Abstract

Age-related macular degeneration (AMD) is the leading cause of blindness among older people, and diet has been postulated to alter risk of AMD. To evaluate associations between red meat and chicken intake and AMD, the authors conducted a cohort study of 6,734 persons aged 58–69 years in 1990–1994 in Melbourne, Australia. Meat intake was estimated from a food frequency questionnaire at baseline. At follow-up (2003–2006), bilateral digital macular photographs were taken and evaluated for AMD (1,680 cases of early AMD, 77 cases of late AMD). Logistic regression was used to estimate odds ratios, adjusted for age, smoking, and other potential confounders. Higher red meat intake was positively associated with early AMD; the odds ratio for consumption of red meat ≥10 times/week versus <5 times/week was 1.47 (95% confidence interval: 1.21, 1.79; P-trend < 0.001). Similar trends toward increasing prevalence of early AMD were seen with higher intakes of fresh and processed red meat. Conversely, consumption of chicken ≥3.5 times/week versus <1.5 times/week was inversely associated with late AMD (odds ratio = 0.43, 95% confidence interval: 0.20, 0.91; P-trend = 0.007). These results suggest that different meats may differently affect AMD risk and may be a target for lifestyle modification.

Age-related macular degeneration (AMD) is the leading cause of severe vision loss in people aged 50 years or older in the developed world (1, 2). Early AMD is characterized clinically by yellow deposits known as drusen and alterations in retinal pigmentation. These are thought to indicate an unhealthy retina which predisposes to late visually threatening complications, or late AMD (3). Late AMD develops when there is ingrowth of new blood vessels that bleed into the subretinal space (exudative or “wet” type) or when the macula atrophies (geographic atrophy or “dry” type).

This progressive late-onset degenerative disease affects central vision, which is essential for independent activities of daily living. The loss of central vision not only results in a significant loss of quality of life but also places a massive burden on health-care resources (4, 5). With the aging US population, it is estimated that by 2020 the number of Americans with late AMD will increase by 50% to 3 million (1). A similar doubling of AMD prevalence in Australia, with direct costs to the community reaching over A$1 billion by the year 2020, is also expected (4).

Risk factors for AMD include age, family history, and smoking, with smoking being the only modifiable risk factor consistently found (6). To date, the pathogenesis of AMD remains unknown; and with treatment available only for neovascular complications of AMD (7), the identification of modifiable risk factors has enormous public health implications. Meat consumption has been associated with AMD in several studies primarily evaluating fat intake and AMD (8–10); despite this association, thus far no study has focused specifically on meat intake or evaluated this association in detail. High meat intake has been associated with higher levels of N-nitroso compounds (11), heme iron (12), and advanced glycation end products (13), which could result in oxidative damage and could be toxic to the retina; oxidative damage has been postulated to be one of the key pathogenic processes in AMD.

Meat forms a large part of the diet in Australia and many other Western countries where AMD is common; thus, clarifying the role of meat consumption in AMD is important. We examined the relation between meat consumption and AMD among 6,734 Melbourne, Australia, residents who were aged 58–69 years upon enrollment in 1990–1994.

MATERIALS AND METHODS

Study design and participants

Participants were selected from the Melbourne Collaborative Cohort Study, a prospective cohort study of 41,528 Melbourne residents (17,049 men) aged 40–69 years at the time of recruitment between 1990 and 1994. Participants were recruited via electoral rolls (voting is compulsory in Australia) and advertisements; details have been published elsewhere (14, 15). In this study, we focused on participants who were aged 58–69 years at baseline and were Australian or British-born. A total of 13,217 participants satisfied these eligibility criteria, of whom 11,617 were alive and residing in the state of Victoria on May 1, 2003. During the follow-up period, May 2003–July 2006, a further 638 participants died and another 157 participants left Victoria (Figure 1). From these, 6,734 participants (58%) attended follow-up.

Figure 1.

Participation in the Melbourne Collaborative Cohort Study (MCCS), Melbourne, Victoria, Australia, 1990–2006.

Figure 1.

Participation in the Melbourne Collaborative Cohort Study (MCCS), Melbourne, Victoria, Australia, 1990–2006.

Participants were excluded if they reported extreme energy intakes (<1st percentile and >99th percentile), which suggested that food frequency questionnaires (FFQs) had been improperly completed (n = 110); if they reported having had acute myocardial infarction, angina, or diabetes at baseline and thus were likely to have changed their diets (n = 624); if they had missing dietary data (n = 1); or if they had missing or nongradable macula photographs due to refusal, loss of an eye, or poor photograph quality secondary to cataract or small pupils (n = 395). After these exclusions, data from 5,604 participants were available for analysis.

Baseline dietary assessment

At baseline, dietary intake over the previous 12 months was estimated using a 121-item FFQ specifically developed for the Melbourne Collaborative Cohort Study. This FFQ was based on data from weighed food records in a sample of 810 Melbourne residents of similar age and ethnicity as the study cohort (16). The FFQ included 18 items on intake of fresh red meat, processed red meat, and chicken. Fresh red meat was defined as veal or beef schnitzel, roast beef or veal, beef steak, rissoles (meatballs) or meatloaf, mixed dishes with beef, roast lamb or lamb chops, mixed dishes with lamb, roast pork or pork chops, and rabbit or other game (rarely consumed). Processed meat included salami or continental sausage, sausages or frankfurters, bacon, ham (including prosciutto), corned beef, and manufactured luncheon meats. Chicken included roast or fried chicken, boiled or steamed chicken, and mixed dishes with chicken.

The completed FFQ was optically scanned into the database. Nutrient intakes were calculated using standard sex-specific portion sizes from the weighed food records. The energy and nutrient contents in food were derived from Australian food composition tables (17). The fatty acid composition of foods was obtained from the Royal Melbourne Institute of Technology fatty acid database (18). Carotenoid data were obtained from the 1998 US Department of Agriculture database (19). For estimation of the reproducibility of the FFQ, between July 1992 and June 1993, 275 participants were invited to participate in a study that required completing a second FFQ 12 months after baseline. Of these persons, 242 (88%) participated. The weighted kappa statistics for the reproducibility of the quartiles of meat intake were 0.42 (95% confidence interval (CI): 0.30, 0.55) for fresh meat, 0.60 (95% CI: 0.48, 0.73) for processed meat, and 0.44 (95% CI: 0.32, 0.56) for chicken (20). The FFQ also collected “yes/no” responses for use of supplements (vitamin C, vitamin E, cod liver oil, or fish oil), with no details on dosage. Thus, vitamin and fat intakes were derived only from diet.

