Dairy, fruit, and vegetable intakes and functional limitations and disability in a biracial cohort: the Atherosclerosis Risk in Communities Study

¹ From the Departments of Nutrition (DKH, JS, and PSH), Epidemiology (JS) and Biostatistics (JC), University of North Carolina at Chapel Hill.

² Supported by grant 1RO3 AG022062-01 from the National Institute on Aging (to DKH, JS, and JC).

³ Reprints not available. Address correspondence to J Stevens, Department of Nutrition, CB #7461, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599. E-mail: June_Stevens{at}unc.edu.

ABSTRACT  
Background: Dairy, fruit, and vegetable intakes may be associated with functional limitations and disability through their role in muscle function, osteoporosis, and prevention of the oxidative damage associated with aging and chronic disease.

Objective: The associations between dairy, fruit, and vegetable intakes and functional limitations and disability were examined in African Americans and whites (baseline age: 45–64 y; n = 9404) in the Atherosclerosis Risk in Communities (ARIC) Study.

Design: Logistic regression analyses were used to ascertain the associations between usual dairy, fruit, and vegetable intakes obtained at baseline by using a food-frequency questionnaire and lower-extremity function, activities of daily living (ADLs), and instrumental ADLs (IADLs) self-reported 9 y later in models stratified by race and sex.

Results: Baseline dairy, fruit, and vegetable intakes tended to be inversely associated with impaired lower-extremity function, ADLs, and IADLs 9 y later, particularly in African American women. For example, in African American women, baseline dairy intakes were inversely associated with impaired ADLs and IADLs [odds ratio (95% CI): 0.60 (0.40, 0.90) and 0.69 (0.48, 0.98) in the 3rd and 1st tertiles of intake, respectively; P for trend < 0.05]. Combined baseline intakes of fruit and vegetables were also inversely associated with impaired lower-extremity function, ADLs, and IADLs [odds ratio (95% CI): 0.67 (0.47, 0.95), 0.52 (0.36, 0.76), and 0.64 (0.45, 0.90), respectively; P for trend < 0.05].

Conclusions: Dairy, fruit, and vegetable intakes may be inversely associated with functional limitations and disability. Further research is needed to ascertain the effect of diet on subsequent functional limitations and disability.

Key Words: Diet • functional limitations • activities of daily living • ADLs • instrumental ADLs • IADLs • ARIC Study

INTRODUCTION  
Although there have been recent declines in the disability rate among US elderly, the absolute number of disabled elderly is expected to triple between 1985 and 2050 simply because of the dramatic increase in the number of elderly that will occur in that time (1). Disability among the elderly leads to a poorer quality of life and has been shown to predict dependency and institutionalization as well as greater health care needs and greater utilization of the health care system (2). Previous research on functional limitations and disability identified several modifiable risk factors, including obesity, lack of physical activity, alcohol consumption, and smoking (3, 4). However, little is known about the role of diet.

A literature review identified nutritional status as one of several key risk factors for functional limitations and disability that deserves further investigation (3). Poor nutrition may lead to lack of appetite, functional limitations and disability, changes in quality of life, morbidity, and mortality (5). Functional limitations and disability may, however, also lead to poor nutrition because of an impaired ability to shop, cook, and feed oneself. Poor diet is an important modifiable risk factor for the development of several chronic diseases associated with aging, such as cardiovascular disease, diabetes, osteoporosis, and cancer (6). Dietary factors, particularly B vitamins and antioxidants, were previously shown to play a role in cognitive function (6, 7).

Previous studies examining the association between diet and functional limitations and disability used nutritional risk screening tools (8-13), healthy eating indexes (14, 15), dietary biomarkers (16-20), and macronutrient or micronutrient intakes (or both) (20-24). However, to our knowledge, no studies have examined the association of specific types of foods or food groups with functional limitations and disability. Moreover, despite the fact that African Americans have a higher prevalence of functional limitations and disability than do whites (25), studies examining the associations between diet and functional limitations and disability have been conducted almost exclusively in predominantly white populations. Dietary intakes may differ between African Americans and whites because of socioeconomic and cultural differences leading to differences in food preferences, customs surrounding eating, beliefs about the association between diet and disease, and other factors (26-28). Sex differences in functional limitations and disability (2, 4, 29) and diet (27, 28) also have been documented.

