![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
,2
* Centre de recherche, Institut universitaire de gériatrie de Montréal, Montreal, Canada;
Département de nutrition, Université de Montréal, Montreal, Canada;
Institut universitaire de gériatrie de Sherbrooke, Université de Sherbrooke, Sherbrooke, Canada; and
** School of Human Nutrition and Dietetics, McGill University, Montreal, Canada
2To whom correspondence should be addressed. E-mail: bryna.shatenstein{at}umontreal.ca.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • dietary assessment • elderly • functional foods • past diet • questionnaires
Degenerative changes of aging are believed to result, in part, from the oxidative destruction of cells and tissues (1). A group of common food items referred to as "functional foods" (FF) has recently emerged as an active area of research (2,3). Functional foods are usually defined as healthful foods or food ingredients that have a potential health benefit beyond their nutrient content when consumed regularly in typical quantities as part of a varied diet (2,4). Their physiologically active components generally include antioxidants. Recent data from the Survey in Europe on Nutrition and the Elderly study show that antioxidant status is positively related to the number of vegetable servings consumed (5).
For example, whole grains, garlic (allyl compound) and soybeans are considered to decrease the risk of both heart disease and cancer (6,7). Carrots (carotenoids) contain antioxidants that act as modulators of cell growth regulation, regulators of gene expression and immunoregulators (8,9). Epidemiological evidence has linked tomatoes (lycopene) to protection against cardiovascular disease (10,11) and cancer (12). Tea (polyphenols) may prevent cancer and coronary heart disease (13,14). Fiber, both soluble and insoluble, may reduce the risk of heart disease and some types of cancer (2). Finally, vegetables and fruit are associated with lower risk of cancer (15,16).
Nutrition monitoring in the elderly, and research linking overall food consumption—including FF—with different models of aging (i.e., pathological, normal or successful) have the potential for reducing chronic diseases and improving quality of life by providing specific information to aid in the development of targeted dietary and other intervention strategies (17). Dietary assessment instruments must accurately capture and reflect seniors’ dietary reality, through the use of a flexible approach to data collection that respects individual preferences and abilities. Any dietary assessment method may be used with elderly subjects if practical data collection concerns are considered (e.g., literacy, degree of autonomy, cognitive integrity). Indeed, diminished memory, low vision, medication use impairing cognitive function and physical limitations (e.g., arthritis) may require the use of complementary dietary evaluation approaches (18,19). A food frequency questionnaire is considered appropriate for obtaining usual long-term dietary information from adults, particularly where the objective is to rank subjects and evaluate associations between dietary habits and health outcomes (20). Accurate dietary assessment by use of a food frequency questionnaire, however, requires a subject who is motivated, attentive and able to remember and conceptualize food intakes (21).
Three approaches are typically used to evaluate past food intake: 1) current food intake is taken to represent past consumption (22); 2) past food frequency consumption is obtained by retrospective recall (23); and 3) perception of dietary change is reported in relation to current consumption (24). Recall ability appears to be determined by diet stability and attentiveness to diet (25). When dietary habits have changed over time (for example, because of illness), recalled reports more closely estimate past consumption than current reports (25). Better recall validity is obtained when respondents are asked to respond only by nonquantitative consumption frequency categories (26). Finally, the reference time period should consider the food frequency questionnaire (FFQ) objectives and the age of the study participants (27).
A pilot study was conducted to evaluate the usefulness of the food frequency approach for assessing usual diet and FF consumption in an aging cohort. Its objectives were to 1) test and validate an FFQ adapted to an elderly, cognitively intact home-dwelling population and, more specifically, to 2) develop and test a method for measuring lifelong functional food intakes. The latter objective is the main focus of the present report.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
A convenience sample of healthy, home-dwelling elderly persons was recruited from a probability sample stratified by age and sex, initially recruited from provincial health insurance lists. This latter group had participated in an earlier pilot study testing recruitment strategies, measurement techniques and study instruments, to prepare for a longitudinal study of nutrition and healthy aging (28). Of the 69 participants in that pilot study, 66 had indicated their willingness (on their consent forms) to be recontacted for further research. The present pilot study was conducted at the Institut universitaire de gériatrie de Sherbrooke and Institut universitaire de gériatrie de Montréal, and coordinated from Montreal. It was approved by the Ethics Review Boards in both institutions, and all participants provided informed consent before data were collected.
