To identify correlates of 6th and 7th graders' (age 10–12 years) fruit and vegetable intake, to investigate parent–child correlations of fruit and vegetable intake, and to compare parents' and children's reports of children's accessibility, skills and preferences with respect to fruit and vegetables.

The results presented are based on the baseline survey of the ‘Fruits and Vegetables Make the Marks Project’, where 38 schools participated.

Fruit and vegetable intake was measured by food frequency questions. Theoretical factors, based on Social Cognitive Theory, potentially correlated to intake were measured, including behavioural skills, accessibility, modelling, intention, preferences, self-efficacy and awareness of 5-a-day recommendations.

In total, 1950 (participation rate 85%) 6th and 7th graders and 1647 of their parents participated.

Overall, 34% of the variance in the pupils' reported fruit and vegetable intake was explained by the measured factors. The strongest correlates to fruit and vegetable intake were preferences and accessibility. The correlation between the children's and their parents' fruit and vegetable intake was 0.23. The parents perceived their children's accessibility to be better than what was reported by the children (P<0.01), while the children reported their skills to be better than what was perceived by their parents (P<0.01).

The results from this study clearly point to a need for nutrition interventions aimed at parents. An important next step will be to investigate whether the identified correlates predict future fruit and vegetable intake, and whether they mediate any changes in intake in an intervention study.

To examine ethnic differences in body mass index (BMI), food habits and physical activity, and determine the factors contributing to differences in BMI.

In 2000–2001, 7343 (response rate 88%) 15- and 16-year-old students, enrolled in lower secondary schools in Oslo, participated in the cross-sectional Oslo Health Study. Of these participants, 1719 were defined as ethnic minorities.

Significant gender and ethnic differences in mean BMI were observed. Of the ethnic minority adolescents, 5.8% were underweight (<5th percentile of the US Centers for Disease Control and Prevention (CDC)/National Center for Health Statistics (NCHS) reference distribution) and 9.1% were overweight (>85th percentile of the US CDC/NCHS reference distribution). BMI was not significantly associated with either socio-economic factors or physical activity. Food habits and physical activity differed with ethnicity but not with socio-economic factors. An ordinal regression showed that girls from East Asia (odds ratio (OR) 0.4) and boys from sub-Saharan Africa (OR 0.4) had lower BMI than the Western group. Among girls, higher BMI was associated with less frequent consumption of chocolates and sweets, full-fat milk and breakfast (OR 2.4, 1.7 and 1.7, respectively). Higher BMI, for both boys and girls, was associated with current and past dieting (OR 3.7 and 4.2, respectively).

Adolescent food habits and physical activity varied by gender and ethnicity but not with socio-economic factors. BMI was associated with ethnicity, gender and food habits, but no significant relationship was observed with socio-economic factors or physical activity. Ethnicity, in addition to gender, should be taken into consideration when studying BMI and associated factors among adolescents.

To determine the correlates of a high-fat diet in urban, suburban and rural areas of Quebec, Canada.

A secondary analysis of data collected as part of a 5-year multi-factorial, multi-setting, community-intervention project.

Urban, suburban and rural settings of the province of Quebec, 1997.

Data were analysed from a sample of 5214 participants (2227 males, 2987 females). A food-frequency questionnaire was completed and a global index of food quality was calculated. Logistic regression was used to identify correlates of a diet high in total fats, saturated fat and cholesterol.

In both genders, lower level of education, smoking status, French and English languages compared with other languages spoken at home, and a rural environment were associated with poor diet quality. Having no intention to eat low-fat dairy products more often was associated with a high-fat diet. In men, obesity (body mass index >30 kg m−2) and absence of reported health problems were correlates of a high-fat diet, while, in women, lower physical activity was a correlate.

Future health interventions in Quebec should target people with low education, smokers and those living in a rural environment. Obese men and sedentary women should have access to specific dietetic resources.

To identify and characterise dietary patterns in a middle-aged Irish population sample and study associations between these patterns, sociodemographic and anthropometric variables and major risk factors for cardiovascular disease.