Assessment of demographic, lifestyle, and anthropometric factors

At baseline, a structured interview was used to obtain further information on demographic and lifestyle factors. The information obtained included age, sex, smoking status, and country of birth. Height, weight, and blood pressure were directly measured.

Ophthalmic data

During follow-up between May 2003 and July 2006, participants had both eyes photographed using a digital Canon CR6-45NM Non-Mydriatic Retinal Camera (Canon Australia Pty. Ltd., Sydney, New South Wales, Australia). Four 45° nonstereoscopic retinal photographs of the disc and macula of each eye were taken for each participant. Participants sat for 3 minutes in a dark room without natural light to maximize physiologic pupil dilation, and photographs were then taken in the same room. The images were viewed immediately and repeated if unsatisfactory. AMD was later graded with the aid of “OptoLite/OptoMize Pro” software (Digital HealthCare Image Management Systems, Cambridge, United Kingdom) on high-resolution screens, where lesions within a grading circle of 6,000 μm, which was centered on the fovea, were recorded and verified. Graders were physicians with additional training in AMD grading. Photographs were blinded, and all retinal pathologies were graded twice, with disagreements being resolved by a senior ophthalmologist/grader. The kappa statistics for intergrader and intragrader reliability for our 2 graders ranged from 0.64 to 0.76 and from 0.60 to 1.00, respectively. To test the robustness of our results, and because there are many definitions of AMD, we employed 2 different definitions of early AMD based on the 2 most commonly used cutoff points for drusen size (21, 22): 1) the presence of drusen ≥63 μm, with or without the presence of hyper-/hypopigmentation (definition 1); and 2) the presence of drusen ≥125 μm, with or without the presence of hyper-/hypopigmentation (definition 2). Late AMD was defined as evidence of choroidal neovascularization or geographic atrophy. Geographic atrophy was defined as an area of retinal pigment epithelium depigmentation ≥175 μm that was roughly round or oval with sharp margins and visible underlying choroidal vessels.

Statistical analysis

All meat variables were analyzed categorically, based on approximate quartile groupings, with the first quartile being used as the reference category. Tests for trend across categories of meat intake were calculated using a pseudocontinuous variable—that is, the values 1, 2, 3, and 4 as continuous variables.

Early or late AMD was defined per person according to the status of the worse eye. Statistical analyses were performed using Stata, version 9.1 (Stata Corporation, College Station, Texas). Using multiple logistic regression, we estimated odds ratios for relations between intakes of different types of meat at baseline and the presence at follow-up of early AMD1 (definition 1), early AMD2 (definition 2), and late AMD, as compared with no AMD by any definition. All analyses were adjusted for age, sex, smoking (current, past, never), and energy intake (base model). The following potential confounders were added provisionally to the base model but were retained only if they changed the β coefficients (loge odds ratios) by ≥5%: vitamin C, vitamin E, β-carotene, zinc, lutein/zeaxanthin, trans-unsaturated fatty acids, omega 3 fatty acids, saturated fat, cholesterol, total fat, alcohol, vegetable intake, fish intake, supplement use (vitamin C, vitamin E, cod liver oil, or fish oil), education, body mass index, and protein intake at baseline (23). Only body mass index and zinc, protein, and vitamin intake met this criterion. Consequently, the final model included these terms plus the terms in the base model.

Ethics approval

The human research ethics committees of the Cancer Council Victoria and the Royal Victorian Eye and Ear Hospital approved the study protocols, as well as the manner in which informed consent was obtained from participants. Study participants gave written consent for the investigators to access their medical records and to collect and store their macular photographs, anthropometric measurements, and biologic samples. They also consented to passive follow-up conducted through record linkage to electoral rolls, electronic phone books, the Victorian Cancer Registry, and death records. These procedures were carried out in concordance with the principles outlined in the Declaration of Helsinki.

RESULTS

Of the 6,734 eligible Melbourne Collaborative Cohort Study participants who participated in the AMD study, 6,339 (94.1%) had gradable macular photographs. At the time retinal photographs were taken, participants had a median age of 74 years (range, 66–85). We identified 1,861 cases of early AMD1 (29.4%), 1,011 cases of early AMD2 (16.0%), and 88 cases of late AMD (1.4%), which included 36 cases of exudative AMD and 52 cases of geographic atrophy. After excluding participants with outlier energy intakes and a history of acute myocardial infarction, angina, or diabetes at baseline, we included 5,604 participants with complete data in the analyses; of these, there were 1,680 cases of early AMD1 (30%), 910 cases of early AMD2 (16.2%), and 77 cases of late AMD (1.4%).

Table 1 gives the characteristics of the participants and nonparticipants eligible to take part in this study. The mean baseline age was 64 years; approximately 30% were former smokers, and less than one-quarter had completed high school. Importantly, red meat consumption far exceeded chicken and fish intakes. Specifically, approximately one-quarter of the population consumed red meat at least 10 times per week. Participants and nonparticipants who were alive were similar in age, smoking history, educational level, and dietary habits (intakes of meat, fatty acids, and antioxidants). However, deceased persons had been older than the participants, had smoked more, and had more often been male.

Table 1.

Demographic and Dietary Characteristics of Participants and Nonparticipants, Melbourne Collaborative Cohort Study, Melbourne, Victoria, Australia, 1990–1994a

 Participants (n = 5,999)
 
Nonparticipants
 
 Alive (n = 4,150)
 
Deceased (n = 1,033)
 
 No. or Mean (SD) No. or Mean (SD) No. or Mean (SD) 
Demographic characteristics       
    Mean age at baseline, years 63.7 (3.3)  64.5 (3.4)  65.3 (3.2)  
    Mean age at eye evaluation, years 74.4 (3.5)  NA  NA  
    Female sex 3,666 61 2,762 67 455 44 
    Current smoker 336 382 163 16 
    Former smoker 1,921 32 1,362 33 453 44 
    Completed high school 1,454 24 936 23 256 25 
    Mean body mass indexb 26.2 (3.9)  26.8 (4.3)  26.6 (4.8)  
    Does not exercise 910 15 736 18 181 18 
Dietary characteristics       
    Mean energy intake, kJ/day 9,462.9 (3,054.6)  9,181.5 (3,096.9)  9,607.2 (3,292.9)  
    Red meat intake, times/week       
        ≤4.5 1,470 24.5 981 23.6 215 20.8 
        5–6.5 1,475 24.6 1,019 24.6 232 22.5 
        7–9.5 1,671 27.9 1,182 28.5 305 29.5 
        ≥10 1,383 23.1 968 23.3 281 27.2 
    Processed red meat intake, times/week       
        ≤1 1,130 18.8 801 19.3 162 15.7 
        1.5 1,369 22.8 918 22.1 246 23.8 
        2–3.5 2,116 35.3 1,483 35.7 354 34.3 
        ≥4 1,384 23.1 948 22.8 271 26.2 
    Chicken intake, times/week       
        ≤1 2,034 33.9 1,465 35.3 386 37.4 
        1.5 1,132 18.9 748 18.0 198 19.2 
        2–3 1,577 26.3 1,064 25.6 255 24.7 
        ≥3.5 1,256 20.9 873 21.0 194 18.8 
    Intake of fish <2 times/week 3,763 63 2,509 60 812 79 
    Median alcohol intake, g/day 4.3c  1.9c  4.3c  
    Intake of vegetables <5 times/week 2,453 40.9 1,961 47.3 545 52.8 
    Mean fat intake, g/day 82.7 (30.5)  81.1 (31.4)  84.8 (32.4)  
 Participants (n = 5,999)
 