Data collection instrument: food frequency.
The FFQ for Quebec home-dwelling elderly persons (FFQ-QE) was modified from a validated 73-item self-administered instrument developed in French and English by use of a list-based approach applied to Quebec population dietary data, to assess usual food consumption over the past 12 mo in adult Quebecers (FFQ-QA) (29,30). The questionnaire booklet included detailed instructions for completing the questionnaire and food portion photos, adapted in black and white from "Portion Photos of Popular Foods" (31), on the facing pages to aid in portion size estimation.
An interviewer-administered version of the FFQ-QA was developed, and changes were made to address needs of an elderly home-dwelling population with possible limitations (e.g., vision, reading skills, etc.). The list-based approach described above was reapplied to population dietary data for people aged 55+ y, and foods not contributing sufficiently to Quebec seniors’ diets (e.g., pizza, potato chips, salty snacks) were eliminated, whereas others (grapes) were added. Participant and interviewer instructions, food frequency categories and reference portion sizes for nutrient calculation purposes were also modified, culminating in the pretest version of the FFQ-QE. Portion size adjustments were based on age- and gender-specific portion data reported for NHANES II participants aged 65+ y (32). Plasticized full-size color food portion photos (31) were prepared for the face-to-face interviews, and minor changes reflecting the new FFQ reference portions were made to the reference portions on the questionnaire photos. The FFQ-QE queries consumption of 86 foods in 8 different categories (breads and cereals; meats, poultry, fish, eggs and vegetable proteins; soups; potatoes and rice; vegetables and fruits; dairy products; sweets and salty snacks; beverages). Frequency was ascertained by asking subjects to indicate the number of times they usually eat each food per d/wk/mo. In recognition of difficulties related to portion size estimation among older people (33), participants were asked to indicate whether they considered their usual portion size to be small, medium or large, compared to others of their age (Cynthia Thomson, University of Arizona, personal communication, 2002).
Functional foods.
A nonquantitative section was developed, tested and added to the FFQ to permit qualitative assessment of lifetime FF intakes (FF-FFQ). The evaluation targeted basic foods rather than nutraceuticals or other newer foods to focus on FF available for decades. Specific items were selected from those commonly reported in the literature and on credible websites (FFH, www.ag.uiuc.edu/ffh; IFIC, http://www.ificinfo.health.org/infofsn.htm), and from position papers such as those of the American Dietetic Association or Health Canada. We included foods from reports providing information on bioactive constituents, recommended minimal daily intakes, potential health benefits and experimental data where possible, while considering their availability in the local market at different time periods in Quebec, the presence of familiar food names and population consumption patterns by use of the list-based approach as described above. Some of these foods were already present in the frequency section of the FFQ-QE (whole wheat bread, high fiber cereals, cruciferous vegetables, among others), whereas others were constituents of broader food categories. The FF-FFQ ascertains consumption of 33 common food items (Table 1) considered by credible sources as FF at 5 sentinel age periods in older persons’ lives (currently, and at ages 65, 45, 25 and 10 y).
|
Self-assessment of response quality was included in the pilot version of the instrument as an internal qualitative reliability check. After completing the FF section of the instrument, participants were asked whether they had experienced difficulties remembering consumption of 1) particular foods or 2) foods at a particular age, and 3) their degree of confidence in their answers. These questions were based on dichotomous responses and a Likert scale.
All materials (recruitment procedures, letters, consent forms, questionnaires, participant materials) were prepared and pretested with elderly community-dwelling volunteers in both French and English and corrected for the final version. The FFQ-QE was validated against four nonconsecutive 24-h diet recalls. These included three collected in person and over the telephone by a dietitian-interviewer during the summer 2000 pilot study (28), and a fourth telephone-administered 24-h recall during the present FFQ-QE study. By use of SIDE/IML software (University of Iowa, 1996) running under SAS (v6.12; SAS Institute, Cary, NC), a statistical transformation was done on nutrient values obtained from each of the four dietary recalls to permit estimation of the usual intake distribution in the study population, adjusted for seasonal and intraindividual variability. A subsample of 20 volunteers completed the FF section a second time (4 wk later) to test its reliability.