Previous studies suggested that there may be an association between functional limitations and disability and calcium, vitamin D, and antioxidants through their role in muscle function, osteoporosis, and preventing the oxidative damage associated with aging and chronic disease (7, 16, 18, 20, 30-32). These nutrients are commonly found in dairy, fruit, and vegetables. Therefore, the purpose of this study is to examine baseline dairy, fruit, and vegetable intakes in middle-aged African American and white men and women and their functional limitations and disability 9 y later.

SUBJECTS AND METHODS  
Subjects
The Atherosclerosis Risk in Communities (ARIC) Study is a prospective, multicenter investigation of atherosclerosis and cardiovascular disease (33). Baseline data were collected between 1987 and 1989 in 2635 African American women, 1631 African American men, 6050 white women, and 5428 white men (n = 15 744). Participants were selected to be representative of adults aged 45–64 y in 4 communities in the United States: Forsyth County, NC; Jackson, MS; the northwestern suburbs of Minneapolis; and Washington County, MD. Approximately 46% of the eligible participants in Jackson and 65% of those in the other 3 communities entered the cohort (34). Participants who entered the cohort were more educated, had higher incomes, and were less likely to be current smokers than were those who did not enter the cohort. Participants were reexamined in the clinic at a maximum of 4 visits at 3-y intervals.

The study was approved by the institutional review boards of the 4 participating centers and the Public Health Institutional Review Board on Research Involving Human Subjects at the University of North Carolina at Chapel Hill. Written informed consent was obtained from all subjects.

Functional limitations and disability
Functional limitations and disability were assessed by self-report at visit 4 (1996–1998) only. The questionnaire identified 12 activities, and participants were asked to indicate whether they had no difficulty, some difficulty, or much difficulty or were unable to do these activities at all when they were by themselves and did not have the use of aids. The 12 activities were grouped into lower-extremity function, activities of daily living (ADLs), and instrumental ADLs (IADLs), as was done in the third National Health and Nutrition Examination Survey (35). Lower-extremity function consisted of the following 5 activities: walking one-quarter of a mile; walking up 10 steps without resting; stooping, crouching, or kneeling; lifting or carrying something as heavy as 10 pounds; and standing up from an armless straight chair. ADLs consisted of the following 4 activities: walking from one room to another on the same level; getting in or out of bed; eating or drinking from a glass; and dressing oneself. IADLs consisted of the following 3 activities: doing chores around the house; preparing own meals; and managing money. Lower-extremity function, ADLs and IADLs were categorized as "no impairment" (ie, participants reported no difficulty on any activity) and "impaired" (ie, participants reported any difficulty—some difficulty, much difficulty, or unable to do—on any activity.

Dietary assessment
Participants' usual dietary intake over the preceding year was assessed by using an interviewer-administered, 66-item, semiquantitative food-frequency questionnaire (FFQ) at visit 1. The questionnaire was based on the 61-item instrument designed and validated by Willett et al (36). The frequency of intake was ranked in 9 categories ranging from "almost never" to "> 6 times/d." Intakes of daily nutrients and individual food items from the FFQ were calculated at the Channing Laboratory, Harvard Medical School. The reliability of the FFQ for intake of nutrients was previously examined among African Americans and whites in the ARIC Study (37).

Food groups were created by using frequencies of consumption of foods from the following categories: dairy [skim (1% fat) or low-fat (2% fat) milk, whole milk, yogurt, ice cream, cottage cheese or ricotta cheese, and other cheeses, plain or as part of a dish]; fruit (fresh apples or pears; oranges; orange or grapefruit juice; peaches, apricots, or plums; bananas; and other fruit); vegetables [string or green beans; broccoli; cabbage, cauliflower, or brussel sprouts; carrots; corn; spinach, collards, or other greens; peas or lima beans; dark yellow, winter squash; sweet potatoes; tomatoes or tomato juice; beans or lentils; and potatoes (mashed or baked, not fried)]; and fruit and vegetables combined. Standard serving sizes from the FFQ were used to compile the number of servings from each food group. Vegetables in mixed meat dishes (ie, sandwiches or mixed dish, stew, casserole, lasagna, or spaghetti sauce, etc., that contain beef, pork, or lamb as well as vegetables) were not included in the vegetable group for these analyses. Daily servings of each food group were categorized into sex-specific tertiles by using the average of the tertile cutoffs in whites and African Americans.