A cross-sectional study.

A group of 1473 men and women were sampled from 17 general practice lists in the South of Ireland. A total of 1018 attended for screening, with a response rate of 69%. Participants completed a detailed health and lifestyle questionnaire and provided a fasting blood sample for glucose, lipids and homocysteine. Dietary intake was assessed using a standard food-frequency questionnaire adapted for use in the Irish population. The food-frequency questionnaire was a modification of that used in the UK arm of the European Prospective Investigation into Cancer study, which was based on that used in the US Nurses' Health Study. Dietary patterns were assessed primarily by K-means cluster analysis, following initial principal components analysis to identify the seeds.

Three dietary patterns were identified. These clusters corresponded to a traditional Irish diet, a prudent diet and a diet characterised by high consumption of alcoholic drinks and convenience foods. Cluster 1 (Traditional Diet) had the highest intakes of saturated fat (SFA), monounsaturated fat (MUFA) and percentage of total energy from fat, and the lowest polyunsaturated fat (PUFA) intake and ratio of polyunsaturated to saturated fat (P:S). Cluster 2 (Prudent Diet) was characterised by significantly higher intakes of fibre, PUFA, P:S ratio and antioxidant vitamins (vitamins C and E), and lower intakes of total fat, MUFA, SFA and cholesterol. Cluster 3 (Alcohol & Convenience Foods) had the highest intakes of alcohol, protein, cholesterol, vitamin B12, vitamin B6, folate, iron, phosphorus, selenium and zinc, and the lowest intakes of PUFA, vitamin A and antioxidant vitamins (vitamins C and E). There were significant differences between clusters in gender distribution, smoking status, physical activity, body mass index, waist circumference and serum homocysteine concentrations.

In this general population sample, cluster analysis methods yielded two major dietary patterns: prudent and traditional. The prudent dietary pattern is associated with other health-seeking behaviours. Study of dietary patterns will help elucidate links between diet and disease and contribute to the development of healthy eating guidelines for health promotion.

To examine differences in food habits among Native Americans with and without diabetes.

A cross-sectional epidemiological study in which participants underwent a physical examination and answered an extensive interviewer-administered questionnaire to assess differences in food servings, preparation and eating habits.

Participants aged ≥25 years were randomly selected from three reservations in Minnesota and Wisconsin. There were 990 persons without diabetes, 294 with a prior diagnosis of diabetes, and 80 with fasting glucose >125 mg dl−1 but no prior diabetes diagnosis.

Persons with prior diabetes diagnosis were less likely than those without diabetes to report eating fast-food meals two or more times per week, eat visible fat on meat or the skin on poultry, eat fried chicken or fried fish, to add fat to cooked vegetables and drink whole milk. Persons with previously undiagnosed diabetes were more likely than previously diagnosed persons to report eating fast-food meals two or more times per week, eat visible fat on meat and the skin on poultry, drink whole milk and eat fried fish, but were less likely to drink low-fat milk. Previously undiagnosed persons were more likely than either diagnosed persons or persons without diabetes to consume lard from cooked foods and use it when cooking.

Persons with diagnosed diabetes showed healthier eating patterns than those without diabetes, while undiagnosed persons showed some less favourable patterns. Because virtually all persons with diabetes in these communities receive nutrition education, the results suggest that nutrition education programmes for diabetics may be associated with healthier eating patterns.

To examine statistical models that have been used to predict the cessation of breast–feeding.

In nutritional epidemiology, a knowledge of risk factors that lead to breast-feeding cessation is essential to promote optimal infant health by increasing or sustaining breast–feeding rates. However, a number of methodological issues complicate the measurement of such risk factors. It is important when building multivariate models that variables entered into the model are not intervening variables, factors on the causal pathway or surrogate outcomes. Inclusion of these types of variable can lead to inaccurate models and biased results. A factor often cited to predict breast–feeding is ‘intention to breast–feed’ prior to the birth of the infant, although this factor is directly on the causal decision–making pathway. Another factor often cited is the age of introduction of formula feeding, which is actually part of the outcome variable because formula feeding defines the difference between full, complementary and no breast-feeding. Rather than include these as risk factors in multivariate models, factors removed from the causal pathway such as influences of educational practices, including advice to complementary feed, and beliefs and attitudes of families and health-care practitioners should be measured.