Study procedure.
Recruitment, communication and interviewing procedures were identical at both sites. Two bilingual (French, English) dietitians with research training and prior experience in collecting dietary data from elderly people were hired and trained as research assistants. The 66 individuals who had agreed to consider participating in further research received a letter describing the present study and informing them that they would soon be contacted. They were telephoned within a week of mailing by the research assistants and invited to take part in the new pilot study. Those who agreed were randomly allocated to one of the questionnaire modes. During this contact, they were given the option of requesting an alternative method and asked whether they preferred to be interviewed at home or at the Research Centre.
Three modes of questionnaire completion were tested: face-to-face interview, telephone interview and self-administered by mail. Instructions for completing the self-administered questionnaire were carefully pretested to ensure comprehensibility. Materials (including consent forms) needed for completion of the questionnaire were mailed to participants allocated to the telephoned interviews or self-administered version, and preaddressed, stamped envelopes were provided for returning consent forms (telephone and self-administered) and questionnaires (self-administered). Those completing the questionnaire at home were asked to return it within 10 d. Upon receipt, the mailed FFQ-QE were verified for missing or implausible data and participants were telephoned to clarify problems. The fourth 24-h recall was conducted at the same time for these subjects, whereas respondents with plausible questionnaire responses were telephoned at a prearranged time to complete their recalls.
Analyses.
Nutrient analysis of the FFQ was done by use of DietSys software (32,34), which had been modified to reflect changes to the foods listed on the questionnaire, their corresponding portion sizes and use of the 1997 Canadian Nutrient File (35).
Consumption of FF was summed for FF reported as eaten either regularly or daily at each age period, from the 33 FF listed. This was done by frequency (daily, regularly) for all foods (n = 33) at all age periods (n = 5) for a total of 165 responses, and by discrete age period (33 responses at each one). To examine intake patterns at different ages, FF were ranked in descending order by percentage of participants reporting consumption of each FF in both frequency categories and each age period. To assess reliability, the percentage of identical responses furnished by the reliability subgroup on two occasions was determined by use of the same summing criteria for both FF-FFQ completed by them.
Nutrient values obtained from the FFQ-QE and from the adjusted 24-h recall data were classified into quartiles, and classification was compared by examining the percentage of respondents falling into each joint classification category. Joint classification was considered 1) identical, when in the same quartile; 2) contiguous, if in an adjacent quartile; 3) different, if in a more distant quartile; or 4) opposite, indicating frank misclassification, if classified in opposing extreme quartiles.
A t test for independent samples (P < 0.05) was used to compare men and women on their weekly consumption of different foods and nutrient values obtained from the FFQ-QE, the number of FF consumed and the percentage of identical responses in the reliability subsample on the FF section of the instrument. The relative validity of the FFQ-QE was evaluated by use of Spearman correlation analysis (attributed to nonnormal nutrient distributions) between nutrient values obtained from this instrument and the usual nutrient intakes assessed by the adjusted 24-h recalls. SPSS for Windows (v10; SPSS Inc., Chicago, IL) was used for all analyses, except for the SIDE/IML nutrient adjustments, which were done with SAS (v6.12; SAS Institute). Microsoft Excel was used to transfer data between SAS and SPSS.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Some 56% of the 51 respondents (51% female) were aged 70–79 y, and the remainder were over 80 y. The oldest participants were aged 86 y. Around 82% were francophone, mirroring the linguistic breakdown in Quebec (data not shown).
The overall refusal rate was 16%. Those who refused cited lack of interest or ill health. Despite efforts to allocate subjects equally to the 3 administration methods, 39% completed a face-to-face interview, 25% were done over the telephone and 35% of questionnaires were self-administered. Reasons for preferring methods other than face-to-face (and preferred method) included: visual problems (telephone), too demanding (self-administered), too busy or wanted to take his/her time (self-administered). Reasons for preferring a method other than self-administered (preferred method) included: disliked writing (face-to-face), felt isolated (face-to-face). On average, the entire questionnaire took 68 ± 29 min with the interviewer, 56 ± 12 min over the telephone and participants reported 57 ± 26 min for self-administered questionnaires. Interviewers reported that the FF component took 
25 min. It took 13 d to receive the completed self-administered FFQ-QE by mail (data not shown).
