![[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}
|
|
|
|
|
||
Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NC 27516-3997
3To whom correspondence should be addressed. E-mail: dore{at}email.unc.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • children • diet trends • income • behavior modification
Throughout the 20th century, the Russian population faced many social and economic stresses. These include the hardship during the Russian revolution, the national restructuring of the socialist era and the economic, political and social instability during the shift to a free market system in 1992. The years following the 1992 reforms can be best characterized as a period of emerging income inequality and economic crises (1,2). The most prominent crisis was the unexpected collapse of the Russian economy in 1998. The annual inflation rate in June of 1998 was 
10%, but by December of 1998 it had reached 150% (3). By December of 1999, the annual inflation rate had dropped back to 25%. During this short-lived crisis, not only did the Russian ruble plummet in value but unemployment increased dramatically.
Political influences and macro-level economic stresses have greatly affected the Russian diet. Beginning in the 1960s, Russian Communist leaders pushed for alterations in consumption patterns to more closely resemble "the Western diet of progress." By 1989, intakes of red meat and milk increased dramatically, whereas intakes of potatoes and bread decreased (4,5). The focus placed on providing affordable, energy-dense "rich" foods established a diet that was high in overall fat and saturated fat and was consumed for >20 y by individuals at all age and income levels. By 1992, fat constituted 40% of energy intake for Russian adults and 37% for children (6,7). The 1992 reforms, however, strongly affected the economic stability of the Russian household, which in turn affected food availability. Due to the removal of government subsidies, maintenance of the accustomed diet became a financial challenge for many Russian households, leading to a conflict between preference and affordability.
Preferences and dietary patterns are thought to be highly resistant to change. Consequently, populations facing major economic stress may compensate by making food-related behavioral modifications that conserve the structure of the diet. To date, there is little information in any country concerning changes in household dietary patterns in response to economic constraints. This study attempted to identify diet and diet-related changes that might have buffered the effect of economic crisis on the usual diet of Russian children. We used nationally representative multiple cross sections of data collected before, during and after the economic collapse of 1998. We first compared trends over time in the consumption of specific food groups among children of high vs. low income households. We then identified trends in food-related behaviors including home food preparation, and changes in cost per unit energy. We were particularly interested in identifying behavioral modifications that families utilized to buffer the decrease in purchasing power associated with the 1998 crisis.
The diet of children was the focus of this analysis for several important reasons. Eating habits are established primarily in childhood; therefore, dietary fluctuations during this period are likely to have a lifelong nutritional and health effect (8,9). Given the impressionable nature of children, establishing healthy eating habits at an early age is more plausible than attempting to influence dietary patterns of adults.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The Russian Longitudinal Monitoring survey (RLMS)3 began in 1992 to assess the effect of political and economic reforms on the healthcare, sociodemographic and behavioral characteristics of households and individuals. To ensure national representation of the households, survey researchers used a probability proportionate to size (PPS) procedure to select primary sampling units (PSU) from the eight regions of Russia. Included in these PSU were the three metropolitan areas of St. Petersburg, Moscow and Moscow Oblast. These areas were selected because they included a substantial percentage of the Russian population. From each PSU, districts were selected using PPS; then households were randomly selected from each district. The surveys were conducted so that interviewers returned to the same individual dwelling units as those visited in previous rounds regardless of any changes in occupancy. Consequently, families that remained in the same dwelling unit over time were repeatedly interviewed, resulting in cross sections that had an overlap of individuals. Subjects living in the same dwelling unit and sharing the same household income were included as members in a household for each survey round. For more details see Kosolapov et al. (10).
To date, the RLMS has collected 11 rounds of data in two phases. This study focuses on rounds preceding (1994, 1996), during (1998) and after (2000) the recent economic collapse. The cross-sectional study samples included all individuals ages 6–10 y living in surveyed households (n = 932, 803, 688, 552 from each round, respectively). The distribution of ages among the different samples stratified by income level was comparable. The RLMS protocols were reviewed and approved by the Institutional Review Board of the University of North Carolina at Chapel Hill and the Russian Academy of Sciences.
Dietary data.
