12 and folate).
      • Chiu T.H.
      • Huang H.Y.
      • Chen K.J.
      • et al.
      Relative validity and reproducibility of a quantitative FFQ for assessing nutrient intakes of vegetarians in Taiwan.
      Vegetarians were defined as those who completely avoided meat and fish for at least 1 month before the study enrollment date. Smoking status was assessed through the lifestyle section of the questionnaire, in which participants were asked whether they currently or have ever smoked. Current and past smokers were also asked the duration of their smoking history and the average number of packs per year. Pack-year was calculated by multiplying the total-years with smoking habit by the number of packs per year. Height and weight were assessed using an electronic scale (Tanita TBF-410, calibrated annually) as the participants stood in an upright position wearing light clothes without shoes. Venous blood was drawn after 12 hours of overnight fasting and used to assess serum cholesterols, triglyceride, and glucose, using the INTEGRA 800 System (Roche Diagnostics). Comorbidities, including diabetes, hypertension, hyperlipidemia, and prescription of corticosteroid were identified from the NHI database.

      Statistical Analysis

      Baseline characteristics were compared between vegetarians and nonvegetarians using independent sample t tests and χ2 tests for continuous and categorical variables, respectively. Due to the skew nature of the dietary data, the Wilcoxon rank-sum test was used to compare the food and nutrient intakes of the vegetarians and nonvegetarians. The stratified Cox regression analysis was used to assess the association between dietary patterns and cataract risk, with age (continuous variable) as the underlying time scale, and adjustment for sex, education, smoking (pack-years), alcohol drinking, exercise habits, Tzu Chi volunteer status, prescription of corticosteroids, history of diabetes, hypertension, and hyperlipidemia, and body mass index (BMI; calculated as kg/m2), with stratification for BMI category (<24 or ≥ 24, cutoff point for normal weight and overweight set by the Ministry of Health and Welfare of Taiwan
      • Healthy B.M.I.
      Health Promotion and Administration, Ministry of Health and Welfare.
      ). BMI category was stratified within Cox model to ensure that the overall model did not violate proportional hazard assumption. Subgroup analysis was conducted by sex, Tzu Chi volunteer status, hypertension, diabetes, and hyperlipidemia.

      Sensitivity Analysis

      Sensitivity analyses were conducted by exclusion of cataract cases identified within 6 months or 1 year of study enrollment.

