Precision and bias of food frequency-based measures of fruit and vegetable intakes.

Accurate assessment of fruit and vegetable intakes is critical for cancer control research and public health surveillance. This report examines the bias and precision of two commonly used food frequency questionnaire methods to assess fruit and vegetable intakes: (a) the 5 A Day method, based on seven items; and (b) the summation method, based on adding total servings of all fruit and vegetable items on a comprehensive (100+ item) food frequency questionnaire. Data are from three studies in which 24-h dietary recalls, food records, or serum carotenoid concentrations could be used as criterion measures (n = 260, 1031, and 342). Studies differed markedly in distributions of participants' age, race/ethnicity, sex, and socioeconomic status. Mean intakes of total fruit and vegetables based on the 5 A Day method were consistently lower than those from either the summation method (3.11 versus 4.06), 24-h recalls (3.32 versus 4.07), or food records (3.11 versus 3.46; all P < 0.01), and this was due primarily to underestimates of vegetable intake. Correlations of the 5 A Day and summation measures with all criterion measures were similar and were consistently higher for fruit (range, 0.33-0.57) than for vegetables (range, 0.24-0.32). These results, which were consistent across diverse participant samples, suggest that the 5 A Day method yields both biased and imprecise measures of vegetable intake and that research to improve this measure is needed.

A randomized trial of a tailored, self-help dietary intervention: the Puget Sound Eating Patterns study.

BACKGROUND: This study evaluated a tailored, multiple-component self-help intervention designed to promote lower fat and higher fruit and vegetable consumption. METHODS: Participants were 1,459 adults selected at random, stratified by sex and age (18-34, 35-54, 55-69), from enrollees of a large health maintenance organization. After completing a baseline telephone survey, participants were randomized to receive the intervention (consisting of a computer-generated personalized letter, a motivational phone call, a self-help manual, a package of supplementary materials, computer-generated behavioral feedback based on a self-administered food frequency questionnaire, and newsletters) or to receive no materials. Evaluation was based on 1,205 (86.5%) participants who completed both a 3- and a 12-month follow up survey. RESULTS: The intervention effect +/- SE for fat, based on a diet habits questionnaire, was -0.10 +/- 0.02 (P < 0.001), corresponding to a reduction of approximately 0.8 percentage points of percentage energy from fat. For fruits and vegetables, the intervention effect was 0.47 +/- 0.10 servings/day (P < 0.001). Intervention effects were similar across age and sex groups. CONCLUSIONS: Tailored, self-help interventions can effectively promote dietary change among both men and women and among younger as well as older adults.

The prevalence and impact of 'atypical' days in 4-day food records.

OBJECTIVE: To understand how days with atypical food intake affect estimates of usual nutrient intake from 4-day food records. PARTICIPANTS/SETTING: Secondary analyses of 4-day food records (4DFRs) (n = 2,560) collected from 1,090 women, aged 50 to 79 years, who participated in the Women's Trial Feasibility Study in Minority Populations, a randomized dietary intervention trial. DESIGN: Food records were classified as atypical if participants marked one or more day's food intake as "more than usual" or "less than usual." Total amounts and nutrient densities (percent of energy or grams per 1,000 kcal) were examined for all macronutrients, fiber, vitamin C, beta carotene, and calcium. STATISTICAL ANALYSIS: Contingency tables were used to examine associations of demographic characteristics with the likelihood of completing a 4DFR with atypical intake days. Analysis of variance was used to test whether nutrient intake differed among records with and without atypical days. Student t tests were used to identify any differences in total energy and percent energy from fat among typical and atypical intake days. RESULTS: Approximately 16% of records included at least 1 atypical day. Reporting less-than-usual intake was associated with younger age, higher income, and higher body mass index. Black women were less likely to report more-than-usual intake than whites and Hispanics. Records with less-than-usual intake had lower intakes of all nutrients analyzed except alcohol; however, there were no differences in nutrient densities. Records with more-than-usual intake had higher intakes of alcohol and all nutrients except beta carotene and vitamin C, with higher nutrient density measures of alcohol and decreased nutrient density measures of protein, vitamin C, and fiber. CONCLUSIONS: Atypical intake days are common in 4DFRs and they have a large effect on mean total intakes of most nutrients. APPLICATIONS: It is important for researchers to collect information on atypical intake days included in a 4-day food record. Strategies are needed to incorporate information on atypical intake days when analyzing and interpreting research results.