Nonparticipants
 
 Alive (n = 4,150)
 
Deceased (n = 1,033)
 
 No. or Mean (SD) No. or Mean (SD) No. or Mean (SD) 
Demographic characteristics       
    Mean age at baseline, years 63.7 (3.3)  64.5 (3.4)  65.3 (3.2)  
    Mean age at eye evaluation, years 74.4 (3.5)  NA  NA  
    Female sex 3,666 61 2,762 67 455 44 
    Current smoker 336 382 163 16 
    Former smoker 1,921 32 1,362 33 453 44 
    Completed high school 1,454 24 936 23 256 25 
    Mean body mass indexb 26.2 (3.9)  26.8 (4.3)  26.6 (4.8)  
    Does not exercise 910 15 736 18 181 18 
Dietary characteristics       
    Mean energy intake, kJ/day 9,462.9 (3,054.6)  9,181.5 (3,096.9)  9,607.2 (3,292.9)  
    Red meat intake, times/week       
        ≤4.5 1,470 24.5 981 23.6 215 20.8 
        5–6.5 1,475 24.6 1,019 24.6 232 22.5 
        7–9.5 1,671 27.9 1,182 28.5 305 29.5 
        ≥10 1,383 23.1 968 23.3 281 27.2 
    Processed red meat intake, times/week       
        ≤1 1,130 18.8 801 19.3 162 15.7 
        1.5 1,369 22.8 918 22.1 246 23.8 
        2–3.5 2,116 35.3 1,483 35.7 354 34.3 
        ≥4 1,384 23.1 948 22.8 271 26.2 
    Chicken intake, times/week       
        ≤1 2,034 33.9 1,465 35.3 386 37.4 
        1.5 1,132 18.9 748 18.0 198 19.2 
        2–3 1,577 26.3 1,064 25.6 255 24.7 
        ≥3.5 1,256 20.9 873 21.0 194 18.8 
    Intake of fish <2 times/week 3,763 63 2,509 60 812 79 
    Median alcohol intake, g/day 4.3c  1.9c  4.3c  
    Intake of vegetables <5 times/week 2,453 40.9 1,961 47.3 545 52.8 
    Mean fat intake, g/day 82.7 (30.5)  81.1 (31.4)  84.8 (32.4)  

Abbreviations: NA, not applicable; SD, standard deviation.

a

Excludes subjects with extreme energy intakes, acute myocardial infarction, angina, or diabetes at baseline and subjects with missing dietary data.

b

Weight (kg)/height (m)2.

c

Median value.

Similar to other observational studies (6), we found age and smoking to be significant risk factors for AMD. The odds ratios for any AMD (adjusted for smoking) in successive 5-year age groups, using the youngest age group (66–70.9 years) as the referent, were 1.08 (95% CI: 0.91, 1.28), 1.54 (95% CI: 1.29, 1.83), and 2.46 (95% CI: 1.78, 3.36), respectively. Current smokers, who comprised 6% of the cohort, had an odds ratio of 3.07 (95% CI: 1.52, 6.21) for late AMD in comparison with never smokers, after adjustment for age.

Table 2 displays multivariate-adjusted odds ratios for AMD at follow-up associated with baseline meat intakes. High red meat intake was positively associated in a significant dose-response fashion with early AMD, regardless of the definition used. Although associations were weaker for processed meat than for fresh red meat, higher consumption of salami or continental sausages was strongly associated with both early and late AMD. Total chicken intake was not associated with early AMD, but it was inversely associated with late AMD. Further adjustment for chicken and fish consumption in analyses of red meat intake and vice versa did not materially change the results. Fish consumption was not associated with AMD (data not shown).

Table 2.

Odds Ratiosa for Age-related Macular Degeneration According to Frequency of Meat Consumption, Melbourne Collaborative Cohort Study, Melbourne, Victoria, Australia, 1990–2006

Baseline Dietary Intake AMD Status at Follow-up
 
Early AMD (Definition 1) (Drusen ≥63 μm) (n = 1,680)
 
Early AMD (Definition 2) (Drusen ≥125 μm) (n = 910)
 
Late AMD (n = 77)
 