      Results

      Table 1 compares the baseline characteristics of vegetarians and nonvegetarians. Vegetarians had a higher proportion of females, never alcohol drinkers, and a lower percentage of lifetime cigarette smoking. They also had a lower prevalence of diabetes, hyperlipidemia, and hypertension, were less likely to engage in habitual leisure time physical activities, and had lower education levels. In terms of cardiometabolic risk factors, vegetarians had lower BMI, fasting glucose, total and low-density lipoprotein, and high-density lipoprotein cholesterol levels compared with nonvegetarians. All differences were statistically significant with the exception of triglyceride levels.
      Table 1Comparison of baseline characteristics between vegetarians and nonvegetarians of the Tzu Chi Health Study included in the prospective analysis of dietary patterns and cataract incidence (n = 4,436)
      CharacteristicsNonvegetarians (n = 3,095)Vegetarians (n = 1,341)P Value
      mean ± SD
      SD = standard deviation.
      Age, y52.8 ± 7.253.0 ± 6.90.388
      n (%)
      Sex<0.001
      Male1,470 (47.5)328 (24.5)
      Female1,625 (52.5)1,013 (75.5)
      Education level<0.001
      Elementary school or less636 (20.5)317 (23.6)
      Secondary school1,631 (52.7)748 (55.8)
      College or higher828 (26.8)276 (20.6)
      mean ± SD
      Smoking history<0.001
      Pack-year, thousand1.6 ± 4.40.5 ± 2.4
      n (%)
      Alcohol drinking<0.001
      Ever565 (18.3)112 (8.4)
      Never2,530 (81.7)1,229 (91.6)
      With habitual leisure time physical activities<0.001
      Yes2,169 (70.1)860 (64.1)
      No926 (29.9)481 (35.9)
      Tzu Chi volunteers
      Tzu Chi volunteers are devoted Buddhists who need quit smoking and alcohol drinking, and to go through a 2-year training program to be certified. These volunteers have been encouraged to consume a vegetarian diet.
      <0.001
      Yes2,210 (71.4)1,256 (93.7)
      No885 (28.6)85 (6.3)
      Hypertension
      Identified by International Classification of Diseases, 9th Revision, Clinical Modification22 in the National Health Insurance Database in Taiwan. Corticosteroid prescription was also identified in the same database.
      <0.001
      Yes680 (22.0)231 (17.2)
      No2,415 (78.0)1,110 (82.8)
      Diabetes mellitus
      Identified by International Classification of Diseases, 9th Revision, Clinical Modification22 in the National Health Insurance Database in Taiwan. Corticosteroid prescription was also identified in the same database.
      <0.001
      Yes279 (9.0)75 (5.6)
      No2,816 (91.0)1,266 (94.4)
      Hyperlipidemia
      Identified by International Classification of Diseases, 9th Revision, Clinical Modification22 in the National Health Insurance Database in Taiwan. Corticosteroid prescription was also identified in the same database.
      <0.001
      Yes620 (20.0)182 (13.6)
      No2,475 (80.0)1,159 (86.4)
      Corticosteroid prescription
      Identified by International Classification of Diseases, 9th Revision, Clinical Modification22 in the National Health Insurance Database in Taiwan. Corticosteroid prescription was also identified in the same database.
      0.365
      Yes2,499 (80.7)1,067 (79.6)
      No596 (19.3)274 (20.4)
      mean ± SD
      BMI
      BMI = body mass index calculated by height (meter) divided by the square of weight (kilograms). Height and weight were both assessed using the Tanita TBF-410 scale, calibrated annually.
      24.1 ± 3.223.0 ± 3.0<0.001
      Total cholesterol,
      To convert mg/dL cholesterol to mmol/L, multiply by 0.02586.
      Blood lipid and glucose levels were obtained from fasting serum and assessed by the INTEGRA 800 System (Roche Diagnostics).
      mg/dL
      197.7 ± 36.1180.1 ± 33.5<0.001
      HDL,
      To convert mg/dL cholesterol to mmol/L, multiply by 0.02586.
      Blood lipid and glucose levels were obtained from fasting serum and assessed by the INTEGRA 800 System (Roche Diagnostics).
      HDL = high-density lipoprotein cholesterol concentration.
      mg/dL
      54.1 ± 14.752.3 ± 13.6<0.001
      LDL,
      To convert mg/dL cholesterol to mmol/L, multiply by 0.02586.
      Blood lipid and glucose levels were obtained from fasting serum and assessed by the INTEGRA 800 System (Roche Diagnostics).
      LDL = low-density lipoprotein cholesterol concentration.
      mg/dL
      130.4 ± 33.0115.9 ± 29.5<0.001
      TG,
      Blood lipid and glucose levels were obtained from fasting serum and assessed by the INTEGRA 800 System (Roche Diagnostics).
      TG = trigycerides concentration.
      To convert mg/dL triglyceride to mmol/L, multiply by 0.01129.
      mg/dL
      117.2 ± 87.3114.1 ± 82.00.251
      Fasting glucose,
      Blood lipid and glucose levels were obtained from fasting serum and assessed by the INTEGRA 800 System (Roche Diagnostics).
      To convert mg/dL glucose to mmol/L, multiply by 0.0555.
      mg/dL
      94.6 ± 19.192.0 ± 14.3<0.001
      a SD = standard deviation.
      b Tzu Chi volunteers are devoted Buddhists who need quit smoking and alcohol drinking, and to go through a 2-year training program to be certified. These volunteers have been encouraged to consume a vegetarian diet.
      c Identified by International Classification of Diseases, 9th Revision, Clinical Modification
      International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Center for Disease Control and Prevention.
      in the National Health Insurance Database in Taiwan. Corticosteroid prescription was also identified in the same database.