Nutrient intakes and validity of the FFQ-QE.
Selected results obtained from the FFQ-QE are presented in Table 2. Proportional macronutrient intakes were similar for lipids and carbohydrates, although men had significantly greater proportional intakes of protein than women. Mean (median) daily protein intakes were greater among men than women [79 ± 18 g (79) versus 67 ± 21 g (64), P < 0.05 (data not shown)].
|
40% of these could be classified into identical quartiles, >75% into identical and contiguous quartiles, a small proportion were classified differently overall and a very small percentage were frankly misclassified. With the exception of vitamin B-12 and fatty acids, correlations were statistically significant [P < 0.05 to P < 0.001 (Table 3)].
|
At the time of interview ("current"), both men and women reported regularly ("several times per month") eating 18 to 20 of the 33 FF listed. However, the number of FF eaten regularly decreased as respondents went further back in time, and, in this frequency category, women consistently reported consuming statistically higher numbers of FF than men at each age period in the past. Approximately 6 of the 33 FF were reported as eaten daily at the current age period, and this declined with time in almost linear fashion. Women reported consuming significantly greater numbers of FF daily than men at ages 45 and 25 y (Table 4).
|
|
Overall, 73% of participants’ responses to the FF on two separate occasions were identical (all age periods together), with mean reliability decreasing by 10% from their present age to age 10 y. Women (n = 11) had statistically higher percentages of identical responses overall compared to men (n = 9). This was also observed for the numbers of FF reported eaten at ages 65 and 45 y. Reliability did not differ by mode of questionnaire administration (Table 6).
|
Approximately half of the participants indicated having some problems with the different time periods queried in the FF section of the questionnaire. As might be expected, the proportions reporting difficulties with time periods increased steadily over time. However, only 18% of respondents indicated having trouble remembering particular foods. On the scale ranging from 0 (lowest confidence) to 10 (highest confidence), none indicated his/her degree of confidence to be low (5 or under); 50% were moderately confident of their answers, and almost 40% scored their level of confidence as high (9 or 10). Overall mean confidence in their answers was 8 (Table 7).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Assessment of intakes is crucial to providing evidence that FF do indeed benefit consumers. Whereas websites, published articles, position papers and numerous industry-source documents describe the health-giving properties of FF and contribute to debate on the effectiveness of whole foods versus extracts of their active components, methods for assessing their intakes remain within the realm of traditional dietary assessment methodology. In addition, assessing diet over time is limited by methodological issues such as data collection methodology, nutrient composition databases and protocols for probing for information about food sources of various nutrients (36). To date, evidence for the beneficial impact of FF comes mostly from case-control studies that use retrospective methods such as FFQ for evaluating food consumption. The FFQ context is considered advantageous for use in elderly populations because it does not rely on short-term memory and involves a relatively low respondent burden (37). Dietary intake patterns have typically been ascertained in cross-sectional studies comparing age groups (36). However, if these beneficial foods do have an impact on health in aging, assessment of their consumption should logically cover a long period in the past, as well as the present. However, remembering past food consumption is a difficult exercise that calls on cognitive strategies such as inference, social desirability and memory (27), which are employed more reliably among those with stable diets and a particular interest in food (25). In our study, interviewers reported that participants had some difficulties remembering their consumption patterns of all the foods listed in the FF section (Table 7). For more remote time periods, some participants initially answered "I imagine that I ate it," "It could be" or "Probably" but with probing were able to provide more certain responses. Their stated confidence in their responses was relatively good, but this must be taken with some reserve given that accurate assessment of respondents’ stated certainty of their confidence in their answers is not possible (38).
Judicious use of memory cues helped respondents situate their functional food intakes within the context of important life events or milestones, and clarified problems remembering foods eaten at different time periods. Indeed, memory performance can improve among older people when they use "encoding strategies" that help them organize their memories by forming verbal or visual associations with the information to be remembered. Requiring participants to evaluate information to be remembered, and interviewers’ use of strategies to support participants’ retrieval of information further enhance processing (39). More work needs to be done through the use of a "life events calendar" approach (40,41) to call on sentinel moments or milestones as a way of helping elderly people remember important past events such as dietary intakes.