Covariates
Body mass index (BMI; in kg/m²) was calculated by using measured weight and height at baseline. Participants were asked the highest grade or year of school they had completed, and education level was categorized as less than a high school diploma, high school diploma or vocational school, and college or graduate or professional school. Baseline smoking status was obtained by questionnaire and categorized as current, former, or never.

Exclusions
Race was assessed by self-identification at baseline by a single choice from a checklist (ie, white, African American, American Indian or Alaskan Indian, or Asian or Pacific Islander). African American participants in Minnesota and Maryland were excluded (n = 55) because there were too few to support modeling. Participants who died before visit 4 (n = 1441), missed visit 4 (n = 2661), or lacked information on functional limitations and disability at visit 4 (n = 138) were excluded. Participants from whom >10 responses on the FFQ were missing and those with extremely low (<600 kcal in men and <500 kcal in women) or extremely high (>4200 kcal in men and >3600 kcal in women) energy intakes were also excluded (n = 230). Information on functional limitations and disability were not collected at baseline. As a proxy for these measures, we excluded from our analyses participants who used a wheelchair, crutches, walker, or cane (n = 61); those with prevalent coronary artery disease, stroke, cancer, or chronic lung disease (n = 1481); and those with poor self-rated health at baseline (n = 105) in an attempt to exclude participants who may have had functional limitations or disability at baseline. We excluded from our analyses participants who used a wheelchair, crutches, or walker or walked with a cane (n = 61); those with prevalent coronary artery disease, stroke, cancer, or chronic lung disease (n = 1481); and those with poor self-rated health at baseline (n = 105). We also excluded participants for whom pertinent covariates were missing (n = 168). The final sample size for these analyses was 9404, or 59.7% of the original sample (n = 3943 white women, 1327 African American women, 3413 white men, and 721 African American men).

Statistical analysis
Logistic regression was used to examine the associations between tertiles of food group intake and impaired lower-extremity function, ADLs, and IADLs (SAS PROC LOGISTIC software, version 8.2; SAS Institute, Cary, NC; 38). The lowest tertile of intake for each food group was set as the reference group. Three-way interactions between sex, race, and each food group variable and two-way interactions between sex or race and each food group were tested, and significant interactions were found at = 0.05. Of the 3-way interactions, there were significant sex x race xdairy interactions for lower-extremity function (P = 0.05), ADLs (P < 0.0001), and IADLs (P = 0.02) and significant sex x race x fruit and vegetables combined interactions for ADLs (P = 0.04). Multiple two-way interactions were also significant. Therefore, all analyses were stratified by sex and race. Models were adjusted for baseline age, education level, smoking status, total energy intake, BMI, and field center (ie, the community). Analyses adjusting for baseline physical activity, diabetes and hypertension status, and income did not change appreciably the significance or direction of the associations and therefore are not presented here. Tests for linear trends across increasing tertiles of food group intake used the median value in each tertile as a continuous variable in the logistic regression models.

RESULTS  
Information on baseline diet intake and functional limitations and disability was available for 9404 African American and white men and women at follow-up 9 y later (range: 7.4–11.9 y). Participants who were excluded were significantly (P < 0.01) more likely than were participants included in the study sample to be men (46.1% and 44.0%, respectively), African American (34.4% and 21.8%), and older (55.0 ± 5.8 and 53.6 ± 5.6 y); to have a higher mean BMI (28.0 ± 5.8 and 27.5 ± 5.1); and to have less than a high school education (33.0% and 17.8%). In addition, participants excluded with the use of the proxy functional-limitation variables were significantly (P < 0.0001) more likely than were participants included in the sample to be older (55.5 ± 5.6 and 53.6 ± 5.6 y, respectively), to be current smokers (26.0% and 21.0%), and to have less than a high school education (25.1% and 17.8%) at baseline. These excluded participants also had a significantly (P < 0.0001) higher prevalence than did included participants of impaired lower-extremity function (61.8% and 47.9%, respectively), ADLs (23.7% and 13.5%), and IADLs (31.9% and 17.9%) at follow-up.