      d BMI = body mass index calculated by height (meter) divided by the square of weight (kilograms). Height and weight were both assessed using the Tanita TBF-410 scale, calibrated annually.
      e To convert mg/dL cholesterol to mmol/L, multiply by 0.02586.
      f Blood lipid and glucose levels were obtained from fasting serum and assessed by the INTEGRA 800 System (Roche Diagnostics).
      g HDL = high-density lipoprotein cholesterol concentration.
      h LDL = low-density lipoprotein cholesterol concentration.
      i TG = trigycerides concentration.
      j To convert mg/dL triglyceride to mmol/L, multiply by 0.01129.
      k To convert mg/dL glucose to mmol/L, multiply by 0.0555.
      Table 2 provides the dietary and nutrient intakes of vegetarians and nonvegetarians. Compared with nonvegetarians, vegetarians consumed more soy, vegetables, nuts, whole grains, dietary fiber, vitamin C, folate, and vitamin A equivalent. No statistically significant difference was found between these 2 dietary groups for intakes of fruits and refined grains. The use of dietary supplements was similar between vegetarians and nonvegetarians (data not shown).
      Table 2Dietary intakes of vegetarians and nonvegetarians as assessed by the food frequency questionnaire
      The food frequency questionnaire used to assess the food and nutrient intakes had been tested previously for reliability and validity in a subcohort of the Tzu Chi Health Study.24
      in the Tzu Chi Health Study (n = 4,436)
      Food and nutrientsNonvegetariansVegetariansP Value
      P value is assessed using Wilcoxon rank-sum test.
      median (P25, P75
      P25, P75 = 25th percentile, 75th percentile.
      )→
      Energy, kcal1,713 (1,301, 2,256)1,704 (1,288, 2,253)<0.001
      Meat, servings
      One serving of meat, fish, and soy is defined by 7 g of protein.
      0.6 (0.2, 1.4)
      Fish, servings
      One serving of meat, fish, and soy is defined by 7 g of protein.
      0.5 (0.2, 1.1)
      Soy, servings
      One serving of meat, fish, and soy is defined by 7 g of protein.
      0.9 (0.5, 1.6)1.5 (0.8, 2.5)<0.001
      Dairy, servings
      One serving of meat, fish, and soy is defined by 7 g of protein.
      0.2 (0, 0.8)0.2 (0, 0.7)0.001
      Milk, servings
      One serving of milk is defined by 8 g of protein.
      0.2 (0, 0.7)0.2 (0, 0.6)0.001
      Eggs, servings
      One serving of eggs is defined as 1 regular size egg.
      0.3 (0.1, 0.6)0.3 (0.1, 0.5)<0.001
      Vegetables, servings
      One serving of vegetables is defined as 100 g of weight food.
      3.8 (2.3, 5.8)4.6 (2.9, 6.8)<0.001
      Fruits, servings
      One serving of fruits is defined as 15 g of carbohydrates.
      1.0 (0.5, 2.0)1.0 (0.6, 2.0)0.064
      Nuts, servings
      One serving of nuts is defined as 5 g fat.
      0.2 (0, 0.6)0.3 (0.1, 1.0)<0.001
      Whole grain, servings
      One serving of whole grains or refined grains is defined as 15 g of carbohydrates.
      1.1 (0.3, 2.8)1.7 (0.7, 4.1)<0.001
      Refined grain, servings
      One serving of whole grains or refined grains is defined as 15 g of carbohydrates.
      8.5 (5.0, 12.3)8.4 (5.0, 11.9)0.292
      Dietary fiber, g19.3 (14.2, 26.2)23.0 (16.5, 30.8)<0.001
      Vitamin C, mg165 (113, 234)170 (119, 243)0.011
      Folate, μg423 (281, 637)490 (320, 719)<0.001
      Vitamin A, μg RE
      RE = retinol equivalents.
      2,127 (1,327, 3,291)2,509 (1,555, 3,852)<0.001
      a The food frequency questionnaire used to assess the food and nutrient intakes had been tested previously for reliability and validity in a subcohort of the Tzu Chi Health Study.
      • Chiu T.H.
      • Huang H.Y.
      • Chen K.J.
      • et al.
      Relative validity and reproducibility of a quantitative FFQ for assessing nutrient intakes of vegetarians in Taiwan.
      b P value is assessed using Wilcoxon rank-sum test.
      c P25, P75 = 25th percentile, 75th percentile.
      d One serving of meat, fish, and soy is defined by 7 g of protein.
      e One serving of milk is defined by 8 g of protein.
      f One serving of eggs is defined as 1 regular size egg.
      g One serving of vegetables is defined as 100 g of weight food.
      h One serving of fruits is defined as 15 g of carbohydrates.
      i One serving of nuts is defined as 5 g fat.
      j One serving of whole grains or refined grains is defined as 15 g of carbohydrates.
      k RE = retinol equivalents.
      During the mean 5.7 years of follow-up, 476 incident cases of cataracts were identified. Table 3 shows the association between a vegetarian diet and the occurrence of cataracts. A vegetarian diet was associated with a 20% lower risk of cataract incidence (hazard ratio [HR] 0.80, 95% CI 0.65 to 0.99; P = 0.04) after adjustment for demographic factors, lifestyle factors, and comorbidities.
      Table 3Hazard ratios and 95% CIs of incidence of cataracts in vegetarians vs nonvegetarians (reference group) in the Tzu Chi Health Study, calculated from Cox Proportional Hazards Regression with age as time scale
      VariableNonvegetariansVegetariansP Value
      No. of cataract cases/person-year347/17,458129/7,645
      hazard ratio (95% CI)
      Crude10.83 (0.68 to 1.02)0.08
      Model 1
      Model 1: adjusted for sex.
      10.77 (0.63 to 0.95)0.02
      Model 2
      Model 2: Model 1 plus additional adjustment for education, alcohol drinking, smoking pack-years, physical activities, and Tzu Chi volunteer status.
      10.79 (0.64 to 0.97)0.03
      Model 3