Possible reasons for the decrease in numbers of FF eaten at the different ages relative to the 33 candidate FF may include memory lapses, lack of familiarity with certain foods eaten when younger, differing availability of these foods at past time periods and changing preferences for different foods over time. Clearly, it is easier to report on foods that one never ate, or remember those consumed on a daily basis, than those that are subject to more varied consumption. In trying to remember past food intake for a particular period, people tend to make "telescopic" errors, which may result when they "import" events that happened more recently than the reference period (27). Bias related to misclassification of dietary intakes is of particular concern in a study assessing past dietary intakes reported by elderly people (36). Still, participants were free of cognitive dysfunction (≥79 on the 3MS, 42) at entry into the study and the questionnaire was administered soon after recruitment. The reliability study, although small in scope, provides additional robust support for the precision of their responses.
Women reported significantly greater consumption of FF than men. Although it is tempting to suggest that women typically selected higher quality diets over their lifetime than men, or that they have better memories than their male counterparts, it must be remembered that women of that generation were typically responsible for all family food purchases and preparation. Thus, the probable explanation lies in their better overall food knowledge and a lifetime habit of being more attentive to food [such as individuals with specific dietary habits (25)] as compared to men of their own age cohort, supporting others’ observations. For example, in a South African FFQ validation study that used anthropometric measurement, heart rate monitoring and 24-h urinary nitrogen excretion, repeatability between FFQ administered at a 6-mo interval was shown to be good among women but inconsistent in men (43). The authors suggested that gender differences in food awareness, purchasing and preparation activities could be responsible for this observation. However, others have reported no differences in accuracy of recall by gender (25) or even better reports among men (44).
Scant comprehensive dietary data have been collected over decades and even fewer have been published (36). To our knowledge no data have been published on the lifetime consumption of a series of FF in people aged 70 y and older. Consequently, we cannot compare our findings to those of others. Although temporal consumption of specific food groupings could be examined in individual studies, it is unlikely that others have looked at FF intakes at age periods similar to ours or from the same qualitative perspective. For example, it was recently reported that a nonquantitative, open-ended 7-item vegetable and fruit "screener" showed intakes of total vegetable and fruit consumption to be highest in people aged 65 y or more and to have increased over a 6-y period. However, the authors could not rule out demographic population shifts as a cause for positive changes in consumption (45).
A number of methodological issues must be considered when interpreting results. This research was conducted as a pilot study to generate data necessary for assessing feasibility issues, and to provide variance estimates needed for calculation of sample sizes, to prepare a protocol for a cohort study on nutrition and successful aging. Thus, the sample was relatively small, but was representative of the target population. Although all requests for administration mode (interviewer, telephone, self-administered) were accommodated because of the relatively small sample size, results of all three modes were examined together. Interviewers reported that face-to-face and telephone interviews were similar in terms of administration and response quality. However, subjects completing the self-administered instruments experienced some confusion in noting frequencies and/or quantities in the FFQ food list. In addition, there were some inconsistencies in responses for like foods between the main FFQ food list and the FF list. Although these discrepancies were queried and clarified in follow-up telephone calls, it is not clear what represents the "truth." This observation provides evidence to support the usefulness of a trained interviewer using probing strategies when collecting dietary data from elderly people (39).
The primary concern, in our opinion, is the issue of validity of reported past intakes. It is reported that food memory is influenced by dietary knowledge and practical food experience, as well as current diet (25). In the present context, this must be considered in conjunction with the availability of foods in the past and social conventions governing food-related activities in this population (43). It is also possible that subclinical age-related declines in attention and memory that precede dementia could contribute to reporting and/or food consumption bias (46). Admittedly, we were unable to substantiate participants’ reports of past (functional) food intakes. However, their responses showed an internal consistency and logical coherence from a temporal perspective. In particular, their reports of FF consumed at the age of 10 y were clearly indicative of a child’s diet. Because the majority of participants (64%) provided their answers spontaneously in an interview setting, it is highly unlikely that their responses were contrived to simulate food patterns of a youngster.