The baseline characteristics of the final sample are shown in Table 1. The mean age at baseline was 53.6 y (range: 44–66 y). As expected, African American women had a higher mean BMI than did white men and women and African American men. More African American men than the other race-and-sex groups were current smokers. African American men and women were less likely to have completed high school than were white men and women. Men had fewer mean daily servings of fruit and vegetables than did women, and African Americans had fewer mean daily servings of dairy than did whites.

View this table:
TABLE 1. Selected baseline characteristics by race and sex group: the ARIC Study¹

 
At follow-up, African American women had the highest prevalence of impaired lower-extremity function, ADLs, and IADLs, with white women, African American men, and white men following in declining order of prevalence (Table 2). Although the mean age at follow-up was only 62.6 y (range: 52–75 y), the prevalence of impaired lower-extremity function ranged from 37.2% in white men to 66.9% in African American women. The prevalence of impaired ADLs ranged from 10.5% in white men to 21.8% in African American women, and that of impaired IADLs ranged from 10.8% in white men to 29.8% in African American women.

View this table:
TABLE 2. . Impaired lower-extremity function, activities of daily living (ADLs), and instrumental ADLs (IADLs) at 9-y follow-up by race and sex group: the ARIC Study¹

 
Among the white women, those in the highest tertile of dairy and fruit intakes at baseline had significantly lower odds of impaired lower-extremity function and IADLs at follow-up than did those in the lowest tertile after adjustment for baseline age, education, smoking status, energy intake, BMI, and field center (P for trend < 0.05; Table 3). There were no significant associations between baseline vegetable intakes and impaired lower-extremity function, ADLs, or IADLs at follow-up. However, there was a trend toward lower odds of impaired lower-extremity function and IADLs at follow-up among white women with the highest baseline intakes of fruit and vegetables combined in the fully adjusted model (P for trend < 0.10).

View this table:
TABLE 3. . Adjusted odds ratios (ORs) and 95% CIs of impaired lower-extremity function, activities of daily living (ADLs), and instrumental ADLs (IADLs) by tertile (T) of dairy, fruit, and vegetable intakes in white and African American women: the ARIC Study¹

 
The odds of impaired ADLs and IADLs at follow-up were significantly lower among African American women in the highest tertile of baseline dairy intakes than among those in the lowest tertile after adjustment for baseline age, education, smoking status, energy intake, BMI, and field center (P for trend < 0.05; Table 3). Those in the highest tertile of baseline fruit intake had significantly lower odds of impaired lower-extremity function, ADLs, and IADLs at follow-up than did those in the lowest tertile in age-adjusted models (P for trend < 0.05). These associations remained significant in the fully adjusted model for lower-extremity function and ADLs but not for IADLs. High baseline intake of vegetables was also significantly associated with lower odds of impaired IADLs at follow-up in the fully adjusted model (P for trend < 0.05). The odds of impaired lower-extremity function, ADLs, and IADLs at follow-up were significantly lower in those with the highest baseline intakes of fruit and vegetables combined than in those with the lowest intakes in the fully adjusted model (P for trend < 0.05).