OR 95% CI P-Trend OR 95% CI P-Trend OR 95% CI P-Trend 
Red meat, times/week          
    ≤4.5 1.00 Reference <0.001 1.00 Reference 0.019 1.00 Reference 0.486 
    5–6.5 1.03 0.92, 1.30  1.20 0.98, 1.48  0.98 0.51, 1.87  
    7–9.5 1.20 1.02, 1.43  1.16 0.93, 1.43  0.84 0.42, 1.68  
    ≥10.0 1.47 1.21, 1.79  1.39 1.09, 1.78  1.47 0.70, 3.11  
Fresh red meat, times/week          
    ≤2.5 1.00 Reference 0.006 1.00 Reference 0.044 1.00 Reference 0.368 
    3–4 1.14 0.96, 1.36  1.22 0.98, 1.51  0.86 0.44, 1.68  
    4.5–6 1.19 1.00, 1.42  1.27 1.02, 1.59  0.88 0.44, 1.76  
    ≥6.5 1.34 1.10, 1.65  1.31 1.01, 1.69  1.48 0.69, 3.19  
Processed red meat, times/week          
    ≤1 1.00 Reference 0.040 1.00 Reference 0.170 1.00 Reference 0.875 
    1.5 0.88 0.73, 1.05  0.95 0.76, 1.19  0.80 0.39, 1.62  
    2–3.5 1.08 0.91, 1.28  1.09 0.89, 1.34  0.98 0.51, 1.85  
    ≥4 1.13 0.94, 1.37  1.12 0.89, 1.42  0.98 0.47, 2.05  
Chicken, times/week          
    ≤1 1.00 Reference 0.228 1.00 Reference 0.922 1.00 Reference 0.007 
    1.5 1.02 0.86, 1.20  0.97 0.78, 1.19  0.73 0.39, 1.37  
    2–3 1.03 0.88, 1.20  0.98 0.81, 1.19  0.49 0.26, 0.92  
    ≥3.5 1.13 0.95, 1.36  1.02 0.81, 1.28  0.43 0.20, 0.91  
Type of fresh red meat          
    Beef or veal schnitzel          
        Never or <1 time/month 1.00 Reference 0.193 1.00 Reference 0.183 1.00 Reference 0.592 
        1–3 times/month 1.04 0.90, 1.19  1.06 0.90, 1.26  0.93 0.54, 1.61  
        ≥1 time/week 1.14 0.94, 1.37  1.16 0.92, 1.46  0.81 0.36, 1.81  
    Beef or veal roast          
        Never or <1 time/month 1.00 Reference 0.373 1.00 Reference 0.419 1.00 Reference 0.072 
        1–3 times/month 1.04 0.91, 1.18  1.07 0.91, 1.27  0.61 0.34, 1.09  
        ≥1 time/week 1.08 0.91, 1.26  1.07 0.88, 1.31  1.83 1.06, 3.16  
    Beef steak          
        Never or <1 time/month 1.00 Reference 0.042 1.00 Reference 0.108 1.00 Reference 0.587 
        1–3 times/month 1.09 0.93, 1.28  1.05 0.86, 1.28  0.49 0.26, 0.94  
        ≥1 time/week 1.17 1.00, 1.36  1.16 0.96, 1.40  0.81 0.47, 1.38  
    Beef rissoles or meatloaf          
        Never or <1 time/month 1.00 Reference 0.074 1.00 Reference 0.119 1.00 Reference 0.823 
        1–3 times/month 1.13 0.99, 1.28  1.15 0.98, 1.35  1.13 0.68, 1.87  
        ≥1 time/week 1.13 0.96, 1.33  1.14 0.92, 1.40  1.03 0.53, 1.99  
    Mixed beef dishes (stews, curry, meat sauces)          
        Never or <1 time/month 1.00 Reference 0.065 1.00 Reference 0.181 1.00 Reference 0.533 
        1–3 times/month 1.02 0.86, 1.21  1.11 0.89, 1.38  0.61 0.32, 1.13  
        ≥1 time/week 1.14 0.97, 1.35  1.16 0.94, 1.43  0.76 0.42, 1.38  
    Lamb chops or roast          
        Never or <1 time/month 1.00 Reference 0.240 1.00 Reference 0.109 1.00 Reference 0.944 
        1–3 times/month 1.00 0.84, 1.19  1.12 0.90, 1.39  0.81 0.41, 1.61  
        ≥1 time/week 1.08 0.92, 1.27  1.19 0.96, 1.46  0.97 0.52, 1.78  
    Pork chops or roast          
        Never or <1 time/month 1.00 Reference 0.165 1.00 Reference 0.660 1.00 Reference 0.183 
        1–3 times/month 1.05 0.92, 1.20  1.00 0.85, 1.18  1.27 0.77, 2.10  
        ≥1 time/week 1.15 0.94, 1.40  1.07 0.84, 1.38  1.53 0.76, 3.10  
Type of processed red meat          
    Bacon          
        Never or <1 time/month 1.00 Reference 0.171 1.00 Reference 0.784 1.00 Reference 0.988 
        1–3 times/month 0.99 0.87, 1.13  0.99 0.84, 1.16  0.98 0.59, 1.64  
        ≥1 time/week 1.14 0.97, 1.34  1.04 0.85, 1.27  1.01 0.54-1.91  
    Salami or continental sausage          
        Never or <1 time/month 1.00 Reference 0.001 1.00 Reference 0.017 1.00 Reference 0.030 
        1–3 times/month 1.16 0.99, 1.38  1.18 0.96, 1.45  1.42 0.76, 2.64  
        ≥1 time/week 1.45 1.12, 1.88  1.36 1.00, 1.86  2.37 1.05, 5.37  
    Other luncheon meats          
        Never or <1 time/month 1.00 Reference 0.141 1.00 Reference 0.031 1.00 Reference 0.521 
        1–3 times/month 1.18 0.99, 1.39  1.20 0.98, 1.47  0.81 0.40, 1.65  
        ≥1 time/week 1.09 0.89, 1.33  1.23 0.97, 1.57  0.81 0.35, 1.93  
Baseline Dietary Intake AMD Status at Follow-up
 
Early AMD (Definition 1) (Drusen ≥63 μm) (n = 1,680)
 
Early AMD (Definition 2) (Drusen ≥125 μm) (n = 910)
 
Late AMD (n = 77)
 