      Model 3: Model 2 plus additional adjustment for hypertension, diabetes, hyperlipidemia, corticosteroid prescription, body mass index, stratified by body mass index (<24 or ≥24, the cut point for normal and overweight defined by Ministry of Health and Welfare of Taiwan25).
      10.80 (0.65 to 0.99)0.04
      a Model 1: adjusted for sex.
      b Model 2: Model 1 plus additional adjustment for education, alcohol drinking, smoking pack-years, physical activities, and Tzu Chi volunteer status.
      c Model 3: Model 2 plus additional adjustment for hypertension, diabetes, hyperlipidemia, corticosteroid prescription, body mass index, stratified by body mass index (<24 or ≥24, the cut point for normal and overweight defined by Ministry of Health and Welfare of Taiwan
      • Healthy B.M.I.
      Health Promotion and Administration, Ministry of Health and Welfare.
      ).
      Table 4 shows the subgroup analysis by sex, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, corticosteroid prescription, and BMI. No significant interaction was found in all of our subgroup analyses, suggesting that the association between a vegetarian diet and risk of cataracts was not significantly heterogeneous by sex, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, corticosteroid prescription, or BMI. The vegetarian-cataract association was significant among those with BMI greater than or equal to 24 (HR 0.70, 95% CI 0.50 to 0.99; P = 0.04), but not those with BMI less than 24 (HR 0.86, 95% CI 0.65 to 1.14; P = 0.29). The magnitude of protective association appeared to be larger in non–Tzu Chi volunteers (HR 0.63, 95% CI 0.30 to 1.32; P = 0.22) than Tzu Chi volunteers (HR 0.82, 95% CI 0.65 to 1.03; P = 0.08). The HRs of the diet–cataracts association ranged between 0.63 and 0.86 (all trending toward protection); however, most of these values did not reach statistical significance, likely due to small sample size.
      Table 4Subgroup analysis for risk of cataract in vegetarians vs nonvegetarians in the Tzu Chi Health Study (n = 4,436)
      Subgroups
      Subgroups were assessed at baseline.
      Cataract cases, nPerson-yearsHR
      HR = hazard ratios, calculated by Cox proportional hazard regression, adjusted: sex, education, alcohol drinking, smoking pack-years, physical activities, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, corticosteroid prescription, body mass index, stratified by body mass index (< 24 or ≥24).
      (95% CI)
      P ValueP Value for interaction
      All47625,1030.80 (0.65 to 0.99)0.04
      Sex
      Male19110,2130.70 (0.45 to 1.08)0.100.56
      Female28514,8900.81 (0.63 to 1.05)0.11
      Tzu Chi volunteer status
      Yes36419,5730.82 (0.65 to 1.03)0.080.27
      No1125,5300.63 (0.30 to 1.32)0.22
      Hypertension
      Hypertension, diabetes, and hyperlipidemia were identified from health claim database at the Health and Welfare Data Science Center, Ministry of Health of Taiwan.23
      Yes1574,8170.79 (0.53 to 1.17)0.240.99
      No31920,2860.82 (0.63 to 1.06)0.13
      Diabetes
      Hypertension, diabetes, and hyperlipidemia were identified from health claim database at the Health and Welfare Data Science Center, Ministry of Health of Taiwan.23
      Yes721,8200.68 (0.36 to 1.30)0.250.82
      No40423,2820.81 (0.64 to 1.02)0.07
      Hyperlipidemia
      Hypertension, diabetes, and hyperlipidemia were identified from health claim database at the Health and Welfare Data Science Center, Ministry of Health of Taiwan.23
      Yes1314,3050.77 (0.50 to 1.21)0.260.76
      No34520,7970.80 (0.62 to 1.02)0.07
      Corticosteroid prescription
      Corticosteroid prescription is the presence of records of corticosteroid prescription at any time except after cataract diagnosis, as identified from health claim database at the Health and Welfare Data Science Center, Ministry of Health of Taiwan.23
      Yes37320,2480.79 (0.62 to 1.01)0.060.89
      No1034,8550.77 (0.49 to 1.22)0.27
      Body mass index
      The cut point for normal and overweight (body mass index 24) is defined by Ministry of Health and Welfare of Taiwan.25
      <2425114,4010.86 (0.65 to 1.14)0.290.46
      ≥2422510,7010.70 (0.50 to 0.99)0.04
      a Subgroups were assessed at baseline.
      b HR = hazard ratios, calculated by Cox proportional hazard regression, adjusted: sex, education, alcohol drinking, smoking pack-years, physical activities, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, corticosteroid prescription, body mass index, stratified by body mass index (< 24 or ≥24).
      c Hypertension, diabetes, and hyperlipidemia were identified from health claim database at the Health and Welfare Data Science Center, Ministry of Health of Taiwan.
      • Hsieh C.Y.
      • Su C.C.
      • Shao S.C.
      • et al.
      Taiwan's National Health Insurance Research Database: Past and future.
      d Corticosteroid prescription is the presence of records of corticosteroid prescription at any time except after cataract diagnosis, as identified from health claim database at the Health and Welfare Data Science Center, Ministry of Health of Taiwan.
      • Hsieh C.Y.
      • Su C.C.
      • Shao S.C.
      • et al.
      Taiwan's National Health Insurance Research Database: Past and future.
      e The cut point for normal and overweight (body mass index 24) is defined by Ministry of Health and Welfare of Taiwan.
      • Healthy B.M.I.
      Health Promotion and Administration, Ministry of Health and Welfare.
      The sensitivity analysis (Table 5; available at www.jandonline.org) showed similar trends as our main results, but statistically less significant due to smaller case numbers.