The accurate quantification of modifiable risk factors is essential for designing public health education campaigns that are effective in sustaining or increasing breast–feeding duration.

To investigate feeding practices in infants under the age of 4 months in Liverpool, England with particular reference to the cost of infant feeding.

A cross–sectional survey consisting of self–completion questionnaires and interviews.

Subjects' homes within Central and South Liverpool Primary Care Trust areas.

One hundred and forty–nine women (aged 18 to 43 years) and their infants (mean age 13 weeks).

The average weekly cost of breast–feeding was £11.58 compared with £9.60 for formula–feeding. Many breast– and formula–feeding women spent money however on items that were not needed or used only once or twice. This was especially true of first–time mothers. Characteristics significantly associated with higher spending were: feeding method – mothers that had or were partially breast–feeding (P=0.001), education – those educated to degree level (P=0.028), socio–economic status – those in social classes I and II (P=0.002) and age – those aged 30 years and over (P=0.003).

This study demonstrates that while breast–feeding is often promoted as being free, this is not the case. Better information needs to be given to parents to avoid wasting money on items that are unnecessary, or where cheaper alternatives are available.

In sub-Saharan Africa, underweight and micronutrient deficiencies account for an estimated 25% of the burden of disease. As the coverage of national health systems expands, increased opportunities exist to address the needs of children and women, the most vulnerable to these deficiencies, through high-quality nutrition services.

To assess health providers' knowledge and practice with regard to essential nutrition services for women and children in Burkina Faso, Mozambique and Niger, in order to assist the development of a standard guide and tools to assess and monitor the quality of the nutrition services delivered through national health systems.

The three surveys reveal the extent of missed opportunities to deliver nutrition services during routine prenatal, postnatal and child-care consultations for the prevention and treatment of highly prevalent nutritional deficiencies.

A commitment to improving the quality of facility-based nutrition services is necessary to impact on the health outcomes of women and children ‘covered’ by national health systems. Rigorous assessment and monitoring of the quality of nutrition services should inform health programme and policy development. Building on the lessons learned in these three assessments, Helen Keller International has developed a standard Guide and Tools to assess the quality of the nutrition services delivered through national health systems. These tools can be adapted to assess ongoing nutrition services in health facilities, provide a framework for nutrition programming, inform the development of pre-service as well as in-service nutrition training curricula for providers, and evaluate the impact of nutrition training on providers' practices.

To compare the relationships between food (nutrient) intakes and biochemical markers of nutritional status of asymptomatic HIV-infected with HIV-uninfected subjects, to gain more information on the appropriate diet for HIV-infected persons at an early stage of infection.

Cross-sectional population-based survey.

North West Province, South Africa.

Two hundred and sixteen asymptomatic HIV-infected and 1550 HIV-uninfected men and women volunteers aged 15 years and older, recruited as ‘apparently healthy’ subjects from 37 randomly selected sites.

Food and nutrient intakes, measured with a validated food-frequency questionnaire, and nutritional status indicated by anthropometric and biochemical variables, measured by a standardised methodology.

The prevalence of HIV infection in the study population was 11.9%. The anthropometric indices and nutrient intakes of HIV-infected and uninfected subjects did not differ significantly, indicating that these 216 HIV-infected subjects were at an early stage of infection. Of the biochemical nutritional status variables, high-density lipoprotein cholesterol and total cholesterol, haemoglobin, albumin and triglycerides were significantly lower in infected subjects. They also had higher globulin and liver enzyme levels than uninfected subjects. In infected subjects, serum albumin correlated significantly with serum lipids, serum vitamin A, serum vitamin E, serum iron, total iron-binding capacity and haemoglobin. The significant positive correlations of the liver enzymes with serum lipids, albumin, vitamin A and iron, observed in HIV-uninfected subjects, disappeared in the infected subjects. Polyunsaturated fat intake showed significant positive correlations with the increased liver enzymes in infected subjects. A principal components analysis indicated that, in infected subjects, increased liver enzymes correlated with higher consumption of maize meal and lower consumption of meat and vegetables.