Differences in fat-related dietary patterns between black, Hispanic and White women: results from the Women's Health Trial Feasibility Study in Minority Populations.

OBJECTIVE: This report examines how sources of fat and patterns of fat-related dietary habits differed between black, Hispanic and white women participating in a randomized trial of a low-fat diet intervention. DESIGN: The intervention consisted of group sessions, which met weekly for 6 weeks, biweekly for 6 weeks and monthly for 9 months, and included didactic nutrition education and activities to provide motivation for sustained dietary change. Outcomes included total fat and fat from nine food groups from a food frequency questionnaire (FFQ), and a summary scale and five subscales that measure fat-related dietary habits. SETTING/SUBJECTS: Data are from 1702 post-menopausal women, recruited from clinical centres in Atlanta, Birmingham and Miami, with dietary assessments at baseline and 6 months post-randomization. RESULTS: Total fat intake was similar across race/ethnic groups at baseline, yet there were many differences in sources of fat and fat-related dietary habits. For example, blacks consumed less fat from dairy foods and more fat from meats than whites. Effects of the intervention on total fat intake or the summary fat-related dietary habits scale did not differ across race/ethnicity groups. There were, however, many differences in how the intervention affected sources of fat and fat-related dietary habits. For example, the intervention effect for added fats (e.g. butter and salad dressings) was -8.9 g for blacks and -12.0 g for whites (P<0.05). The intervention effect for adopting low-fat meat purchasing and preparation methods was larger for blacks than whites, and the intervention effect for replacing high-fat foods with fruits and vegetables was larger for Hispanics than whites. CONCLUSIONS: This study demonstrates that, if properly designed, a single nutrition intervention programme can work well even in groups with culturally diverse dietary patterns.

Feasibility of using volunteer research staff to deliver and evaluate a low-fat dietary intervention: the American Cancer Society Breast Cancer Dietary Intervention Project.

This report presents the results of a study to examine the feasibility of using volunteers as research staff for a randomized trial of whether reduction in dietary fat intake could prevent or delay breast cancer recurrence. We examined whether volunteers could be trained to recruit study participants, deliver a complex and intensive dietary intervention, and monitor intervention effectiveness. Volunteers, who were mostly employed nurses and dietitians, screened 521 women, of whom 293 were eligible and 144 were randomized. Participants were postmenopausal women under age 75, who had recently been diagnosed with breast cancer and treated with either mastectomy or lumpectomy. At 1 year postrandomization, 77% of intervention and 75% of control participants remained active in the study. Intervention effects (change in intervention group minus change in control group) at 3, 6, and 12 months postrandomization were 5.9, 8.4, and 7.2% energy from fat and 1.7, 3.0, and 3.5 kg body weight (all P < 0.001). These results were similar to those from other studies that used paid, professional staff to deliver and monitor interventions. Results from this feasibility study suggest that volunteer-based health organizations can provide research opportunities for health practitioners and can conduct high-quality research at lower costs.

Relationship of food groups and water intake to colon cancer risk.

The association between food groupings and adenocarcinoma of the colon was investigated in a population-based case-control study of men and women ages 30-62 years. Colon cancer cases (238 men and 186 women) diagnosed from 1985 to 1989 were identified from the Seattle-Puget Sound Surveillance, Epidemiology, and End Results Registry. Controls (224 men and 190 women) were selected using a random digit telephone dialing method. Dietary information was gathered using an 80-item food frequency questionnaire. Foods were grouped and analyzed by quartile of intake, with adjustment for age and total energy intake. Among women, a reduced risk of colon cancer was associated with a high intake of fruits and vegetables [adjusted odds ratio (OR) for highest versus lowest quartile, 0.48; 95% confidence interval (CI), 0.26-0.86; P for trend, P = 0.02]. Inverse associations were also observed for the consumption of total (hot and cold) cereals (OR, 0.47; 95% CI, 0.25-0.91; P = 0.05), dairy products (OR, 0.40; 95% CI, 0.21-0.79; P = 0.05), and water (OR for > 5 glasses/day versus < or = 2 glass/day, 0.55; 95% CI, 0.31-0.99; P = 0.004). Among men, colon cancer risk was inversely associated with the intake of breads and cereals (OR, 0.43; 95% CI, 0.22-0.82; P = 0.02) and hot cereal (OR for weekly versus never eating, 0.53; 95% CI, 0.32-0.87; P = 0.01). Water consumption was marginally associated with a decreased colon cancer risk among men as well (OR for > 4 glasses/day versus < or = 1 glass/day, 0.68; 95% CI, 0.38-1.22; P = 0.16). Total meat consumption was associated with an increased risk of distal colon cancer among men (OR, 2.20; 95% CI, 1.08-4.48; P = 0.01). These results were not confounded by body mass index or other measured health behaviors. Results of this research support previous findings which associate intake of fruits, vegetables, grains, and dairy products with reduced colon cancer risk, and meat intake with an increased colon cancer risk. This study also reports a new finding of a possible inverse association of water consumption (glasses of plain water per day) with colon cancer risk.