      Discussion

      The present prospective cohort study found that a vegetarian diet is associated with approximately 20% lower risk of cataracts. There was no significant effect modification found by sex, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, BMI, or corticosteroid prescription, although the association seems to be stronger among overweight individuals.
      The results in this study are more in line with the EPIC-Oxford Cohort, which found a 26% reduced risk of cataracts in vegetarians compared with nonvegetarians,
      • Appleby P.N.
      • Allen N.E.
      • Key T.J.
      Diet, vegetarianism, and cataract risk.
      than with the positive cross-sectional association found for Asians in Leicester.
      • Das B.N.
      • Thompson J.R.
      • Patel R.
      • Rosenthal A.R.
      The prevalence of age related cataract in the Asian community in Leicester: A community based study.
      Despite lower vitamin B-12 status and its corresponding elevation of homocysteine in vegetarians in both this cohort
      • Chiu T.H.T.
      • Chang H.R.
      • Wang L.Y.
      • Chang C.C.
      • Lin M.N.
      • Lin C.L.
      Vegetarian diet and incidence of total, ischemic, and hemorrhagic stroke in 2 cohorts in Taiwan.
      and the EPIC Oxford cohort,
      • Gilsing A.M.
      • Crowe F.L.
      • Lloyd-Wright Z.
      • et al.
      Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: Results from a cross-sectional analysis of the EPIC-Oxford cohort study.
      vegetarian and vegan diets were associated with protective rather than harmful association. These findings contrast that of the Blue Mountains Eye Study, in which both serum homocysteine elevation and methylenetetrahydrofolate reductase C677T polymorphism (CT/TT vs CC—polymorphisms that influence homocysteine level) were associated with a higher risk of cortical cataracts.
      • Tan A.G.
      • Kifley A.
      • Mitchell P.
      • et al.
      Associations between methylenetetrahydrofolate reductase polymorphisms, serum homocysteine levels, and incident cortical cataract.
      The findings from this current study suggest that the potential benefits of vegetarian diets might outweigh the risk related to homocysteine. Unfortunately, in both the EPIC-Oxford and this current study, only a limited number of participants had plasma homocysteine available,
      • Chiu T.H.T.
      • Chang H.R.
      • Wang L.Y.
      • Chang C.C.
      • Lin M.N.
      • Lin C.L.
      Vegetarian diet and incidence of total, ischemic, and hemorrhagic stroke in 2 cohorts in Taiwan.
      ,
      • Gilsing A.M.
      • Crowe F.L.
      • Lloyd-Wright Z.
      • et al.
      Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: Results from a cross-sectional analysis of the EPIC-Oxford cohort study.
      precluding further analysis on how homocysteine might influence cataract risk in vegetarians.
      Vegetarian diets are defined by meat avoidance. Meat intake is positively associated with cataract risk in previous case–control studies.
      • Theodoropoulou S.
      • Samoli E.
      • Theodossiadis P.G.
      • et al.
      Diet and cataract: A case-control study.
      ,
      • Ghanavati M.
      • Behrooz M.
      • Rashidkhani B.
      • Ashtray-Larky D.
      • Zameni S.D.
      • Alipour M.
      Healthy Eating Index in patients with cataract: A case-control study.
      Cooking meat in high temperature generates heterocyclic amines, which has been associated with increased oxidative stress,
      • Carvalho A.M.
      • Miranda A.M.
      • Santos F.A.
      • Loureiro A.P.
      • Fisberg R.M.
      • Marchioni D.M.
      High intake of heterocyclic amines from meat is associated with oxidative stress.
      an important risk factor for cataract formation, particularly for nuclear cataracts.
      • Truscott R.J.
      Age-related nuclear cataract-oxidation is the key.
      Moreover, meat contains heme iron, which is more bioavailable than plant-based non-heme iron,
      • Carpenter C.E.
      • Mahoney A.W.
      Contributions of heme and nonheme iron to human nutrition.
      and a higher iron level can increase oxidative stress.
      • Galaris D.
      • Pantopoulos K.
      Oxidative stress and iron homeostasis: Mechanistic and health aspects.
      It should be noted, however, that meat consumption of nonvegetarians in this study tended to be lower than the general population,
      • Chiu T.H.
      • Huang H.Y.
      • Chiu Y.F.
      • et al.
      Taiwanese vegetarians and omnivores: Dietary composition, prevalence of diabetes and IFG.
      and this could have potentially lead to an underestimation in the findings. This trend is echoed in the findings that the magnitude of protective association between a vegetarian diet (vs nonvegetarian diet) and the risk of cataracts is larger in non–Tzu Chi volunteers (who might be more similar to the general population) than Tzu Chi volunteers (HR 0.63 vs 0.82), although both the subgroup analysis and test of interaction showed no significant difference, possibly due to small sample size.
      Vegetarians in this study also consumed more leafy greens and all vegetables, soy, nuts, and whole grains than their nonvegetarian counterparts.
      • Chiu T.H.
      • Huang H.Y.
      • Chiu Y.F.
      • et al.
      Taiwanese vegetarians and omnivores: Dietary composition, prevalence of diabetes and IFG.
      These plant foods are major sources of antioxidants.
      • Carlsen M.H.
      • Halvorsen B.L.
      • Holte K.
      • et al.
      The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
      A prospective cohort study showed that Swedish women consuming a diet with higher total antioxidant capacity had fewer cataract diagnoses and extractions.
      • Rautiainen S.
      • Lindblad B.E.
      • Morgenstern R.
      • Wolk A.
      Total antioxidant capacity of the diet and risk of age-related cataract: A population-based prospective cohort of women.