This survey indicated that asymptomatic HIV-infected subjects who followed a diet rich in animal foods had smaller decreases in serum albumin, haemoglobin and lipid variables, and smaller increases in liver enzymes, than those who consumed a diet based on staple foods. This suggests that animal foods are associated with improved nutritional status in HIV-infected persons. These results should be confirmed with intervention studies before dietary recommendations for asymptomatic HIV-infected individuals can be made.

To study the prevalence of anaemia and its association with measures of iron deficiency (ID) among a group of pregnant women.

Cross-sectional survey.

Pregnant women identified through house-to-house visits and participating in community-based antenatal care activities in a rural location of Mymensingh, Bangladesh.

The estimates are based on 214 reportedly healthy pregnant women in their second trimester. Information on socio-economic status and reproductive history were obtained through home visits and venous blood samples were collected at antenatal care centres. Haemoglobin concentration (Hb) was measured by HemoCue, serum ferritin (sFt) by radioimmunoassay and serum transferrin receptor (sTfR) by enzyme-linked immunosorbent assay methods. ID was defined as presence of either low sFt (<12 μg l−1) or high sTfR (>8.5 mg l−1).

The prevalence of anaemia (Hb <110 g l−1) was 50%, but severe anaemia (Hb >70 g l−1) was absent. Low sFt was observed in 42%, high sTfR in 25%, either low sFt or high TfR in 54% and both low sFt and high TfR in 13% of the pregnant women. Two out of three anaemic women had an indication of ID, which was present in 80% of women with moderate (Hb 70–99 g l−1) and 50% with mild (Hb 100–109 g l−1) anaemia. Four out of 10 non-anaemic women (Hb >110 g l−1) also had ID, but the prevalence was significantly lower than that observed in anaemic women (P=0.001).

Despite the high prevalence of anaemia, severe cases were absent. The prevalence of ID increased at lower Hb. However, an increased prevalence was also found among women in the highest category of Hb.

To explore the effect of BRAC (formerly Bangladesh Rural Advancement Committee) initiatives towards promoting gender and social equity in health among children of poor mothers who are BRAC members.

A cohort of 576 children from the prospective study of a BRAC– International Centre for Diarrhoeal Disease Research, Bangladesh joint research project was analysed. Data were collected three times during 1995–1996 with approximately 4-month intervals. Stunting, defined as height-for-age below minus two standard deviations from the reference median, was the outcome health measure. The study children were stratified into three groups according to their mother's social and BRAC membership status: poor and BRAC member (BM), poor non-member (TG) and non-poor non-member (NTG).

Matlab, rural area of Bangladesh.

Children aged 6–72 months.

The overall prevalence of stunting was 76%; the highest prevalence was found among TG (84.6%) children and no significant difference was observed between BM and NTG children (67.3% and 69.4%, respectively). In all groups, a significantly larger proportion of girls was stunted compared with boys in the first round. Group-level analysis showed that stunting decreased among all children except BM boys at the end of third round, with the largest decline among BM girls. In contrast, stunting prevalence increased among BM boys. A similar trend was found in the individual-level analysis, where a larger proportion of BM girls recovered from stunting compared with other groups and no recovery was observed among BM boys. At the end of the third round, the nutritional status of BM girls was almost equal to that of the BM boys, while gender inequity remained large among TG and NTG children.

The BRAC initiative appeared to contribute to a significant equity gain in health for girls, as well as to decreased differences in ill health between the poor and the non-poor.

Recent research on access to food among low-income populations in industrialised countries has begun to focus on neighbourhood food availability as a key determinant of dietary behaviour. This study examined the relationship between various measures of food store access and household fruit and vegetable use among participants in the Food Stamp Program, America's largest domestic food assistance programme.