OR 95% CI P-Trend OR 95% CI P-Trend OR 95% CI P-Trend 
Red meat, times/week          
    ≤4.5 1.00 Reference <0.001 1.00 Reference 0.019 1.00 Reference 0.486 
    5–6.5 1.03 0.92, 1.30  1.20 0.98, 1.48  0.98 0.51, 1.87  
    7–9.5 1.20 1.02, 1.43  1.16 0.93, 1.43  0.84 0.42, 1.68  
    ≥10.0 1.47 1.21, 1.79  1.39 1.09, 1.78  1.47 0.70, 3.11  
Fresh red meat, times/week          
    ≤2.5 1.00 Reference 0.006 1.00 Reference 0.044 1.00 Reference 0.368 
    3–4 1.14 0.96, 1.36  1.22 0.98, 1.51  0.86 0.44, 1.68  
    4.5–6 1.19 1.00, 1.42  1.27 1.02, 1.59  0.88 0.44, 1.76  
    ≥6.5 1.34 1.10, 1.65  1.31 1.01, 1.69  1.48 0.69, 3.19  
Processed red meat, times/week          
    ≤1 1.00 Reference 0.040 1.00 Reference 0.170 1.00 Reference 0.875 
    1.5 0.88 0.73, 1.05  0.95 0.76, 1.19  0.80 0.39, 1.62  
    2–3.5 1.08 0.91, 1.28  1.09 0.89, 1.34  0.98 0.51, 1.85  
    ≥4 1.13 0.94, 1.37  1.12 0.89, 1.42  0.98 0.47, 2.05  
Chicken, times/week          
    ≤1 1.00 Reference 0.228 1.00 Reference 0.922 1.00 Reference 0.007 
    1.5 1.02 0.86, 1.20  0.97 0.78, 1.19  0.73 0.39, 1.37  
    2–3 1.03 0.88, 1.20  0.98 0.81, 1.19  0.49 0.26, 0.92  
    ≥3.5 1.13 0.95, 1.36  1.02 0.81, 1.28  0.43 0.20, 0.91  
Type of fresh red meat          
    Beef or veal schnitzel          
        Never or <1 time/month 1.00 Reference 0.193 1.00 Reference 0.183 1.00 Reference 0.592 
        1–3 times/month 1.04 0.90, 1.19  1.06 0.90, 1.26  0.93 0.54, 1.61  
        ≥1 time/week 1.14 0.94, 1.37  1.16 0.92, 1.46  0.81 0.36, 1.81  
    Beef or veal roast          
        Never or <1 time/month 1.00 Reference 0.373 1.00 Reference 0.419 1.00 Reference 0.072 
        1–3 times/month 1.04 0.91, 1.18  1.07 0.91, 1.27  0.61 0.34, 1.09  
        ≥1 time/week 1.08 0.91, 1.26  1.07 0.88, 1.31  1.83 1.06, 3.16  
    Beef steak          
        Never or <1 time/month 1.00 Reference 0.042 1.00 Reference 0.108 1.00 Reference 0.587 
        1–3 times/month 1.09 0.93, 1.28  1.05 0.86, 1.28  0.49 0.26, 0.94  
        ≥1 time/week 1.17 1.00, 1.36  1.16 0.96, 1.40  0.81 0.47, 1.38  
    Beef rissoles or meatloaf          
        Never or <1 time/month 1.00 Reference 0.074 1.00 Reference 0.119 1.00 Reference 0.823 
        1–3 times/month 1.13 0.99, 1.28  1.15 0.98, 1.35  1.13 0.68, 1.87  
        ≥1 time/week 1.13 0.96, 1.33  1.14 0.92, 1.40  1.03 0.53, 1.99  
    Mixed beef dishes (stews, curry, meat sauces)          
        Never or <1 time/month 1.00 Reference 0.065 1.00 Reference 0.181 1.00 Reference 0.533 
        1–3 times/month 1.02 0.86, 1.21  1.11 0.89, 1.38  0.61 0.32, 1.13  
        ≥1 time/week 1.14 0.97, 1.35  1.16 0.94, 1.43  0.76 0.42, 1.38  
    Lamb chops or roast          
        Never or <1 time/month 1.00 Reference 0.240 1.00 Reference 0.109 1.00 Reference 0.944 
        1–3 times/month 1.00 0.84, 1.19  1.12 0.90, 1.39  0.81 0.41, 1.61  
        ≥1 time/week 1.08 0.92, 1.27  1.19 0.96, 1.46  0.97 0.52, 1.78  
    Pork chops or roast          
        Never or <1 time/month 1.00 Reference 0.165 1.00 Reference 0.660 1.00 Reference 0.183 
        1–3 times/month 1.05 0.92, 1.20  1.00 0.85, 1.18  1.27 0.77, 2.10  
        ≥1 time/week 1.15 0.94, 1.40  1.07 0.84, 1.38  1.53 0.76, 3.10  
Type of processed red meat          
    Bacon          
        Never or <1 time/month 1.00 Reference 0.171 1.00 Reference 0.784 1.00 Reference 0.988 
        1–3 times/month 0.99 0.87, 1.13  0.99 0.84, 1.16  0.98 0.59, 1.64  
        ≥1 time/week 1.14 0.97, 1.34  1.04 0.85, 1.27  1.01 0.54-1.91  
    Salami or continental sausage          
        Never or <1 time/month 1.00 Reference 0.001 1.00 Reference 0.017 1.00 Reference 0.030 
        1–3 times/month 1.16 0.99, 1.38  1.18 0.96, 1.45  1.42 0.76, 2.64  
        ≥1 time/week 1.45 1.12, 1.88  1.36 1.00, 1.86  2.37 1.05, 5.37  
    Other luncheon meats          
        Never or <1 time/month 1.00 Reference 0.141 1.00 Reference 0.031 1.00 Reference 0.521 
        1–3 times/month 1.18 0.99, 1.39  1.20 0.98, 1.47  0.81 0.40, 1.65  
        ≥1 time/week 1.09 0.89, 1.33  1.23 0.97, 1.57  0.81 0.35, 1.93  

Abbreviations: AMD, age-related macular degeneration; CI, confidence interval; OR, odds ratio.

a

Adjusted for age, sex, smoking (current, past, never), energy intake, vitamin C, vitamin E, β-carotene, zinc, lutein/zeaxanthin, body mass index, and energy-adjusted protein intake.

DISCUSSION

In this cohort study of persons aged 58–69 years at baseline, we observed a highly statistically significant positive association between baseline red meat intake and early AMD and an inverse association between chicken intake and late AMD. Our findings suggest that intakes of specific meats may have different effects on the risk of AMD.

In several studies focusing mostly on associations between fat intake and AMD, investigators reported on red meat intake as a potential risk factor for AMD (8–10, 24, 25). Results of these 3 cohort studies (8–10) and 2 cross-sectional studies (24, 25) are provided. Note that definitions of AMD vary between studies; thus, their results may not be directly comparable. Nonetheless, generally, positive associations were observed in the cohort studies. The Nurses’ Health Study and the Health Professionals Follow-up Study revealed an association of higher beef, pork, and lamb consumption with increased risk of any AMD; the odds ratio for intake more than 4–6 times per week as compared with 3 or fewer times per month was 1.35 (95% CI: 1.07, 1.69; P-trend = 0.01) (9). A hospital-based cohort study also found the highest quartile of red meat intake (versus the lowest) to be associated with increased risk of AMD progression (odds ratio = 2.09, 95% CI: 0.98, 4.47; P-trend = 0.11) (10). Similarly, in the Reykjavik Eye Study, an Icelandic population-based cohort study, investigators found frequent consumption of meat and meat products to be associated with increased 5-year risk of early AMD, after adjustment for age, sex, and smoking, although there was some evidence that more frequent meat consumption was associated instead with lower risk of soft drusen only (8). For consumption of meat 1–2 times per fortnight as compared with 4–7 times per week for early AMD (defined as large drusen or pigmentary abnormalities), the odds ratio was 0.53 (95% CI: 0.28, 0.99). However, lower meat consumption decreased the risk of pigmentary abnormalities but increased the risk of soft drusen (8). Conversely, cross-sectional studies found no associations between meat intake and AMD; the odds ratios for consumption of battered, fried, and processed meat more than 7 times per week as compared with less than twice per week and for the fifth quintile of beef and pork intake versus the first quintile were 1.1 (95% CI: 0.8, 1.5) (24) and 1.0 (95% CI: 0.6, 1.5) (25), respectively. None of the above studies (8–10, 24, 25) investigated the association between consumption of different types of red meat and AMD in detail, and we are unaware of any literature investigating the associations between chicken intake and AMD. Furthermore, including chicken in our model did not alter the odds ratio for red meat as a risk factor for AMD; thus, the inverse association seen for chicken is unlikely to reflect simply a lower intake of red meat. Although total fat has been associated with increased risk of AMD (9, 26) and meat may have a high fat content, there were no associations between total fat or saturated fat intake and AMD in our cohort. Additionally, controlling for fat intake did not affect our odds ratios relating to meat intake and AMD.