Trained interviewers completed one 24 h-recall for all members (where possible) of each household selected in rounds 5 (1994), 8 (1998) and 9 (2000). In contrast, round 7 (1996) included two nonconsecutive days of data in an attempt to collect information for both a weekday and weekend day. In this study, we analyzed only d 1 of round 7 dietary data to ensure comparability with the other rounds included in the analysis. For all rounds color photos of foods were used to assist in the identification of portion sizes. In all four rounds, a proxy informant was interviewed to collect dietary information for children between the ages of 6 and 13 y. Additionally trained interviewers questioned teachers and cafeteria workers (assisted by color photos of foods) to verify recalled foods consumed during school hours.
To analyze the Russian diet, a team of nutritionists developed a nutrient composition table, which was modified for each round to include newly introduced foods. Nutrient values for foods found in the USDA and West German food composition tables provided a basis for values estimated in the Russian nutrient database. The USDA table was chosen because it has the most extensive list of foods and the German table because it represents European foods similar to what may be found in Russia. Because major regional differences in food composition tend to be in micro- rather than macronutrients, the conglomerate of these two tables was considered adequate for an evaluation of food group intake patterns. Composite foods were broken down and recorded as raw ingredient lists. Mean daily nutrient intakes were then calculated on the basis of the intake of the uncooked and unprocessed ingredients as well as other noncomposite foods.
For this study, food intake patterns were summarized using the following food groups: meat and poultry, grains and starches, dairy products and eggs, and fruits and vegetables. Like meats and poultry, dairy products and eggs are both high quality sources of protein but they were evaluated as a separate group because they are generally lower in cost. Consumption of these products may be less sensitive to economic insecurity and thus may provide stable sources of protein during periods of crisis. Among Russian children, the daily gram intakes per capita and per consumer as well as the percentage of consumers of each food group were calculated. Daily gram intake per capita was calculated as the mean intake in the entire child sample, without accounting for the fact that not all individuals consume a particular food group. Intake per consumer reflects only the amount consumed by those who eat foods from a food group. This distinction is important because when analyzing time trends, average per capita intake might reflect either an increase in the number of individuals who consume particular foods or an increase in the amount consumed by those who eat these foods. Finally, total energy intake was evaluated as relative energy in kJ/kg weight for each child to adjust for body size.
Surveys also included information on the location of preparation and consumption of individual foods. In this study, we were interested in the mean percentage of food that was prepared in the home. Unlike in the United States, this category extends beyond home-cooked meals to include items such as butter, cheese, breads and pastas. Home preparation may be important as an indicator of behavior modifications utilized in the face of periods of unstable economic conditions.
Income and expenditure data.
Measures of household income and household poverty were used to create a poverty index, which in turn was used as an indicator of the economic well-being of individual children. To calculate household income, the sum of the incomes from individual members was added to the monetary value from any sales of harvested produce and/or livestock. Individual income included values for goods, services and money received from a primary or additional place of work; child support; a stipend or pension; unemployment benefits; alimony; insurance payments; investments; sale or rent of personal property; and government subsidies for fuel. Per capita and household incomes were deflated to the selected baseline year 1992. Household poverty was measured as the sum of all individual poverty lines within the household. A Russian poverty line was calculated as the average cost of food items in a Russian food basket adjusted for regional price variations, household size and nutrient requirements for individuals within the household. To create the index relative household income was divided by the household poverty level. The continuous poverty index variable was divided into thirds to identify low and high income families. Mean values adjusted for urban/rural location, age and gender were substituted for those individuals missing income data.
A summary measure of the relative cost of household food purchases, referred to as cost per kilojoule, was constructed as total household ruble food expenditure per household kilojoule intake adjusted for household size [expenditure/(intake · household size)]. This variable allowed us to assess whether families bought cheaper versions of food to enable the purchase of more kJ/ruble before, during and/or after the economic crisis of 1998.
Anthropometry.
Age, weight and height were assessed by trained interviewers at the home of the subjects. For both height and weight, subjects were measured wearing light clothing and no shoes.
Analysis.
Preliminary descriptive data included mean estimates with SD for the age, height (cm) and weight (kg) of the children. Dietary data including the mean percentage of dietary fat and protein were also calculated. The main outcomes of interest, estimated for each year, were mean per capita and per consumer gram intakes for each food group, the percentage of people consuming these foods, the mean relative energy intake (kJ/kg), the percentage of foods produced at home and the cost per kilojoule. Only the high and low income thirds were included in the tables and figures to focus on the contrast between a low income group (potentially highly sensitive to economic shocks) and a high income group consisting of individuals with a substantially greater economic status and stability.