      In another study of cataract patients and age-, sex- matched controls, plant-derived nutrients, including vitamin C, folate, and inositol pentaphosphate, were associated with lower lens opacity.
      • Tarwadi K.V.
      • Chiplonkar S.A.
      • Agte V.
      Dietary and nutritional biomarkers of lens degeneration, oxidative stress and micronutrient inadequacies in Indian cataract patients.
      In a subsample (406 vegetarians and 877 nonvegetarians) of this cohort study, vegetarians were reported to have higher median plasma folate than their nonvegetarian counterparts (13.4 vs 10.6 ng/mL [to convert to nmol/L, multiply by 2.266]; P < 0.01).
      • Chiu T.H.T.
      • Chang H.R.
      • Wang L.Y.
      • Chang C.C.
      • Lin M.N.
      • Lin C.L.
      Vegetarian diet and incidence of total, ischemic, and hemorrhagic stroke in 2 cohorts in Taiwan.
      Folate insufficiency was associated with cataracts in the Elderly Nutrition and Health Survey in Taiwan conducted in 1999 to 2000.
      • Chen K.J.
      • Pan W.H.
      • Huang C.J.
      • Lin B.F.
      Association between folate status, diabetes, antihypertensive medication and age-related cataracts in elderly Taiwanese.
      However, serum folate levels can be a surrogate marker for dietary intakes of vegetables, and it is unclear whether it is folate per se or other components of vegetables that protect against cataracts.
      Diabetes and hypercholesterolemia are also important risk factors for cataracts.
      • Bang C.N.
      • Greve A.M.
      • La Cour M.
      • et al.
      Effect of randomized lipid lowering with simvastatin and ezetimibe on cataract development (from the Simvastatin and Ezetimibe in Aortic Stenosis Study).
      • Kanthan G.L.
      • Mitchell P.
      • Burlutsky G.
      • Wang J.J.
      Fasting blood glucose levels and the long-term incidence and progression of cataract—The Blue Mountains Eye Study.
      • Kuang T.M.
      • Tsai S.Y.
      • Hsu W.M.
      • Cheng C.Y.
      • Liu J.H.
      • Chou P.
      Body mass index and age-related cataract: The Shihpai Eye Study.
      Vegetarian diets have been shown to reduce risk of diabetes in prospective cohort studies.
      • Chiu T.H.T.
      • Pan W.H.
      • Lin M.N.
      • Lin C.L.
      Vegetarian diet, change in dietary patterns, and diabetes risk: A prospective study.
      ,
      • Tonstad S.
      • Butler T.
      • Yan R.
      • Fraser G.E.
      Type of vegetarian diet, body weight, and prevalence of type 2 diabetes.
      A meta-analysis of randomized controlled trials also found that vegetarian diets lower cholesterol levels.
      • Wang F.
      • Zheng J.
      • Yang B.
      • Jiang J.
      • Fu Y.
      • Li D.
      Effects of vegetarian diets on blood lipids: A systematic review and meta-analysis of randomized controlled trials.
      Vegetarians in this study had lower blood glucose, cholesterol, and BMIs. It is possible that the lower cardiometabolic risk factors in vegetarians also contribute to the protective associations with cataracts, although adjustment for these cardiometabolic risk factors did not significantly alter the results. However, in the subgroup analysis by BMI and diabetes, the protective association between a vegetarian diet and cataracts was more pronounced in those with a BMI more than or equal to 24 and those with diabetes, suggesting the inverse association between a vegetarian diet and cataract might potentially be stronger among those with higher cardiometabolic risk. Although the small sample size for subgroup analyses might be responsible for the nonsignificant results, there is no way to know this for sure.
      The strengths of this study include the prospective design and minimal loss to follow-up due to nearly 100% population coverage of the NHI, which could minimize selection bias. Moreover, the similar religious background and the general avoidance of alcohol and tobacco in both vegetarians and nonvegetarians among Tzu Chi volunteers could minimize potential confounding. There are several limitations in this study. Detailed diet was assessed only at baseline and potential change in diet over time might be a major limitation. Of the follow-up data on Tzu Chi volunteers gathered until end of 2014, most participants (89%) who were vegetarians at baseline remained vegetarians, but about 40% of nonvegetarians had some attempts to adopt a vegetarian diet (might be short-term or long-term). Using baseline diet, although imperfect, ensured that nonvegetarians would not be misclassified if they switched to a vegetarian diet for only a short-term. The majority of non–Tzu Chi-volunteer participants (91%) were nonvegetarians and follow-up diet assessments for these individuals were poorly collected, but it is assumed that they were less likely to switch to a vegetarian diet than Tzu Chi volunteers. Although many potential confounders were adjusted, some important risk factors for cataract, such as sunlight or ultraviolet light exposures and use of sunglasses, were not assessed in the study. Randomized controlled trials are needed to further confirm this finding and resolve all potential confounders. Cataract cases were defined by physician diagnosis recorded in medical claim records and not a specific eye examination, thus the subtypes or severity could not be determined. The study was conducted among Buddhists volunteers in an organization in which vegetarianism is highly encouraged and socially supported and smoking cessation is mandatory, therefore, generalizability to other populations would need to be confirmed in other studies. However, the consistency with the EPIC-Oxford finding suggests that lens biology does not differ by ethnicity. However, the small diet variation related to low meat consumption among nonvegetarians could cause underestimation of the actual diet–cataract association, and the potential effect of a vegetarian diet on cataract risk might be even stronger in the general population.