Among the white men, there were significantly higher odds of impaired lower-extremity function and ADLs at follow-up in those in the highest tertile of dairy intake at baseline than in those in the lowest tertile after adjustment for baseline age (P for trend < 0.05; Table 4). This association remained significant for lower-extremity function (P for trend < 0.05) after adjustment for baseline education, smoking status, energy intake, BMI, and field center. Because the dairy items in the 66-item FFQ were not specific enough to separate out low- and high-fat dairy, a single question regarding the frequency of intake of low-fat (2% fat) or skim (1% fat) milk was used as a surrogate. Analyses of low-fat or skim milk showed no significant association between baseline intake of low-fat or skim milk and odds of impaired lower-extremity function or ADLs at follow-up [odds ratio (95% CI): 1.04 (0.87, 1.25) and 0.81 (0.61, 1.06), respectively, in the 3rd tertile compared with the 1st tertile]. White men in the highest tertiles of baseline intake of fruit and fruit and vegetables combined had significantly lower odds of impaired lower-extremity function at follow-up than did those in the lowest tertiles in the fully adjusted model (P for trend < 0.05). There was also a trend toward lower odds of impaired lower-extremity function at follow-up among white men with the highest baseline vegetable intake than among those with the lowest intake in the fully adjusted model (P for trend < 0.10). There were no significant associations between intake of fruit, vegetables, and fruit and vegetables combined at baseline and impaired ADLs and IADLs at follow-up.

View this table:
TABLE 4. . Adjusted odds ratios (ORs) and 95% CIs of impaired lower-extremity function, activities of daily living (ADLs), and instrumental ADLs (IADLs) by tertile (T) of dairy, fruit, and vegetable intakes in white and African American men: the ARIC Study¹

 
The odds of impaired lower-extremity function, ADLs, and IADLs at follow-up were significantly lower among African American men in the highest tertile of baseline intake of dairy than among those in the lowest tertile after adjustment for baseline age (P for trend < 0.05;Table 4). These associations remained significant for lower-extremity function and ADLs after adjustment for baseline education, smoking status, energy intake, BMI, and field center (P for trend < 0.05). African American men in the highest tertile of fruit intake at baseline also had significantly lower odds of impaired lower-extremity function and IADLs than did those in the lowest tertile in age-adjusted models (P for trend < 0.05). These associations remained significant in the fully adjusted model for lower-extremity function only. High intakes of vegetables and fruit and vegetables combined were also significantly associated with lower odds of impaired lower-extremity function in age-adjusted models (P for trend < 0.05). The trend toward lower odds of impaired lower-extremity function remained in the fully adjusted model for fruit and vegetables combined (P for trend < 0.10).

DISCUSSION  
The role of diet in chronic disease prevention is well established (39, 40), but little is known about the role of diet in preventing functional limitations and disability. The present study shows that, in general, intakes of dairy, fruit, and vegetables were inversely associated with functional limitations and disability at follow-up among African American and white men and women in the ARIC Study cohort. Although not always reaching significance, high dairy intake was associated with lower odds of impaired lower-extremity function, ADLs, and IADLs in white women and African American men and women, but not in white men. High fruit intake was also associated with lower odds of impaired lower-extremity function, ADLs, and IADLs in white men and African American men and women and with impaired lower-extremity function and IADLs in white women. There tended to be an association between high vegetable intake and lower odds of impaired ADLs and IADLs in African American women and an association between high vegetable intake and lower odds of impaired lower-extremity function in African American men.

There is little published information on the associations between diet and functional limitations and disability. Several cross-sectional studies found an inverse association between the ability to perform ADLs and IADLs and higher scores on nutritional risk-screening tools, which include questions on eating recommended servings of dairy, fruit, and vegetables (8-12). High nutrition risk scores were also associated with increased incidence of disability at 12- and 22-mo follow-ups (11, 13). Less is known about the role of diet patterns and functional limitations and disability. In a cross-sectional study, elderly men and women with "better" nutrition practices, as ascertained by using a summary score that included questions on food choices and meal patterns, were more likely to have better ADL scores than were their counterparts with "worse" nutrition practices (14). Among middle-aged women, but not men, consuming an "unhealthy" diet containing no whole-grain bread, no low-fat milk, and <1 serving fresh fruit or vegetables/d was associated with increased risk of poor physical function at 5-y follow-up (15). Similarly, we found that consuming a diet high in dairy, fruit, and vegetables is generally associated with lower odds of impaired lower-extremity function, ADLs, and IADLs than is consuming a diet low in these food groups.