Trends were assessed comparing mean values across the 4 y stratified by household income level. The datasets used for this analysis provided neither a pure longitudinal panel nor independent samples; therefore a P for trend analysis using regression methods was not possible (11). Because the RLMS comprises survey data, an analysis would no longer be nationally representative if we were to use a subset of individuals (those children followed in all four time points). However, due to the age range selected for this study, the 1994 and 2000 surveys are independent samples. A trend was therefore considered significant if there was a monotonic trend in at least three consecutive time points and a two-way t test between the 1994 and 2000 values was significant. There are limitations, however, with these testing criteria because the change of most interest, in reference to coping mechanisms, is 1994 to 1998 which cannot be evaluated formally due to the sample overlap. As a result, some important behavior modifications may not have been captured by this analysis of time trends. All data files were created using SAS System (version 8.2, SAS Institute, Cary, NC) and weighted means and SD were calculated using the SVY commands in STATA (version 7.0, College Station, TX) (12,13).
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The mean age fluctuated between 8.4 and 8.7 y for all time points for both low and high income children (Table 1). In 1994, high income children were heavier (1.4 kg: P < 0.05) and in 1998 high income children were both heavier (1.3 kg: P < 0.05) and taller (2.3 cm: P < 0.05) than low income children. Although significant, these weight differences may be biologically minor. Household size between the two income groups differed only in 1994 (3.1 for low income households vs. 2.6 in high income households: P < 0.01). The trend in the percentage of protein and fat intake in the diet did not differ notably across survey rounds for either low or high income children. A significant difference in protein intake between low and high income children existed in 2000 (11.1 vs 11.7%: P < 0.05) although this 0.6% difference may not be biologically important. The percentage of fat differed significantly between low and high income children in all rounds except for 2000. The average percentage of fat intake for low income children from 1994 to 1998 was 28.8 vs. 31.7 for high income children (P1994 < 0.0001, P1996 < 0.01, P1998 < 0.00001).
|
Meat and poultry per capita gram intake was higher for children in the high income families in 1996 (P < 0.01), 1998 (P < 0.0001), and 2000 (P < 0.01) (Fig. 1a). A decreasing time trend, from 88 to 68 g/d, was observed for the low income children (P < 0.01). Gram intake over time was suggestive of an increasing trend for high income children but was not significant.
|
|
175 g/d from 1994 to 2000 for children in the low income group. Children in the high income group started with intakes similar to the low income group at baseline but increased their per capita intake over time to 216 g by 2000 (P < 0.01). An increasing time trend was also observed in the per consumer gram intake for high income children (P < 0.05), whereas the estimated percentage per consumer remained relatively constant (Table 2). This consumer intake time trend suggests that the increasing per capita trend observed was due to an increased intake by individuals already consuming these foods rather than an increase in the number of consumers. Whole milk, eggs and sour cream were the most commonly consumed. Similar to the intakes of dairy products and eggs, the low income mean per capita intake of grains and starches remained relatively constant from 1994 to 2000 (Fig. 1c). The overall per capita gram intake for high income children was suggestive of an increasing trend but was not significant over time (from a low of 333 to 385 g/d). There were no significant differences in per capita gram intake between the low and high income children for any of the four time points. No significant time trends were observed for either income group in per consumer intake (Table 2). For both income groups, the most common products included bread, potatoes, flour, pastas, and baked goods.
The per capita consumption of fruits and vegetables was unique in that the trends were parallel for the low income and high income children (Fig. 1d). An alternating increasing and decreasing trend was observed for both groups. Per capita consumption in the high income group on average ranged from 99 to 135 g/d more than consumption in the low income group and differed for all four years (P1994 < 0.0001, P1996 < 0.0001, P1998 < 0.01, P2000 < 0.0001). These patterns in intake arose from changes in per consumer intake rather than a change in the number of consumers for both income groups (Table 2). The largest decrease in per consumer intake for both income groups occurred between the 1996 and 1998 survey years (50 g for the low income children and 65 g for the high income children). Time trends could not be evaluated formally because the criterion for at least three time points following a monotonic trend was not observed. The most common foods in this group included: onions, root crops (i.e., beets and carrots), canned vegetables and cabbage. The WHO recommendation of 400 g/d of fruits and vegetables was not met by either income group in any of the years observed (15).