How women's adopted low-fat diets affect their husbands.

OBJECTIVES: One way of promoting a reduction in dietary fat intake is by changing the diet of family members. This study investigated the long-term effects of a low-fat dietary intervention on husbands of women who participated in the Women's Health Trial (WHT). METHODS: An average of 12 months after the end of the WHT, a randomly selected sample of participants' husbands was sent dietary and health questionnaires as part of a follow-up study of the maintenance of the low-fat diet among WHT participants. RESULTS: We found an absolute difference in fat intake between groups of 4 percentage points (32.9% energy from fat among intervention husbands [n = 188] vs 36.9% among control husbands [n = 180]). The wife's attitude and fat intake were among the most important predictors of her husband's fat intake, indicating that the effect of the WHT intervention on the husbands of participants was more likely due to their acceptance of lower-fat foods being served at home than to overt actions by the men. CONCLUSIONS: Our results suggest that a dietary intervention aimed at women can have an effect on their husbands and may be a cost-effective approach to healthy dietary change for both women and men.

Long-term maintenance of a low-fat diet: durability of fat-related dietary habits in the Women's Health Trial.

The Women's Health Trial (WHT) was a feasibility study for a randomized trial of a low-fat diet for the prevention of breast cancer. One year after the WHT was terminated, a random sample of 894 participants who had been active in the WHT for an average of 16 months (range = 5 to 37) completed questionnaires about their dietary habits (a 21-item instrument that measures five dimensions of low-fat dietary habits) and food intake (a food frequency questionnaire). Women who participated in the intervention program maintained most of the low-fat dietary habits adopted during the study: mean total fat intake increased from 37.8 g to 41.0 g and scales describing substitution of specially manufactured low-fat foods and modification of meats to be lower in fat increased only slightly (by 0.11 and 0.14, respectively, on a scale of 1 = always to 4 = never). Scales describing avoiding meat and avoiding fats as a flavoring increased by 0.23 and 0.22, respectively, which suggests some recidivism. Women in the control group lowered their dietary fat intake from 65.0 to 57.5 g, but all differences in fat intake and fat-related dietary habits scales between women in the control and intervention groups remained highly statistically significant. In multiple regression models, all five low-fat dietary habits scales were independently associated with percentage of energy from fat, but the strongest association was for avoiding fats as flavorings. These results suggest that substitutions of specially manufactured low-fat foods are easily adopted and maintained dietary changes, but that maintenance of new habits related to avoiding fats as flavorings and avoiding meat will require long-term reinforcement strategies.(ABSTRACT TRUNCATED AT 250 WORDS)

Maintenance of a low-fat diet: follow-up of the Women's Health Trial.

This report examines the maintenance of a low-fat diet 1 year on average after the completion of intervention sessions among participants in the Women's Health Trial (WHT). The WHT was a randomized controlled trial of the feasibility of adoption of a low-fat diet among women of moderate or increased risk of breast cancer, conduced in Seattle, Houston, and Cincinnati in 1985-1988. The women randomized to the low-fat diet attended an intensive dietary intervention program for 5-37 months. Intervention women were highly successful in reducing their dietary fat intake from 40.0% of energy intake at baseline to 26.3% by the end of the trial, based on a food frequency questionnaire (or an estimated 24% adjusted for the inaccuracies of a food frequency questionnaire versus a 4-day diet record). During 1989, 1 year on average after the WHT ended, 448 intervention women and 457 control women (87% of eligibles) completed a follow-up survey to determine the degree of maintenance of the diet. The intervention women maintained the low-fat diet with an increase of only 1.4 percentage points of energy from fat, despite the fact that they had attended no further intervention sessions and had made no commitment to maintain the diet beyond the end of the WHT. Furthermore, the degree of maintenance of the low-fat diet was not dependent on the length of time in the intervention, which suggests that intervention led to a sustained change in eating habits after as little as 5-9 months (8-13 classes).(ABSTRACT TRUNCATED AT 250 WORDS)

Development and validation of a food use checklist for evaluation of community nutrition interventions.

