QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pang, Y. Articles by Ryan, P. B. Search for Related Content PubMed PubMed Citation Articles by Pang, Y. Articles by Ryan, P. B.

---

Articles

A Longitudinal Investigation of Aggregate Oral Intake of Copper^{1 ,2}

Yaohong Pang^*, David L. MacIntosh^*3 and P. Barry Ryan

^* Department of Environmental Health Science, College of Agricultural and Environmental Sciences, University of Georgia, Athens, GA 30602 and Department of Environmental and Occupational Health, Rollins School of Public Health, Emory University, Atlanta, GA 30322

³To whom correspondence should be addressed. E-mail: dmac{at}uga.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Four-day composite solid food and beverage duplicate plates and 1-L samples of drinking water were collected from a stratified random sample of 80 individuals as part of the National Human Exposure Assessment Survey in Maryland. The media were obtained from each participant in up to six equally spaced sampling cycles over a year and analyzed for copper by inductively coupled plasma mass spectrometry. Copper concentrations (µg/kg) and consumption rates (kg/d) of solid food, beverage and drinking water were used to derive average daily aggregate oral intake of copper (µg/d). The mean aggregate copper intake of 263 measurements obtained from 68 people was 923.2 ± 685.6 µg/d (mean ± SD). Intake through solid food accounted for the majority of aggregate daily intake of copper contributing 87% on average. According to results from mixed model analysis of variance procedures, the mean log-transformed average daily copper intake in each medium except beverage exhibited significant (P < 0.05) variability among sampling cycles. Between-person variability accounted for 50% of the total variance in aggregate copper intake. As measured by the coefficient of variation, distributions of copper intake consisting of one observation per individual were more variable than the distribution consisting of the long-term average intake for each person. These results suggest that estimates of the fraction of a population at risk from chronic copper deficiency or excess copper intake can be overestimated if based upon short-term measures of copper intake. In addition, these results indicate that longitudinal information is required for accurate assessment of aggregate oral intake of copper for an individual.

KEY WORDS: • copper • aggregate daily intake • longitudinal • temporal variation • reliability • humans

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
It has been known for a number of years that copper is an essential trace element (1) . Copper is an essential component in enzymes involved with heart function, bone formation, energy metabolism, nerve transmission, elastin synthesis, skin pigmentation, normal hair growth and red blood cell production (2) . Ingestion is the major route of exposure to copper for healthy humans who are not occupationally exposed to copper. All other sources of copper intake, such as inhalation and dermal absorption, are low in comparison with oral intake (3) . There are adverse health effects for humans associated with low intakes as well as high intakes of copper (3) . Copper deficiency can cause a variety of nonspecific diseases and may be difficult to diagnose accurately (2) . Copper toxicity may occur when copper intake is excessive. Epigastric pain, headache, nausea, dizziness, vomiting and diarrhea, tachycardia, respiratory difficulty, hemolytic anemia, massive gastrointestinal bleeding, liver, kidney failure and death have been reported after accidental extreme oral ingestion (3) . Chronic intoxication caused by copper ingestion is seldom observed, probably due to effective homeostatic mechanisms in the gastrointestinal tract, such as the excretion of excess intake by bile and a capacity to link copper to ceruloplasmin (4) .

Few reports have been published on direct measurements of copper intake in community settings, especially variation of intake on the temporal dimension. To improve public health protection worldwide, more information on increased inspection of copper intake by oral pathways should be obtained (3) . In this article, we present the results of a longitudinal investigation of aggregate daily intake of copper. The objectives of our research were to assess aggregate daily intake of copper from ingestion of drinking water, beverage and solid food; to compare copper intake among those pathways; to characterize temporal variability of aggregate copper intake; and to evaluate the reliability of a short-term measure of aggregate intake for assessment of long-term average intake. Aggregate daily intake of copper was assessed from concurrent measurements of copper in drinking water, beverage and solid food samples collected from the study population.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Study population.