      Conclusions

      This study supports previous findings that vegetarian diets are associated with a decreased cataract risk among an Asian population in Taiwan, and also suggests that the association might be stronger among individuals with overweight. In counseling individuals with overweight during medical nutrition therapy, encouraging a plant-based dietary pattern can be beneficial for risk related to cardiovascular diseases, diabetes, hypertension, and potentially cataracts, although a randomized controlled trial will be needed to assess its actual effect.

      Supplementary Materials

      Table 5Sensitivity analysis for risk of cataract in vegetarians vs nonvegetarians in the Tzu Chi Health Study
      VariableCataract cases, nPerson-yearsHR
      HR = hazard ratios calculated by Cox proportional hazard regression, adjusted for sex, education, alcohol drinking, smoking pack-years, physical activities, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, prescription of corticosteroid, and body mass index, stratified by body mass index (<24 or ≥24).
      (95% CI)
      P Value
      Wash out for 1 year
      Cases occurred within 1 year of baseline were excluded from the analysis.
      27525,0760.86 (0.65 to 1.14)0.29
      Wash out for one-half year
      Cases occurred within one-half year of baseline were excluded from the analysis.
      30025,0940.82 (0.63 to 1.08)0.15
      a HR = hazard ratios calculated by Cox proportional hazard regression, adjusted for sex, education, alcohol drinking, smoking pack-years, physical activities, Tzu Chi volunteer status, hypertension, diabetes, hyperlipidemia, prescription of corticosteroid, and body mass index, stratified by body mass index (<24 or ≥24).
      b Cases occurred within 1 year of baseline were excluded from the analysis.
      c Cases occurred within one-half year of baseline were excluded from the analysis.