We developed a new dietary assessment instrument, the Food Behavior Checklist (FBC), which measures food use related to adopting lower-fat and higher-fiber diets. The FBC is a simplification of the 24-hour diet recall that consists of 19 simple yes/no questions about foods consumed during the previous day. To develop the FBC, an expert committee generated a list of foods based on our intervention program, and we used focus groups and random-digit dialing pretests to refine the format and clarify items. To validate the FBC, we compared responses of 96 women on the FBC to information collected during a professionally administered 24-hour diet recall. For most items, agreement between the FBC and 24-hour recall, based on the kappa statistic, was good to excellent. Agreement was poor for items requiring detailed knowledge about food composition (e.g. high-fiber cereal). There was a trend to over-report general food categories (e.g. luncheon meats) but not specific food items (e.g. ice cream). In an embedded randomized study, we found that a set of introductory items designed to serve as a memory retrieval cue did not improve agreement between the FBC and 24-hour recall. These data provide preliminary evidence supporting the use of short checklist questionnaires on the previous day's food use as a means to assess diet at the group or community level. This relatively inexpensive and rapid measure can be used to inform the design of public health nutrition programs and as an evaluation tool in intervention research.

Patterns of dietary behavior associated with selecting diets low in fat: reliability and validity of a behavioral approach to dietary assessment.

This report describes the development and evaluation of a behavioral measure of dietary patterns related to selecting low-fat diets. An 18-item questionnaire, based on an anthropological theory of dietary change, was developed to assess four relevant dimensions of dietary behavior: (a) excluding high-fat ingredients and preparation techniques, (b) modifying high-fat foods, (c) substituting specially manufactured low-fat foods for their higher-fat counterparts, and (d) replacing high-fat foods with low-fat alternatives. In this study, 99 women completed the diet behavior questionnaire twice and, to characterize precisely their dietary fat intake, also completed two 4-day food records and a food frequency questionnaire. Participants were aged 45 to 59 years and were selected to have a wide range of fat intakes (19.4% to 49.4% of calories from fat). Confirmatory factor analysis identified five scales that corresponded to those hypothesized, except for exclusion, which split into avoiding meat and avoiding fat as a seasoning. The scales had high test-retest and internal consistency reliabilities, and correlations with percent of calories from fat ranged from 0.34 to 0.57 (p less than .01). The correlation of the sum of the five scales (18 items) with percent of calories from fat was 0.68 (p less than .001) and, in multiple regression models, the multiple R2 using all factors to predict percent of calories from fat was 0.47. Overall, these findings supported the validity of the theoretical model of dietary patterns related to selecting diets low in fat. We conclude that a standardized, behavioral approach to measuring fat-related dietary behavior may be useful for designing and evaluating nutrition intervention programs.

Rapid assessment of dietary intake of fat, fiber, and saturated fat: validity of an instrument suitable for community intervention research and nutritional surveillance.

Abstract Traditional methods to assess changes in dietary intake, for example food frequency questionnaires or 24-hour dietary recalls, are often not practical: they are lengthy, expensive, and unsuitable for telephone administration. This article describes the development and evaluation of an approach to the rapid assessment of the dietary intake of nutrients of most interest in health promotion research; total fat, saturated fat, dietary fiber, and percent of calories from fat. In this validation study on 97 women, short dietary questionnaires were compared to two criterion measures of usual dietary intake, a food frequency questionnaire and the mean of two four-day diet records. Correlations between self-administered short questionnaires and four-day diet records were 0.52, 0.53, 0.61, and 0.40 for total fat, percent of calories from fat, saturated fat, and dietary fiber, respectively. These correlations are similar to those observed between food frequency questionnaires and four-day diet records, which suggests that this approach to developing and administering short dietary questionnaires may be useful in situations where more expensive and time-consuming methods of dietary assessment are not practical.