Overall energy intake followed trends similar to those observed in the food group intake data (Fig. 2). Although total energy intakes for low income children remained below the WHO recommendation of 293 kJ/kg for this age group (251.0–259.4 kJ/kg), intake remained stable even during the 1998 economic crisis except for a slight increase in the 1998 CI (16). In contrast, the intake for high income children increased from 251.0 kJ/kg in 1996 to 297.1 kJ/kg by 2000 (P < 0.01). High income children consumed more energy in both 1998 (P < 0.05) and 2000 (P < 0.0001) than low income children. According to these data, by 2000, the average energy intake for high but not low income children exceeded the WHO recommendation for total energy intake. Because the age distribution between cross-sectional samples is proportionate, the figures reflect time trends in energy intake rather than age-related change in appetite and preferences.
|
As was shown in the dietary trends, energy intake of low income children changed very little over the 6-y observation period, although there was a change in diet composition. The effect of the crisis may have been mitigated to some extent by various behavioral modifications. The cost per kilojoule (ruble/kJ) decreased significantly for both the low income (0.06–0.03 ruble/kJ) and high income groups (0.09–0.05 ruble/kJ) (Fig. 3). However, at all time points the ruble/kJ value for low income households was less than half the ruble/kJ value for high income households (P < 0.0001 for all four time points) suggesting that low income families were purchasing much less expensive foods. Equally, as national economic stability decreased from 1994 to 1998, the percentage of food that was home-prepared increased more notably in the low income group (from 46 to 53%) than in the high income group (42.5 to 45.5%) (Fig. 4). By 2000, after the crisis, the percentage of foods prepared at home returned to precrisis levels. Because the 1994 and 1998 samples are not independent samples, we cannot make a formal statistical evaluation of the change in home-food preparation levels.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Children from high income families showed a recent significantly increasing trend in total energy intake to levels that exceeded WHO recommendations for many individuals (17). This increasing trend reflects a selective increase in high vs. low energy-dense foods (high fat dairy products and eggs vs. fruits and vegetables). The gram intake of fruits and vegetables remained consistently below the WHO recommended level, while the proportion of the diet from fat consistently exceeded WHO recommended levels of 
30% of the diet from fat (16). Contrary to expectation, cost per kilojoule decreased significantly over time for high income families. A recent study, however, found that prices for select staple foods such as breads, flours, pastas and grains during this time may also have been decreasing, which may have affected this observed trend for both income groups (Stillman and Thomas, unpublished data). As was expected, the 1998 economic crisis did not cause a reduction in intake by high income children; in fact, intake levels increased despite the crisis.
Health and policy implications.
We observed one group of Russian children maintaining their dietary intake over time, albeit deficient for some children, and one group increasing their intake over time. These trends may affect existing patterns of weight change and existing under- and overweight prevalences in Russia. Recent studies found evidence of stunting in the early 1990s and an increase in underweight prevalences in Russian children from 1992 to 1998, which may be a result of patterns of energy inadequacy evident from our study (18,19). Conversely, 35% of Russians ages 30–59 y are currently overweight and 21% are obese (the percentages are even higher among the elderly) (5,20). Considering the increasing information on tracking of dietary patterns and overweight from childhood to adulthood, this raises concerns for the increasing energy intake among high income Russian children observed in our study (21–23). In validation of this concern, recent work by Stillman and Thomas found that as long-term income increased among Russians, there was an increase in energy intake, a change in diet composition (higher percentage of protein and fat) and an increase in BMI (Stillman and Thomas, unpublished data).
The coexistence of under- and overconsumption as well as under- and overweight individuals requires that government programs address both sets of problems simultaneously. This issue becomes increasingly complex in light of recent findings that among households with at least one underweight individual, 58% also had at least one overweight individual (24). To be effective, policy measures must address current dietary deficiencies while minimizing the trend toward diets that are excessive in total intake (specifically fatty animal products) at the individual, household and community levels. It is evident from the findings in this study that supplementing family income in Russia will not ensure the attainment of a healthy balanced diet. Policy measures should instead take advantage of existing household coping mechanisms to efficiently and effectively improve the Russian diet. We have determined that Russian families successfully maintained existing eating patterns by purchasing cheaper foods and increasing home preparation. Similar to other transitional Eastern European countries cost is a main determinant in the purchasing and preparation of foods in Russia (25,26). It seems plausible then that subsidizing lean meat cuts as well as specific fruits and vegetables might influence eating patterns. These subsidies would make accessible foods that are either in decline or are consistently insufficient in diets of low income children. At the same time, these subsidies might promote the consumption of fruits and vegetables as well as low fat high quality protein sources among high income children, resulting in a diet that is less energy dense. Preceding 1992, the practice of subsidizing foods such as meat, dairy products and even bread was widely used by the Russian government, suggesting a familiarity and earlier success with this form of intervention (18).