A high intake of red meat has been associated with higher levels of nitrosamines or N-nitroso compounds (11), heme iron (12), and advanced glycation end products (13). These highly reactive products may act independently or in combination to initiate AMD onset.

Interestingly, red meat but not white meat stimulates endogenous N-nitrosation in humans, leading to higher levels of N-nitroso compounds, which may explain the differences seen between red meat and chicken intake in our study (11, 27, 28). N-nitroso compounds spontaneously decompose or are metabolized into highly reactive methylating agents which could be toxic to the retina (29). Smoking is a risk factor for AMD, and it has been shown that at higher levels of red meat consumption, concentrations of N-nitroso compounds are of the same order as the concentrations of tobacco-specific N-nitroso compounds in cigarette smoke (28). Additionally, processing food by smoking or direct fire-drying often adds to the amount of nitrosamines, which could explain the strong associations we observed with intake of salami or continental sausage (30).

It has been suggested that the observed increase in levels of N-nitroso compounds can be attributed to heme iron, which is present in much higher levels in red meat than in white meat (12, 31). Heme, the iron-containing porphyrin pigment of meat, in itself is also a prooxidant (32). Increased concentrations of iron can generate highly reactive hydroxyl radicals via the Fenton reaction (33). Heme iron is more bioavailable than free iron (34), and AMD-affected maculae have significantly increased total iron concentrations, including chelatable iron, in the retinal pigment epithelial cells and Bruch's membrane, although the mechanism of iron transport and storage in the retina is not fully understood (33).

Foods that have been browned in cooking, as well as tobacco smoke, are sources of exogenous advanced glycation end products (13, 35). Advanced glycation end products have not only been shown to increase oxidative stress and inflammation (36), which have been postulated as key components in AMD pathogenesis, but have also been shown to stimulate vascular endothelial growth factors in retinal pigment epithelial cells (37). Additionally, advanced glycation end products have been localized within soft drusen and the retinal pigment epithelial cells in AMD eyes (38).

Our study had several strengths, including its large size, with dietary data being collected more than 10 years ago, before dietary modification for AMD was commonly recommended to people with signs of AMD. Thus, participants’ diets are unlikely to have been modified by knowledge of the disease. Moreover, 94% of study participants had gradable macular photographs, and outcome measures of AMD were validated and graded using 2 definitions of AMD. Additionally, nonresponse did not bias our findings, because participants’ and nonparticipants’ characteristics were comparable. Even though participants in our study were elderly, such that considerable attrition might be expected, our follow-up rate of 58% over a 13- to 16-year period was comparable to that of other well-conducted cohort studies of younger participants (39, 40). Finally, to account for possible etiologic differences between early and late forms of AMD, these forms of AMD were treated as separate outcomes.

However, our study was not without limitations. Combining the 2 subtypes of late AMD may have attenuated our results towards the null, if meat affected one subtype in a direction opposite that of the other. Unfortunately, with few late AMD cases, we were unable to evaluate them separately. Although our FFQ showed reasonably good repeatability, we were unable to assess the validity of red meat and chicken intake using an alternative dietary method. Nevertheless, comparisons of levels of fatty acids, which meat contains, estimated by the FFQ and in plasma phospholipids, showed corrected correlation coefficients ranging from 0.38 to 0.78 (41), similar to the correlation coefficients in other nutritional studies utilizing FFQs (42–44). In addition, errors in estimates of dietary intake are likely to have been nondifferential, attenuating our results towards the null. Despite the benefit of information on long-term diet, which is probably more relevant to the initiation of late AMD than recent diet, our study was limited by the onetime assessments of diet and maculopathy status. Specifically, the diets the participants followed in their 40s or younger may have been relevant to the AMD cases ascertained in this study. Furthermore, diet and other exposures that occurred during the follow-up period could have been relevant either to the initiation of early AMD or to the progression from early to late AMD.

We attempted to adjust for other potential lifestyle and dietary confounders, but like investigators in other observational studies, we were unable to exclude residual confounding due to inaccurately measured or unmeasured confounders. It is possible that participants who survived were different from those who died before eye evaluation; for example, if participants who had both a high-red-meat diet and AMD survived while participants with both a low-red-meat diet and AMD died before eye evaluation, survival bias could have resulted. However, there are no data in our study (Table 1) or in the literature to suggest this. Meat intake could also be a proxy for other risk factors or for other unknown substances that are associated with AMD. Chicken intake may be associated with a particular lifestyle which is protective against AMD. Another issue with analyses of diet and disease is the issue of “multiple comparisons.” Since most FFQs measure a large number of food items, some proponents argue that the P value should be adjusted according to the number of variables examined; however, this unduly reduces power. Individual associations should be evaluated on their own merits, and conclusions should be drawn in light of consistency with information both internal and external to the study (45–49). In our analyses, we did not adjust for multiple comparisons because our hypotheses were generated a priori; our findings were interpreted in the context of reviews of the literature currently available and were checked for internal consistency using 2 definitions of AMD. Nevertheless, in interpreting our results, the possibility of chance findings must be considered. Additionally, since few cases of late AMD were present, the study may have lacked statistical power to evaluate dietary associations with late AMD.

A high level of red meat consumption may be a novel risk factor for early AMD or may act as a marker for a group of persons with an increased risk from other lifestyle factors. Given the limited epidemiologic data on the relation between meat consumption and AMD and the biologic plausibility of such an association, confirmatory data from other cohort studies are needed.

Abbreviations

    Abbreviations
  • AMD

    age-related macular degeneration

  • CI

    confidence interval

  • FFQ

    food frequency questionnaire

Author affiliations: Centre for Eye Research Australia, University of Melbourne, Melbourne, Victoria, Australia (Elaine W. Chong, Luibov D. Robman, Khin Zaw Aung, Robyn H. Guymer); Cancer Epidemiology Centre, Cancer Council Victoria, Carlton, Victoria, Australia (Julie A. Simpson, Dallas R. English, Graham G. Giles); Centre for Molecular, Environmental, Genetic and Analytic Epidemiology, University of Melbourne, Melbourne, Victoria, Australia (Julie A. Simpson, Dallas R. English); and Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia (Allison M. Hodge).