Further work will address the selection of specific FF and their presentation in the FF list, as well as strategies for regularly updating this list. A lifetime FF index has recently been constructed from this instrument to quantify lifelong FF intakes. This index will be tested in larger samples of elderly individuals followed in longitudinal studies, thereby furthering evaluation of our methodological approach, and clarification of the relationship between lifelong diet and health, with particular emphasis on FF. Differing patterns of consumption of foods with physiologically active components providing health benefits beyond basic nutrition could furnish crucial insights on the role of dietary patterns in aging, and specifically, the importance of FF in the "positive health paradigm."
Members of the Division on Nutrition and Healthy Aging, Québec Network on Aging Research, and their academic affiliation:
Isabelle Dionne, Université de Sherbrooke
Lise Dubois, Université Laval
Guylaine Ferland, Université de Montréal
Tamàs Fülöp, Université de Sherbrooke
Pierrette Gaudreau, Université de Montréal
Katherine Gray-Donald, McGill University
Danielle Jacques, Université de Sherbrooke
Marie-Jeanne Kergoat, Université de Montréal
Abdelouahed Khalil, Université de Sherbrooke
José Antonio Morais, McGill University
Hélène Payette, Université de Sherbrooke
Bryna Shatenstein, Université de Montréal
Daniel Tessier, Université de Sherbrooke
James Richard Wagner, Université de Sherbrooke
|
FOOTNOTES |
|---|
3 Abbreviations used: FF, functional foods; FFQ, food frequency questionnaire; FF-FFQ, nonquantitative functional foods frequency questionnaire; FFQ-QA, food frequency questionnaire-Quebec adult; FFQ-QE food frequency questionnaire-Quebec elderly.
Manuscript received 17 December 2002. Initial review completed 13 January 2003. Revision accepted 18 April 2003.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Martin, A., Prior, R., Shukitt-Hale, B., Cao, G. & Joseph, J. A. (2000) Effects of fruits, vegetables or vitamin E rich diet on vitamins E and C distribution in peripheral and brain tissues: implications for brain function. J. Gerontol. Biol. Sci. 55:B144-B151.
2. Hasler, C. M. (2000) The changing face of functional foods. J. Am. Coll. Nutr. 19:499S-506S.
3. Functional Foods for Health (FFH) University of Illinois at Chicago and University of Illinois at Urbana-Champaign www.ag.uiuc.edu/ffh (accessed January 2001).
4. American Dietetic Association (ADA) (1999) Position of The American Dietetic Association: functional foods. J. Am. Diet. Assoc. 99:1278-1285.[Medline]
5. Schlettwein-Gsell, D., Brubacher, D., Schroll Bjornsbo, K., Inelmen, E. M., Haller, J. & Van Staveren, W. (2002) Fruit and vegetable servings in ten SENECA towns and the impact of the midday meal. Age Nutr 5:17-22.
6. Craig, W. J. (1997) Phytochemicals: guardians of our health. J. Am. Diet. Assoc. 97(suppl.):199S-204S.
7. Milner, J. A. (1996) Garlic: its anticarcinogenic and antitumorigenic properties. Nutr. Rev. 54(suppl.):82S-86S.
8. Rock, C. L. (1997) Carotenoids: biology and treatment. Pharmacol. Ther. 75:185-197.[Medline]
9. Street, D. A., Comstock, G. W., Salkeld, R. M., Schuep, W. & Klag, M. J. (1994) Serum antioxidants and myocardial infarction: are low levels of carotenoids and alpha-tocopherol risk factors for myocardial infarction?. Circulation 90:1154-1161.
10. Clinton, S. K. (1998) Lycopene: chemistry, biology, and implications for human health and disease. Nutr. Rev. 56:35-51.[Medline]
11. Kohlmeier, L., Kark, J. D., Gomez-Gracia, E., Martin, B. C., Steck, S. E., Kardinaal, A. F., Ringstad, J., Thamm, M., Masaev, V., Riemersma, R., Martin-Moreno, J. M., Huttunen, J. K. & Kok, F. J. (1997) Lycopene and myocardial infarction risk in the EURAMIC study. Am. J. Epidemiol. 146:618-626.