Although several studies have investigated the association between nutrients from dairy, fruit, and vegetables and functional limitations and disability and have found inconsistent results (16-21), the association between dairy, fruit, and vegetable intakes and functional limitations and disability is biologically plausible. Dairy, the primary source of calcium and vitamin D, may decrease the risk of functional limitations and disability associated with osteoporosis, osteoporosis-related fractures, and decreased muscle strength (7, 16, 30). Antioxidants found in fruit and vegetables may reduce the accumulation of oxidative damage in tissues and, therefore, slow the onset of functional limitations and disability associated with aging as well as decrease the risk of oxidative-related chronic diseases, such as cardiovascular disease and cancer, that may lead to functional limitations and disability (18, 20, 31, 32).

A strength of this study is that it included large numbers of African American and white men and women for whom there was information on diet and functional limitations and disability. Most previous studies examined these associations in smaller samples. In addition, previous studies examining diet and functional limitations and disability were conducted in predominately white populations, even though African American men and women tend to report more functional limitations and limitations in ADLs and IADLs than do same-sex whites (25). In the ARIC Study cohort, African American men and women had a significantly higher prevalence of functional limitations and disability than did their white counterparts. Observed differences in dietary intake between African Americans and whites may be influenced by socioeconomic and cultural differences (26-28); therefore, it is important to examine the associations between diet and functional limitations and disability in African Americans and in other racial or ethnic groups. The strongest associations between high intakes of dairy, fruit, and vegetables and lower odds of impaired lower-extremity function, ADLs, and IADLs at follow-up were seen in African American women. This finding may reflect differences in the types of dairy, fruit, and vegetables consumed by African Americans and whites and the association between the individual nutrients, such as antioxidants, found in the different types of dairy, fruit, and vegetables and the participants' functional limitations and disability. Previous studies showed that African Americans have higher intakes of dark green vegetables and of vitamins A and C than do whites (28, 41). Although the trends in African American men in this study were similar, there were fewer significant associations between dairy, fruit, and vegetable intakes and impaired lower-extremity function, ADLs, and IADLs, possibly because of limitations in power among African American men due to the smaller sample size.

A limitation of the work presented here is the lack of information on functional limitations and disability at baseline. Information on lower-extremity function, ADLs, and IADLs was not collected until visit 4, 9 y after baseline. Therefore, the true incidence of impaired lower-extremity function, ADLs, and IADLs could not be assessed. To decrease the likelihood of including in the study participants who had impaired lower-extremity function, ADLs, and IADLs at baseline, we applied a somewhat extensive list of exclusions, which included indicators of chronic disease, self-rated health, and mobility (10% of the original sample). These exclusions resulted in a relatively healthy cohort, but we cannot be certain that none of the participants in our analysis had been functionally impaired at baseline. In additional analyses that included participants with prevalent coronary artery disease, stroke, cancer, or chronic lung disease at baseline, the associations between the 4 food groups and functional limitations and disability were similar to the results presented here. Functional limitations and disability could influence food intake and could also be influenced by food intake. Therefore, we cannot rule out the presence of some reverse causality here.

Another limitation was the method used to assess dietary intakes. A single, 66-item FFQ was used to characterize usual intakes of dairy, fruit, and vegetables. However, an FFQ provides an imprecise means of ranking intakes among persons (36). Moreover, assessment of dietary intake at just one period of time may be inadequate to capture important dietary exposure. The imprecision of the dietary data may have reduced the ability of this study to detect associations between dairy, fruit, and vegetable intakes and impaired lower-extremity function, ADLs, and IADLs. Although analyses were adjusted for BMI and physical activity (data not shown), diet may serve as a proxy measure for other, relevant, healthy lifestyle characteristics. Other limitations include the use of self-reported diet, functional limitation, and disability information and loss to follow-up (9% died and 17% were lost to follow-up), which possibly resulted in selection bias for the final samples used for analysis.

To our knowledge, the current study is the first to examine the associations between intakes of dairy, fruit, and vegetables and functional limitations and disability among African Americans and whites. In general, dairy, fruit, and vegetable intakes appear to be inversely associated with impaired lower-extremity function, ADLs, and IADLs. Thus, the current recommendations for daily consumption of 2–3 servings of low-fat dairy, 2–4 servings of fruit, and 3–5 servings of vegetables should be investigated for their potential to reduce the prevalence of functional limitations and disability in an aging population.