This study was supported by a National Health and Medical Research Council (NHMRC) Public Health Scholarship (Elaine W. Chong) and an NHMRC Career Development Award (Robyn H. Guymer). The initial Melbourne Collaborative Cohort Study was funded by NHMRC Program Grant 209057, Capacity Building Grant 251533, and Enabling Grant 396414. The ophthalmic component of this study was funded by the Ophthalmic Research Institute of Australia, the John Reid Charitable Trust, Perpetual Trustees Australia Ltd., and the Royal Victorian Eye and Ear Hospital. The above organizations provided financial support only and did not direct the analysis or the writing of this paper.

The authors thank Theresa Dolphin for her contributions to this study.

This paper was presented in part at the 38th Annual Scientific Congress of the Royal Australian and New Zealand College of Ophthalmologists, Sydney, Australia, November 4–8, 2006.

Conflict of interest: none declared.

References

1.
Friedman
DS
O'Colmain
BJ
Munoz
B
, et al.  . 
Prevalence of age-related macular degeneration in the United States
Arch Ophthalmol
 , 
2004
, vol. 
122
 
4
(pg. 
564
-
572
)
2.
VanNewkirk
MR
Nanjan
MB
Wang
JJ
, et al.  . 
The prevalence of age-related maculopathy: The Visual Impairment Project
Ophthalmology
 , 
2000
, vol. 
107
 
8
(pg. 
1593
-
1600
)
3.
Seddon
J
Chen
CA
Ryan
SJ
Epidemiology of age-related macular degeneration
Retina
 , 
2006
, vol. 
Vol 2
 
4th ed
Philadelphia, PA
Elsevier/Mosby Publishing
(pg. 
1017
-
1027
)
4.
Access Economics Proprietary Ltd
Clear Insight: The Economic Impact and Cost of Vision Loss in Australia
 , 
2004
Melbourne, Australia
Eye Research Australia
 
5.
Brown
MM
Brown
GC
Sharma
S
, et al.  . 
The burden of age-related macular degeneration: a value-based analysis
Curr Opin Ophthalmol
 , 
2006
, vol. 
17
 
3
(pg. 
257
-
266
)
6.
Thornton
J
Edwards
R
Mitchell
P
, et al.  . 
Smoking and age-related macular degeneration: a review of association
Eye
 , 
2005
, vol. 
19
 
9
(pg. 
935
-
944
)
7.
Brown
DM
Kaiser
PK
Michels
M
, et al.  . 
Ranibizumab versus verteporfin for neovascular age-related macular degeneration
N Engl J Med
 , 
2006
, vol. 
355
 
14
(pg. 
1432
-
1444
)
8.
Arnarsson
A
Sverrisson
T
Stefánsson
E
, et al.  . 
Risk factors for five-year incident age-related macular degeneration: The Reykjavik Eye Study
Am J Ophthalmol
 , 
2006
, vol. 
142
 
3
(pg. 
419
-
428
)
9.
Cho
E
Hung
S
Willett
WC
, et al.  . 
Prospective study of dietary fat and the risk of age-related macular degeneration
Am J Clin Nutr
 , 
2001
, vol. 
73
 
2
(pg. 
209
-
218
)
10.
Seddon
JM
Cote
J
Rosner
B
Progression of age-related macular degeneration: association with dietary fat, transunsaturated fat, nuts, and fish intake
Arch Ophthalmol
 , 
2003
, vol. 
121
 
12
(pg. 
1728
-
1737
)
11.
Rowland
IR
Granli
T
Bockman
OC
, et al.  . 
Endogenous N-nitrosation in man assessed by measurement of apparent total N-nitroso compounds in faeces
Carcinogenesis
 , 
1991
, vol. 
12
 
8
(pg. 
1395
-
1401
)
12.
Carpenter
CE
Mahoney
AW
Evaluation of methods used in meat iron analysis and iron content of raw and cooked meats
J Agric Food Chem.
 , 
1995
, vol. 
43
 
7
(pg. 
1824
-
1827
)
13.
Goldberg
T
Cai
W
Peppa
M
, et al.  . 
Advanced glycoxidation end products in commonly consumed foods
J Am Diet Assoc.
 , 
2004
, vol. 
104
 
8
(pg. 
1287
-
1291
)
14.
Giles
GG
Epidemiology of food and disease: The Melbourne Cohort Study
Asia Pac J Clin Nutr
 , 
2004
, vol. 
13
 
suppl
pg. 
S30
 
15.
Giles
GG
English
DR
The Melbourne Collaborative Cohort Study. (IARC Scientific Publication no. 156)
 , 
2002
Lyon, France
International Agency for Research on Cancer
(pg. 
69
-
70
)
16.
Ireland
P
Jolley
D
Giles
G
, et al.  . 
Development of the Melbourne FFQ: a food frequency questionnaire for use in an Australian prospective study involving an ethnically diverse cohort
Asia Pac J Clin Nutr
 , 
1994
, vol. 
3
 
1
(pg. 
19
-
31
)
17.
Lewis
J
Milligan
G
Hunt
A
NUTTAB95—NUTrient Data TABle for Use in Australia
 , 
1995
Canberra, Australia
Australian Government Publishing Service
18.
RMIT Lipid Research Group
Fatty Acid Compositional Database
 
Brisbane
Australia
 
Xyris Software. 2001
19.
US Department of Agriculture and University of Minnesota Nutrition Coordinating Center
USDA-NCC Carotenoid Database for U.S. Foods—1998
 , 
1998
Washington, DC
US Department of Agriculture
20.
English
DR
MacInnis
RJ
Hodge
AM
, et al.  . 
Red meat, chicken, and fish consumption and risk of colorectal cancer
Cancer Epidemiol Biomarkers Prev
 , 
2004
, vol. 
13
 
9
(pg. 
1509
-
1514
)
21.
Bird
AC
Bressler
NM
Bressler
SB
, et al.  . 
An international classification and grading system for age-related maculopathy and age-related macular degeneration. The International ARM Epidemiological Study Group
Surv Ophthalmol
 , 
1995
, vol. 
39
 
5
(pg. 
367
-
374
)
22.
Klein
R
Davis
MD
Magli
YL
, et al.  . 
The Wisconsin Age-Related Maculopathy Grading System
Ophthalmology
 , 
1991
, vol. 
98
 
7
(pg. 
1128
-
1134
)
23.
Brown
CC
Kipnis
V
Freedman
LS
, et al.  . 
Energy adjustment methods for nutritional epidemiology: the effect of categorization
Am J Epidemiol
 , 
1994
, vol. 
139
 
3
(pg. 
323
-
338
)
24.
Heuberger
RA
Mares-Perlman
JA
Klein
R
, et al.  . 
Relationship of dietary fat to age-related maculopathy in the Third National Health and Nutrition Examination Survey
Arch Ophthalmol
 , 
2001
, vol. 
119
 