Future considerations.
We would like to stress that this analysis utilizes observational data and is preliminary in nature; it identifies overall trends in diet patterns and behaviors suggestive of possible food-related coping mechanisms. Future studies are necessary to evaluate a possible direct relationship between diet-related behaviors and diet patterns as well as to identify a possible causal pathway. It would also be useful to evaluate the effect of food price changes in Russia on the above relationship. A limitation of this study that should also be recognized is the use of population means, which may oversimplify the dietary patterns within income groups. There may be, for example, some individuals who consume far above or below the recommended level of energy intake even though the overall mean was consistently <50 kJ/kg below the WHO recommended level. Using the mean intake level as a guide for developing health policies might therefore improve the nutritional status of some low income children while promoting excess intake by others. It may be necessary to identify within this low income group those children who are at the highest risk of nutritional deficiencies to successfully address the problem while minimizing excessive intake. A recent paper by Lokshin and Popkin clarified that the low income Russian group was in fact a dynamic mix of transitional families (those who temporarily fall below the poverty line) and the ultra-poor (those individuals that consistently remain below the poverty line) (2). To design successful health interventions for Russia children, it may be necessary to extend this analysis in the future to carefully identify those individuals at highest risk of under- and overconsumption and minimize the level of misclassification of dietary intake.
|
FOOTNOTES |
|---|
2 Supported by National Institutes of Health (R01-HD30880 and R01-HD38700).
4 Abbreviations used: RMLS, Russian Longitudinal Monitoring Survey; PPS, probability proportionate to size; PSU, primary sampling unit.
Manuscript received 21 February 2003. Initial review completed 17 March 2003. Revision accepted 27 August 2003.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Cornia, G. A. (1994) Poverty, food consumption, and nutrition during the transition to the market economy in Eastern Europe. Am. Econ. Rev. 84:297-302.
2. Lokshin, M. & Popkin, B. M. (1999) The emerging underclass in the Russian Federation: income dynamics 1992–96. Econ. Dev. Cult. Change 47:803-829.
3. Mroz, T. A., Henderson, L., Bontch-Osmolovski, M. & Popkin, B. M. (2003) Monitoring Economic Conditions in the Russian Federation: The Russia Longitudinal Monitoring Survey 1992–2002. Report submitted to the U.S. Agency for International Development 2003 Carolina Population Center University of North Carolina at Chapel Hill, NC.
4. Popkin, B. M., Baturin, A., Kohlmeier, L. & Zohoori, N. (1997) Russia: monitoring nutritional change during the reform period. Wheelock, V. eds. Implementing Dietary Guidelines for Healthy Eating 1997:23-46 Chapman and Hall London, UK. .
5. Popkin, B. M., Zohoori, N., Kohlmeier, L., Baturin, A., Martinchik, A. & Deev, A. (1997) Nutritional risk factors in the former Soviet Union. Bobadilla, J. L. Costello, C. A. Mitchell, F. eds. Premature Death in the New Independent States 1997:314-334 National Academy Press Washington, DC. .
6. Jahns, L., Baturin, A. & Popkin, B. M. () Secular trends in diet and obesity by economic status in the Russian Federation. Eur. J. Clin. Nutr. 57:1295-1302.
7. Jahns, L. & Popkin, B. M. (2001) Prevalence of obesity in Russian children and adolescents: change in the new economy. Obes. Res. 9:60S.
8. Wang, Y., Bently, M. E., Zhai, F. & Popkin, B. M. (2002) Tracking of dietary intake patterns of Chinese from childhood to adolescence over a six-year follow up period. J. Nutr. 132:430-438.
9. Nicklas, T. A. (1995) Dietary studies of children and young adults (1973–1988): the Bogalusa Heart Study. Am. J. Med. Sci. 310:S101-S108.