ACKNOWLEDGMENTS  
We thank the staff and participants in the ARIC Study for their important contributions.

DKH contributed to the experimental design, analysis and interpretation of the data, and writing of the manuscript. JS contributed to the experimental design, analysis and interpretation of data, and critical revision of the manuscript and also obtained funding. JC contributed to the experimental design, analysis and interpretation of data, and critical revision of the manuscript. PSH contributed to the analysis and interpretation of data and provided significant advice in preparing the manuscript. None of the authors had any conflicts of interest.

REFERENCES  

1. Manton KG, Corder L, Stallard E. Chronic disability trends in the elderly United States populations: 1982–1994. Proc Natl Acad Sci U S A 1997;94:2593–8.
2. Fried LP, Guralnik JM. Disability in older adults: evidence regarding significance, etiology, and risk. J Am Geriatr Soc 1997;45:92–100.
3. Stuck AE, Walthert JM, Nikolaus T, Bula CJ, Hohmann C, Beck JC. Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med 1999;48:445–69.
4. Guralnik JM, Fried LP, Salive ME. Disability as a public health outcome in the aging population. Annu Rev Public Health 1996;17:25–46.
5. Amarantos E, Martinez A, Dwyer J. Nutrition and quality of life in older adults. J Gerontol 2001;56A(special issue II):54–64.
6. Meydani M. Nutrition interventions in aging and age-associated disease. Ann N Y Acad Sci 2001;928:226–35.
7. Rosenberg IH, Miller JW. Nutritional factors in physical and cognitive functions of elderly people. Am J Clin Nutr 1992;55(suppl):1237s–43s.
8. Wolinsky FD, Coe RM, Miller DK, Prendergast JM. Measurement of the global and functional dimensions of health status in the elderly. J Gerontol 1984;39:88–92.
9. Johnson MA, Brown MA, Poon LW, Martin P, Clayton GM. Nutritional patterns of centenarians. Int J Aging Hum Dev 1992;34:57–76.
10. Jensen GL, Kita K, Fish J, Heydt D, Frey C. Nutrition risk screening characteristics of rural older persons: relation to functional limitations and health care charges. Am J Clin Nutr 1997;66:819–28.
11. Boult C, Krinke UB, Urdangarin CF, Skarin V. The validity of nutritional status as a marker for future disability and depressive symptoms among high-risk older adults. J Am Geriatr Soc 1999;47:995–9.
12. Sharkey JR, Haines PS, Zohoori N. Community-based screening: association between nutritional risk status and severe disability among rural home-delivered nutrition participants. J Nutr Elder 2000;20:1–15.
13. Bryant LL, Shetterly SM, Baxter J, Hamman RF. Modifiable risks of incident functional dependence in Hispanic and Non-Hispanic white elders: the San Luis Valley Health and Aging Study. Gerontology 2002;42:690–7.
14. Duffy ME, MacDonald E. Determinants of functional health of older persons. Gerontology 1990;30:503–9.
15. Stafford M, Hemingway H, Stansfield SA, Brunner E, Marmot M. Behavioural and biological correlates of physical functioning in middle aged office workers: the UK Whitehall II study. J Epidemiol Community Health 1998;52:353–8.
16. Zamboni M, Zoico E, Tosoni P, et al. Relation between vitamin D, physical performance, and disability in elderly persons. J Gerontol Med Sci 2002;57A:M7–11.
17. Verreault R, Semba RD, Volpato S, Ferrucci L, Fried LP, Guralnik JM. Low serum vitamin D does not predict new disability or loss of muscle strength in older women. J Am Geriatr Soc 2002;50:912–7.
18. Snowdon DA, Gross MD, Butler SM. Antioxidants and reduced functional capacity in the elderly: findings from the Nun Study. J Gerontol Med Sci 1996;51A:M10–6.