A secondary data analysis was conducted using the 1996–97 National Food Stamp Program Survey. The survey employed a 1-week food inventory method, including two at-home interviews, to determine household food use. Separate linear regression models were developed to analyse fruit and vegetable use. Independent variables included distance to store, travel time to store, ownership of a car and difficulty of supermarket access. All models controlled for a full set of socio-economic variables.

A nationally representative sample of participants (n = 963) in the Food Stamp Program.

After controlling for confounding variables, easy access to supermarket shopping was associated with increased household use of fruits (84 grams per adult equivalent per day; 95% confidence interval 5, 162). Distance from home to food store was inversely associated with fruit use by households. Similar patterns were seen with vegetable use, though associations were not significant.

Environmental factors are importantly related to dietary choice in a nationally representative sample of low-income households, reinforcing the importance of including such factors in interventions that seek to effect dietary improvements.

To examine associations of the frequency of eating at fast-food restaurants with demographic, behavioural and psychosocial factors and dietary intake in African American adults.

Self-reported data from a population-based cross-sectional survey of 658 African Americans, aged 20–70 years, in North Carolina. An 11-page questionnaire assessed eating at fast-food restaurants, demographic, behavioural and diet-related psychosocial factors, and dietary intake (fruit, vegetable, total fat and saturated fat intakes, and fat-related dietary behaviours).

The participants were aged 43.9±11.6 years (mean±standard deviation), 41% were male, 37% were college graduates and 75% were overweight or obese. Seventy-six per cent reported eating at fast-food restaurants during the previous 3 months: 4% usually, 22% often and 50% sometimes. Frequency of eating at fast-food restaurants was positively associated with total fat and saturated fat intakes and fat-related dietary behaviours (P < 0.0001) and inversely associated with vegetable intake (P < 0.05). For example, mean daily fat intake was 39.0 g for usually/often respondents and 28.3 g for those reporting rare/never eating at fast-food restaurants. Participants who reported usual/often eating at fast-food restaurants were younger, never married, obese, physically inactive and multivitamin non-users (all P < 0.01). Frequency of eating at fast-food restaurants was positively associated with fair/poor self-rated health, weak belief in a diet–cancer relationship, low self-efficacy for healthy eating, weight dissatisfaction, and perceived difficulties of preparing healthy meals and ordering healthy foods in restaurants (all P < 0.05). Frequency of eating at fast-food restaurants did not differ significantly by sex, education, smoking, ability to purchase healthy foods or knowledge of the Food Guide Pyramid.

Eating at fast-food restaurants is associated with higher fat and lower vegetable intakes in African Americans. Interventions to reduce fast-food consumption and obesity in African Americans should consider demographic and behavioural characteristics and address attitudes about diet–disease relationships and convenience barriers to healthy eating.

We describe the methods used to develop and score a 17-item ‘screener’ designed to estimate intake of fruit and vegetables, percentage energy from fat and fibre. The ability of this screener and a food-frequency questionnaire (FFQ) to measure these exposures is evaluated.

Using US national food consumption data, stepwise multiple regression was used to identify the foods to be included on the instrument; multiple regression analysis was used to develop scoring algorithms. The performance of the screener was evaluated in three different studies. Estimates of intakes measured by the screener and the FFQ were compared with true usual intake based on a measurement error model.

US adult population.

For development of instrument, n = 9323 adults. For testing of instrument, adult men and women in three studies completing multiple 24-hour dietary recalls, FFQ and screeners, n = 484, 462 and 416, respectively.

Median recalled intakes for examined exposures were generally estimated closely by the screener. In the various validation studies, the correlations between screener estimates and estimated true intake were 0.5–0.8. In general, the performances of the screener and the full FFQ were similar; estimates of attenuation were lower for screeners than for full FFQs.

When coupled with appropriate reference data, the screener approach described may yield useful estimates of intake, for both surveillance and epidemiological purposes.