10. Kosolapov, M. S., Swafford, M. S. & Heeringa, S. G. (2002) Design, Methods and Statistical Properties of the Russian Longitudinal Monitoring Survey (RLMS) 2002 Survey Research Center Ann Arbor, MI.
11. Draper, N. R. & Smith, H. (1981) Applied Regression Analysis 2nd ed. 1981:278-284 John Wiley & Sons New York, NY.
12. SAS Institute Inc. (1999) SAS User’s Guide, Version 8e 1999 SAS Institute Cary, NC.
13. STATA (2002) STATA Version 7.0 2002 College Station, TX.
14. World Health Organization (1996) Report of a Joint FAO/WHO Consultation. Preparation and use of food-based dietary guidelines, Nutrition Programme 1996 WHO Geneva, Switzerland.
15. Pomerleau, J., McKee, M., Robertson, A., Vaask, S., Pudule, I., Grinberga, D., Abaravicius, A. & Bartkeviciute, R. (2002) Food security in the Baltic Republics. Public Health Nutr. 5:397-404.[Medline]
16. Trahms, C. M. & Pipes, P. L. (1997) Nutrition in Infancy and Childhood 1997 WCB/McGraw Hill New York, NY.
17. World Health Organization (1985) Report of a Joint FAO-WHO-UNU Expert Consultation. Energy and protein requirements, WHO Technical Report Series no. 724 1985 WHO Geneva, 1985.
18. Mroz, T. & Popkin, B. M. (1995) Poverty and the economic transition in the Russian Federation. Econ. Dev. Cult. Change 44:1-31.
19. Wang, Y., Monteiro, C. & Popkin, B. M. (2002) Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am. J. Clin. Nutr. 75:971-977.
20. Zohoori, N., Gleiter, K. & Popkin, B. M. (2002) Monitoring Health Conditions in the Russian Federation: The Russia Longitudinal Monitoring Survey 1992–2001. Report submitted to the U. S. Agency for International Development 2002 Carolina Population Center University of North Carolina at Chapel Hill, NC.
21. Guo, S. S., Huang, C., Maynard, L. M., Demerathe, E., Towne, B., Chumlea, W. C. & Siervogel, R. M. (2000) Body mass index during childhood, adolescence and young adulthood in relation to adult overweight and adiposity: the Fels Longitudinal Study. Int. J. Obes. Relat. Metab. Disord. 24:1628-1635.[Medline]
22. Eriksson, J., Forsen, T., Tuomilehto, J., Osmond, C. & Barker, D. (2001) Size at birth, childhood growth and obesity in adult life. Int. J. Obes. Relat. Metab. Disord. 25:735-740.[Medline]
23. Hoffmans, M. D., Kromhout, D. & Coulander, C. D. (1989) Body mass index at the age of 18 and its effects on 32-year-mortality from coronary heart disease and cancer. A nested case-control study among the entire 1932 Dutch male birth cohort. J. Clin. Epidemiol. 42:513-520.[Medline]
24. Doak, C. M., Adair, L. S., Monteiro, C. & Popkin, B. M. (2000) Overweight and underweight coexist within households in Brazil, China and Russia. J. Nutr. 130:2965-2971.
25. Pomerleau, J., McKee, M., Robertson, A., Kadziauskiene, K., Abaravicius, A., Vaask, S., Pudule, I. & Grinberga, D. (2001) Macronutrient and food intake in the Baltic republics. Eur. J. Clin. Nutr. 55:200-207.[Medline]
26. Biloukha, O. & Utermohlen, V. (2001) Healthy eating in Ukraine: attitudes, barriers and information sources. Public Health Nutr. 4:207-215.[Medline]
This article has been cited by other articles:
|
|
 |
|
  L. Jahns, A. Carriquiry, L. Arab, T. A. Mroz, and B. M. Popkin Within- and Between-Person Variation in Nutrient Intakes of Russian and U.S. Children Differs by Sex and Age J. Nutr., November 1, 2004; 134(11): 3114 - 3120. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
P < 0.01). (a) Income differences: P1996 < 0.01, P1998 < 0.0001, P2000 < 0.01. Decreasing low income time trend: P < 0.01. (b) Increasing high income time trend: P < 0.01. (c) No significant income differences or time trends. (d) Income differences: P1994 < 0.0001, P1996 < 0.0001, P1998 < 0.01, P2000 < 0.0001.