19. Ortega RM, Manas LR, Andres P, et al. Functional and psychic deterioration in elderly people may be aggravated by folate deficiency. J Nutr 1996;126:1992–9.
20. Cesari M, Pahor M, Bartali B, et al. Antioxidants and physical performance in elderly persons: the Invecchiare in Chianti (InCHIANTI) study. Am J Clin Nutr 2004;79:289–94.
21. Sharkey JR, Giuliani C, Haines PS, Branch LG, Busby-Whitehead J, Zohoori N. Summary measure of dietary musculoskeletal nutrient (calcium, vitamin D, magnesium, and phosphorus) intakes is associated with lower-extremity physical performance in homebound elderly men and women. Am J Clin Nutr 2003;77:847–56.
22. Bianchetti A, Rozzini R, Carabellese C, Zanetti O, Trabucchi M. Nutritional intake, socioeconomic conditions, and health status in a large elderly population. J Am Geriatr Soc 1990;38:521–6.
23. Sonn U, Rothenberg E, Steen B. Dietary intake and functional ability between 70 and 76 years of age. Aging Clin Exp Res 1998;10:324–31.
24. Pradignac A, Schlienger JL, Velten M, Mejean L. Relationships between macronutrient intake, handicaps, and cognitive impairments in free living elderly people. Aging Clin Exp Res 1995;7:67–74.
25. Blackwell DL, Tonthat L. Summary health statistics for the U.S. population: National Health Interview Survey, 1997. Vital Health Stat 10 2002;204:1–92.
26. Dwyer J. Nutritional problems of elderly minorities. Nutr Rev 1994;52:s24–7.
27. Patterson BH, Harlan LC, Block G, Kahle L. Food choices of Whites, Blacks, and Hispanics: data from the 1987 National Health Interview Survey. Nutr Cancer 1995;23:105–19.
28. Shimakawa T, Sorlie P, Carpenter MA, et al. Dietary intake patterns and sociodemographic factors in the Atherosclerosis Risk in Communities Study. Prev Med 1994;23:769–80.
29. Newman AB, Brach JS. Gender gap in longevity and disability in older persons. Epidemiol Rev 2001;23:343–50.
30. Heaney RP. Calcium, dairy products and osteoporosis. Am Coll Nutr 2000;19:83s–99s.
31. Meydani M. Dietary antioxidants modulation of aging and immune-endothelial cell interaction. Mech Ageing Develop 1999;111:123–32.
32. Lampe JW. Health effects of vegetables and fruit: assessing mechanisms of action in human experimental studies. Am J Clin Nutr 1999;70(suppl):475s–90s.
33. ARIC Investigators. The Atherosclerosis Risk in Communities (ARIC) Study: design and objectives. Am J Epidemiol 1989;129:687–702.
34. Jackson R, Chambless L, Yang K. Differences between respondents and nonrespondents in a multicenter community-based study vary by gender and ethnicity. J Clin Epidemiol 1996;49:1441–6.
35. Ostchega Y, Harris TB, Hirsch R, Parsons VL, Kington R. The prevalence of functional limitations and disability in older persons in the US: data from the National Health and Nutrition Examination Survey III. J Am Geriatr Soc 2000;48:1132–5.
36. Willett WC, Sampson L, Stampfer MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 1985;122:51–65.
37. Stevens J, Metcalf PA, Dennis BH, Tell GS, Shimakawa T, Folsom AR. Reliability of food frequency questionnaire by ethnicity, gender, and education. Nutr Res 1996;16:735–45.
38. Stokes ME, Davis CS, Koch GG. Categorical data analysis using the SAS system. 2nd ed. Cary, NC: SAS Institute Inc, 2000.
39. Key TJ, Allen NE, Spencer EA, Travis RC. The effect of diet on risk of cancer. Lancet 2002;360:861–8.
40. Kromhout D, Menotti A, Kesteloot H, Sans S. Prevention of coronary heart disease by diet and lifestyle: evidence from prospective cross-cultural, cohort, and intervention studies. Circulation 2002;105:893–8.
41. Swanson CA, Gridley G, Greenberg RS, et al. A comparison of diets of blacks and whites in three areas of the United States. Nutr Cancer 1993;20:153–65.