12. Tzonou, A., Signorello, L. B., Lagiou, P., Wuu, J., Trichopoulos, D. & Trichopoulou, A. (1999) Diet and cancer of the prostate: a case-control study in Greece. Int. J. Cancer 80:704-708.[Medline]
13. Yang, C. S., Chung, J. Y., Yang, G.-Y., Chhabra, S. K. & Lee, M. J. (2000) Tea and tea polyphenols in cancer prevention. J. Nutr. 130:472S-478S.
14. Stensvold, I., Tverdal, A., Solvoll, K. & Foss, O. P. (1992) Tea consumption: relationship to cholesterol, blood pressure, and coronary and total mortality. Prev. Med. 21:546-553.[Medline]
15. Block, G., Patterson, B. & Subar, A. (1992) Fruit, vegetables, and cancer prevention: a review of the epidemiological evidence. Nutr. Cancer 18:1-29.[Medline]
16. American Cancer Society. Advisory Committee on Diet, Nutrition, and Cancer Prevention (1996) Guidelines on diet, nutrition and cancer prevention: reducing the risk of cancer with healthy food choices and physical activity. CA Cancer J. Clin. 46:325-341.[Abstract]
17. Health Canada () Nutrition for Health: An Agenda for Action http://www.hc-sc.gc.ca/hppb/nutrition/pube/action/agenda.htm (Spring 1996; accessed March 2003).
18. van Staveren, W. A., de Groot, L., Blauw, Y. H. & Reggy, P. J. (1994) Assessing diets of elderly people: problems and approaches. Am. J. Clin. Nutr. 59(suppl.):221S-223S.
19. Payette, H., Guigoz, Y. & Vellas, B. J. (1999) Study design for nutritional assessments in the elderly. Yu, Byung, Pal eds. Methods in Aging Research 1999:301-320 CRC Press Boca Raton, FL. .
20. Willett, W. (1998) Nutritional Epidemiology. 2nd ed. Monographs in Epidemiology and Biostatics 30 1998 Oxford University Press New York, NY/Oxford, UK.
21. Barrett-Connor, E. (1991) Nutrition epidemiology: how do we know what they ate?. Am. J. Clin. Nutr. 54:182S-187S.
22. Dwyer, J. T., Gardner, J., Halvorsen, K., Krall, E. A., Cohen, A. & Valadian, I. (1989) Memory of food intake in the distant past. Am. J. Epidemiol. 130:1033-1046.
23. Bakkum, A., Bloemberg, B., van Staveren, W. A., Verschuren, M. & West, C. E. (1988) The relative validity of a retrospective estimate of food consumption based on a current dietary history and a food frequency list. Nutr. Cancer 11:41-53.[Medline]
24. Metzner, H. L., Thompson, F. E., Lamphiear, D. E. & Hawthorne, V. M. (1988) Correspondence between perceptions of change in diet and 15-year change in diet reports in the Tecumseh diet methodology study. Nutr. Cancer 11:61-71.[Medline]
25. Lindsted, K. D. & Kuzma, J. W. (1989) Long-term (24-year) recall reliability in cancer cases and controls using a 21-item food frequency questionnaire. Nutr. Cancer 12:135-149.[Medline]
26. Fraser, G. E., Lindsted, K. D., Knutsen, S. F., Beeson, W. L., Bennett, H. & Shavlik, D. J. (1998) Validity of dietary recall over 20 years among California Seventh-Day Adventists. Am. J. Epidemiol. 148:810-818.
27. Dwyer, J. T. & Coleman, A. (1994) Cognitive issues in two dietary survey methods. Wright, J. D. Ervin, B. Briefel, R. eds. Consensus Workshop on Dietary Assessment: Nutrition Monitoring and Tracking the Year 2000 Objectives 1994:98-131 U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, MD. .
28. Payette, H., Members of The Division on Nutrition and Healthy Aging, Québec Network on Aging Research (2001) La cohorte québécoise pour l’étude de la nutrition et du vieillissement réussi. Étude pilote. Study Report 2001.