12
(pg. 
1833
-
1838
)
25.
Mares-Perlman
JA
Brady
WE
Klein
R
, et al.  . 
Dietary fat and age-related maculopathy
Arch Ophthalmol
 , 
1995
, vol. 
113
 
6
(pg. 
743
-
748
)
26.
Seddon
JM
Rosner
B
Sperduto
RD
, et al.  . 
Dietary fat and risk for advanced age-related macular degeneration
Arch Ophthalmol
 , 
2001
, vol. 
119
 
8
(pg. 
1191
-
1199
)
27.
Bingham
SA
Hughes
R
Cross
AJ
Effect of white versus red meat on endogenous N-nitrosation in the human colon and further evidence of a dose response
J Nutr
 , 
2002
, vol. 
132
 
11 suppl
(pg. 
3522S
-
3525S
)
28.
Bingham
SA
Pignatelli
B
Pollock
JR
, et al.  . 
Does increased endogenous formation of N-nitroso compounds in the human colon explain the association between red meat and colon cancer?
Carcinogenesis
 , 
1996
, vol. 
17
 
3
(pg. 
515
-
523
)
29.
Shuker
DE
Margison
GP
Nitrosated glycine derivatives as a potential source of O6-methylguanine in DNA
Cancer Res.
 , 
1997
, vol. 
57
 
3
(pg. 
366
-
369
)
30.
Hotchkiss
JH
Preformed N-nitroso compounds in foods and beverages
Cancer Surv
 , 
1989
, vol. 
8
 
2
(pg. 
295
-
321
)
31.
Cross
AJ
Pollock
JR
Bingham
SA
Haem, not protein or inorganic iron, is responsible for endogenous intestinal N-nitrosation arising from red meat
Cancer Res.
 , 
2003
, vol. 
63
 
10
(pg. 
2358
-
2360
)
32.
Sesink
AL
Termont
DS
Kleibeuker
JH
, et al.  . 
Red meat and colon cancer: dietary haem, but not fat, has cytotoxic and hyperproliferative effects on rat colonic epithelium
Carcinogenesis
 , 
2000
, vol. 
21
 
10
(pg. 
1909
-
1915
)
33.
Hahn
P
Milam
AH
Dunaief
JL
Maculas affected by age-related macular degeneration contain increased chelatable iron in the retinal pigment epithelium and Bruch's membrane
Arch Ophthalmol
 , 
2003
, vol. 
121
 
8
(pg. 
1099
-
1105
)
34.
Balder
HF
Vogel
J
Jansen
MC
, et al.  . 
Heme and chlorophyll intake and risk of colorectal cancer in the Netherlands cohort study
Cancer Epidemiol Biomarkers Prev
 , 
2006
, vol. 
15
 
4
(pg. 
717
-
725
)
35.
Uribarri
J
Cai
W
Sandu
O
, et al.  . 
Diet-derived advanced glycation end products are major contributors to the body's AGE pool and induce inflammation in healthy subjects
Ann N Y Acad Sci.
 , 
2005
, vol. 
1043
 (pg. 
461
-
466
)
36.
Cai
W
Gao
QD
Zhu
L
, et al.  . 
Oxidative stress-inducing carbonyl compounds from common foods: novel mediators of cellular dysfunction
Mol Med
 , 
2002
, vol. 
8
 
7
(pg. 
337
-
346
)
37.
Lu
M
Kuroki
M
Amano
S
, et al.  . 
Advanced glycation end products increase retinal vascular endothelial growth factor expression
J Clin Invest
 , 
1998
, vol. 
101
 
6
(pg. 
1219
-
1224
)
38.
Ishibashi
T
Murata
T
Hangai
M
, et al.  . 
Advanced glycation end products in age-related macular degeneration
Arch Ophthalmol
 , 
1998
, vol. 
116
 
12
(pg. 
1629
-
1632
)
39.
Klein
R
Klein
BE
Moss
SE
, et al.  . 
The relationship of retinopathy in persons without diabetes to the 15-year incidence of diabetes and hypertension: Beaver Dam Eye Study
Trans Am Ophthalmol Soc.
 , 
2006
, vol. 
104
 (pg. 
98
-
107
)
40.
Tan
JS
Wang
JJ
Flood
V
, et al.  . 
Dietary antioxidants and the long-term incidence of age-related macular degeneration: The Blue Mountains Eye Study
Ophthalmology
 , 
2008
, vol. 
115
 
2
(pg. 
334
-
341
)
41.
Hodge
AM
Simpson
JA
Gibson
RA
, et al.  . 
Plasma phospholipid fatty acid composition as a biomarker of habitual dietary fat intake in an ethnically diverse cohort
Nutr Metab Cardiovasc Dis
 , 
2007
, vol. 
17
 
6
(pg. 
415
-
426
)
42.
Hunter
D
Willett
W
Biochemical indicators of dietary intake
Nutritional Epidemiology
 , 
1998
New York, NY
Oxford University Press
(pg. 
174
-
243
)
43.
Hunter
DJ
Rimm
EB
Sacks
FM
, et al.  . 
Comparison of measures of fatty acid intake by subcutaneous fat aspirate, food frequency questionnaire, and diet records in a free-living population of US men
Am J Epidemiol
 , 
1992
, vol. 
135
 
4
(pg. 
418
-
427
)
44.
Ma
J
Folsom
AR
Shahar
E
, et al.  . 
Plasma fatty acid composition as an indicator of habitual dietary fat intake in middle-aged adults. The Atherosclerosis Risk in Communities (ARIC) Study Investigators
Am J Clin Nutr
 , 
1995
, vol. 
62
 
3
(pg. 
564
-
571
)
45.
Willett
W
Willett
W
Issues in analysis and presentation of dietary data
Nutritional Epidemiology
 , 
1998
New York, NY
Oxford University Press
(pg. 
321
-
345
)
46.
Rothman
KJ
No adjustments are needed for multiple comparisons
Epidemiology
 , 
1990
, vol. 
1
 
1
(pg. 
43
-
46
)
47.
Savitz
DA
Olshan
AF
Describing data requires no adjustment for multiple comparisons: a reply from Savitz and Olshan
Am J Epidemiol
 , 
1998
, vol. 
147
 
9
(pg. 
813
-
814
)
48.
Savitz
DA
Olshan
AF
Multiple comparisons and related issues in the interpretation of epidemiologic data
Am J Epidemiol
 , 
1995
, vol. 
142
 
9
(pg. 
904
-
908
)
49.
Feise
RJ
Do multiple outcome measures require p-value adjustment? [Electronic article]
BMC Med Res Methodol
 , 
2002
, vol. 
2
 pg. 
8