      References

        • Pascolini D.
        • Mariotti S.P.
        Global estimates of visual impairment: 2010.
        Br J Ophthalmol. 2012; 96: 614-618
        • Brian G.
        • Taylor H.
        Cataract blindness—Challenges for the 21st century.
        Bull World Health Organ. 2001; 79: 249-256
        • Moreau K.L.
        • King J.A.
        Protein misfolding and aggregation in cataract disease and prospects for prevention.
        Trends Mol Med. 2012; 18: 273-282
        • Vinson J.A.
        Oxidative stress in cataracts.
        Pathophysiology. 2006; 13: 151-162
        • Kumar P.A.
        • Reddy G.B.
        Modulation of alpha-crystallin chaperone activity: A target to prevent or delay cataract?.
        IUBMB Life. 2009; 61: 485-495
        • Makley L.N.
        • McMenimen K.A.
        • DeVree B.T.
        • et al.
        Pharmacological chaperone for alpha-crystallin partially restores transparency in cataract models.
        Science. 2015; 350: 674-677
        • Mathew M.C.
        • Ervin A.M.
        • Tao J.
        • Davis R.M.
        Antioxidant vitamin supplementation for preventing and slowing the progression of age-related cataract.
        Cochrane Database Syst Rev. 2012; : 6CD004567
        • Karppi J.
        • Laukkanen J.A.
        • Kurl S.
        Plasma lutein and zeaxanthin and the risk of age-related nuclear cataract among the elderly Finnish population.
        Br J Nutr. 2012; 108: 148-154
        • Rautiainen S.
        • Lindblad B.E.
        • Morgenstern R.
        • Wolk A.
        Total antioxidant capacity of the diet and risk of age-related cataract: A population-based prospective cohort of women.
        JAMA Ophthalmol. 2014; 132: 247-252
        • Theodoropoulou S.
        • Samoli E.
        • Theodossiadis P.G.
        • et al.
        Diet and cataract: A case-control study.
        Int Ophthalmol. 2014; 34: 59-68
        • Kumar P.A.
        • Reddy P.Y.
        • Srinivas P.N.
        • Reddy G.B.
        Delay of diabetic cataract in rats by the antiglycating potential of cumin through modulation of alpha-crystallin chaperone activity.
        J Nutr Biochem. 2009; 20: 553-562
        • Ghanavati M.
        • Behrooz M.
        • Rashidkhani B.
        • Ashtray-Larky D.
        • Zameni S.D.
        • Alipour M.
        Healthy Eating Index in patients with cataract: A case-control study.
        Iran Red Crescent Med J. 2015; 17: e22490
        • Mares J.A.
        • Voland R.
        • Adler R.
        • et al.
        Healthy diets and the subsequent prevalence of nuclear cataract in women.
        Arch Ophthalmol. 2010; 128: 738-749
        • Moeller S.M.
        • Taylor A.
        • Tucker K.L.
        • et al.
        Overall adherence to the Dietary Guidelines for Americans is associated with reduced prevalence of early age-related nuclear lens opacities in women.
        J Nutr. 2004; 134: 1812-1819
        • Garcia-Layana A.
        • Ciufo G.
        • Toledo E.
        • et al.
        The effect of a Mediterranean diet on the incidence of cataract surgery.
        Nutrients. 2017; 9: 453
        • Carlsen M.H.
        • Halvorsen B.L.
        • Holte K.
        • et al.
        The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide.
        Nutr J. 2010; 9: 3
        • Lim S.H.
        • Fan S.H.
        • Say Y.H.
        Plasma total antioxidant capacity (TAC) in obese Malaysian subjects.
        Malays J Nutr. 2012; 18: 345-354
        • Elmadfa I.
        • Singer I.
        Vitamin B-12 and homocysteine status among vegetarians: A global perspective.
        Am J Clin Nutr. 2009; 89: 1693S-1698S
        • Tan A.G.
        • Kifley A.
        • Mitchell P.
        • et al.
        Associations between methylenetetrahydrofolate reductase polymorphisms, serum homocysteine levels, and incident cortical cataract.
        JAMA Ophthalmol. 2016; 134: 522-528
        • Appleby P.N.
        • Allen N.E.
        • Key T.J.
        Diet, vegetarianism, and cataract risk.
        Am J Clin Nutr. 2011; 93: 1128-1135
        • Das B.N.
        • Thompson J.R.
        • Patel R.
        • Rosenthal A.R.
        The prevalence of age related cataract in the Asian community in Leicester: A community based study.
        Eye (Lond). 1990; 4: 723-726
      1. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Center for Disease Control and Prevention.
        (Accessed October 22, 2020)
        • Hsieh C.Y.
        • Su C.C.
        • Shao S.C.
        • et al.
        Taiwan's National Health Insurance Research Database: Past and future.
        Clin Epidemiol. 2019; : 11349-11358
        • Chiu T.H.
        • Huang H.Y.
        • Chen K.J.
        • et al.
        Relative validity and reproducibility of a quantitative FFQ for assessing nutrient intakes of vegetarians in Taiwan.
        Public Health Nutr. 2014; 17: 1459-1466
        • Healthy B.M.I.
        Health Promotion and Administration, Ministry of Health and Welfare.
        (Accessed August 22, 2020)
        • Chiu T.H.T.
        • Chang H.R.
        • Wang L.Y.
        • Chang C.C.
        • Lin M.N.
        • Lin C.L.
        Vegetarian diet and incidence of total, ischemic, and hemorrhagic stroke in 2 cohorts in Taiwan.
        Neurology. 2020; 94: e1112-e1121
        • Gilsing A.M.
        • Crowe F.L.
        • Lloyd-Wright Z.
        • et al.
        Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: Results from a cross-sectional analysis of the EPIC-Oxford cohort study.
        Eur J Clin Nutr. 2010; 64: 933-939
        • Carvalho A.M.
        • Miranda A.M.
        • Santos F.A.
        • Loureiro A.P.
        • Fisberg R.M.
        • Marchioni D.M.
        High intake of heterocyclic amines from meat is associated with oxidative stress.
        Br J Nutr. 2015; 113: 1301-1307
        • Truscott R.J.
        Age-related nuclear cataract-oxidation is the key.
        Exp Eye Res. 2005; 80: 709-725
        • Carpenter C.E.
        • Mahoney A.W.
        Contributions of heme and nonheme iron to human nutrition.
        Crit Rev Food Sci Nutr. 1992; 31: 333-367
        • Galaris D.
        • Pantopoulos K.
        Oxidative stress and iron homeostasis: Mechanistic and health aspects.
        Crit Rev Clin Lab Sci. 2008; 45: 1-23
        • Chiu T.H.
        • Huang H.Y.
        • Chiu Y.F.
        • et al.
        Taiwanese vegetarians and omnivores: Dietary composition, prevalence of diabetes and IFG.
        PLoS One. 2014; 9: e88547
        • Tarwadi K.V.
        • Chiplonkar S.A.
        • Agte V.
        Dietary and nutritional biomarkers of lens degeneration, oxidative stress and micronutrient inadequacies in Indian cataract patients.
        Clin Nutr. 2008; 27: 464-472
        • Chen K.J.
        • Pan W.H.
        • Huang C.J.
        • Lin B.F.
        Association between folate status, diabetes, antihypertensive medication and age-related cataracts in elderly Taiwanese.
        J Nutr Health Aging. 2011; 15: 304-310
        • Bang C.N.
        • Greve A.M.
        • La Cour M.
        • et al.
        Effect of randomized lipid lowering with simvastatin and ezetimibe on cataract development (from the Simvastatin and Ezetimibe in Aortic Stenosis Study).
        Am J Cardiol. 2015; 116: 1840-1844
        • Kanthan G.L.
        • Mitchell P.
        • Burlutsky G.
        • Wang J.J.
        Fasting blood glucose levels and the long-term incidence and progression of cataract—The Blue Mountains Eye Study.
        Acta Ophthalmol. 2011; 89: e434-e438
        • Kuang T.M.
        • Tsai S.Y.
        • Hsu W.M.
        • Cheng C.Y.
        • Liu J.H.
        • Chou P.
        Body mass index and age-related cataract: The Shihpai Eye Study.
        Arch Ophthalmol. 2005; 123: 1109-1114
        • Chiu T.H.T.
        • Pan W.H.
        • Lin M.N.
        • Lin C.L.
        Vegetarian diet, change in dietary patterns, and diabetes risk: A prospective study.
        Nutr Diabetes. 2018; 8: 12
        • Tonstad S.
        • Butler T.
        • Yan R.
        • Fraser G.E.
        Type of vegetarian diet, body weight, and prevalence of type 2 diabetes.
        Diabetes Care. 2009; 32: 791-796
        • Wang F.
        • Zheng J.
        • Yang B.
        • Jiang J.
        • Fu Y.
        • Li D.
        Effects of vegetarian diets on blood lipids: A systematic review and meta-analysis of randomized controlled trials.
        J Am Heart Assoc. 2015; 4: e002408

      Biography

      T. H. T. Chiu is an assistant professor, Department of Nutritional Sciences, Fu-Jen Catholic University, Taipei, Taiwan.
      C.-C. Chang is a statistician, Department of Medical Research, Buddhist Tzu Chi Medical Foundation, Hualien, Taiwan.
      C.-L. Lin is an attending physician, Department of Cardiology, Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Chiayi County, Taiwan, and an adjunct associate professor, Department of Internal Medicine, College of Medicine, Tzu Chi University, Hualien, Taiwan.
      M.-N. Lin is an attending physician, Department of Family Medicine, Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Chiayi County, Taiwan, and an adjunct assistant professor, Department of Family Medicine, College of Medicine, Tzu Chi University, Hualien, Taiwan.