29. Shatenstein, B., Nadon, S. & Ferland, G. (2002) Challenges in developing an instrument to assess overall dietary and nutritional quality. Can. J. Diet. Prac. Res. 63(suppl.):113(abs.).
30. Shatenstein, B., Nadon, S. & Ferland, G. (2002) Validation of a population-based food frequency questionnaire using four food records. Abstracts of XVI IEA World Congress of Epidemiology meeting. Int. J. Epidemiol. Suppl. August 2002 (abs.).
31. American Dietetic Association (ADA) (1997) Hess, M. A. eds. Portion Photos of Popular Foods 1997 The American Dietetic Association and Center for Nutrition Education University of Wisconsin-Stout .
32. National Cancer Institute (NCI) (1997) Health Habits and History Questionnaire: Diet History and Other Risk Factors. DietSys Version 3.0 User’s Guide. National Cancer Institute, Information Management Services. Block Dietary Data Systems 1997http://www-dccps.ims.nci.nih.gov/ARP/DietSys/uguide.html (accessed July 2000).
33. Ervin, R. B. & Smiciklas-Wright, H. (2001) Accuracy in estimating and recalling portion sizes of foods among elderly adults. Nutr. Res. 21:703-713.
34. Block, G., Coyle, L. M., Hartman, A. M. & Scoppa, S. M. (1994) Revision of dietary analysis software for the Health Habits and History Questionnaire. Am. J. Epidemiol. 139:1190-1196.
35. Health Canada (1982) Canadian Nutrient File 1982 Bureau of Nutritional Sciences, Health & Welfare Canada Ottawa (Ontario), Canada.
36. Wakimoto, P. & Block, G. (2001) Dietary intake, dietary patterns, and changes with age: an epidemiological perspective. J. Gerontol. Ser. A 56A(Special Issue II):65-80.
37. Smith, W., Mitchell, P., Reay, E. M., Webb, K. & Harvey, P.W.J. (1998) Validity and reproducibility of a self-administered food frequency questionnaire in older people. Aust. N. Z. J. Public Health 22:456-463.[Medline]
38. Dwyer, J. T. & Coleman, K. A. (1997) Insights into dietary recall from a longitudinal study: accuracy over four decades. Am. J. Clin. Nutr. 65(suppl.):1153S-1158S.
39. Ervin, R. B. & Smiciklas-Wright, H. (1998) Using encoding and retrieval strategies to improve 24-hour dietary recalls among older adults. J. Am. Diet. Assoc. 98:989-994.[Medline]
40. Schwarz, N. & Sudman, S. (1994) Autobiographical Memory and the Validity of Retrospective Reports. 1994 Springer-Verlag, New York New York.
41. Means, B., Swan, G. E., Jobe, J. B. & Esposito, J. L. (1991) An alternative approach to obtaining personal history data. Biemer, P. P. Groves, R. M. Lyberg, L. E. Mathiowetz, N. A. Sudman, S. eds. Measurement Errors in Surveys 1991:167-183 Wiley, New York New York. .
42. Teng, E. L. & Chui, H. C. (1987) The modified Mini-Mental State (3MS) Examination. J. Clin. Psychiatry 48:314-318.[Medline]
43. Charlton, K. E. & Lambert, E. V. (1999) Validation of a food frequency questionnaire in older South Africans. S. Afr. Med. J. 89:184-189.
44. Hansson, L. M., Galanti, M. R. & Bergstrom, R. (2000) Factors affecting reproducibility of dietary reports using food frequency questionnaires. Eur. J. Clin. Nutr. 54:658-664.[Medline]
45. Stables, G. J., Subar, A. F., Patterson, B. H., Dodd, K., Heimendinger, J., Van Duyn, M. A. & Nebeling, L. (2002) Changes in vegetable and fruit consumption and awareness among US adults: results of the 1991 and 1997 5 a Day for Better Health Program surveys. J. Am. Diet. Assoc. 102:809-817.[Medline]
46. Engelhart, M. J., Geerlings, M. I., Ruitenberg, A., van Swieten, J. C., Hofman, A., Witteman, J.C.M. & Breteler, M.M.B. (2002) Dietary intake of antioxidants and risk of Alzheimer Disease. J. Am. Med. Assoc. 287:3223-3229.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