A stratified random sample of 80 individuals selected from four contiguous counties and the city of Baltimore, Maryland was enrolled in the study in September 1995. All participants provided informed consent under protocols approved by an institutional review board. Details of the sampling strategy and demographic characteristics of the participants are reported elsewhere (5) . Briefly, study participants included 45 females and 23 males ranging in age from 15 to 84 y (mean = 45) and representing three racial groups: African-American (n = 13), Caucasian (n = 54) and Asian/Pacific Islander (n = 1). All participants were healthy except for one that provided a self-report of an unspecified cancer. We are not aware that any participants consumed atypical diets because of disease status or any other reason. Income of all participants was above the federal poverty level. As part of a larger questionnaire, all participants responded in the negative to a question about use of dietary supplements. Samples from selected environmental and biological media, as well as questionnaire data, were collected from each participant in as many as six 1-wk long periods (cycles) approximately equally spaced between September 1995 and September 1996. Cycles 1–6 correspond to September 20 through December 23, 1995, January 15 through February 23, 1996, March 27 through April 20, 1996, April 22 through June 15, 1996, June 18 through July 27, 1996 and July 30 through September 18, 1996, respectively.

Sample collection and analysis.

As described in detail by Ryan et al. (6) , a 1-L water sample was obtained on d 1 of each sampling period from the primary source of drinking water identified by the respondent. Aliquots of samples were directly injected for multielement analysis by inductively coupled plasma mass spectrometry following U.S. Environmental Protection Agency method 200.8 (7) . A questionnaire on daily time budgets and behavior patterns was administered to participants on each day of the 7-d sampling period for a given cycle. Responses to the question "How many glasses of water did you drink today?" were used to obtain drinking water consumption rates. A nominal serving size of 0.3 L (10 fl oz) per glass was used to convert from glasses of water to liters of water consumed per day.

Participants were requested to prepare a duplicate portion of all foods consumed on four consecutive days during each sampling cycle including eating events outside of the home. Beverages were collected and stored separately from solid food samples. The weight of each 4-d solid food and beverage sample was recorded by a field technician in cycles 2–6. Samples were homogenized (solid food separate from beverages) and analyzed by inductively coupled plasma mass spectrometry for selected elements in the U.S. Food and Drug Administration laboratory in Kansas City, Missouri (8) . Participants in the National Human Exposure Assessment Survey in Maryland (NHEXAS-MD)⁴ self-reported the number of servings of various beverages and solid food consumed using a food checklist questionnaire. Details of the food sample collection and questionnaire procedures are reported elsewhere (6 ,9) . The weights of duplicate beverage and solid samples were recorded commencing with cycle 2; therefore, aggregate intake analysis for copper was based on cycles 2–6.

Quality assurance.

To ensure traceability and accuracy of the data, a series of quality assurance steps was performed. A chain-of-custody form followed each sample and questionnaire from the field, to the laboratory, and, finally, to the database manager. A sample data point not accompanied by a completed chain-of-custody, or vice versa, was omitted from subsequent analysis. Of the samples, 9% were reported as incomplete for reasons including illness, travel, not eating at home, limited food availability and fatigue. No adjustments were made to account for this noncompliance with diet collection protocols.

Detection limits (DL) and recovery efficiencies for copper in each medium were determined throughout the study. In drinking water, the DL for copper was constant over the course of the study and equal to 0.2 µg/L. Recovery efficiency from drinking water as measured by samples fortified with copper at 40 µg/L, a concentration that is within the range of observed copper levels in drinking water, was 99.5% (n = 64, SD 5.7%). The average DL for copper in beverage was 0.9 µg/kg and ranged from 0.78 to 1.04 µg/kg, while the DL for solid food samples was 4.8 µg/kg (4.20–5.06 µg/kg). The magnitude of the DL variability in beverage and solid food was small compared with copper concentrations measured in field samples. The spike recoveries centered near 100% and varied little over the study, with cycle-specific means that ranged from 94.0% to 101.0% for beverage and 86.5% to 99.2% for solid food.

Data analysis.

To evaluate temporal variation of copper intake from ingestion of drinking water, beverage and solid food, data analysis was restricted to observations from those individuals with quality-assured data in all three media from at least two monitoring cycles. Each observation in the dataset contained the copper concentration (µg/kg of sample) and average daily consumption (kg of sample/d) for drinking water, beverage and solid food. Copper concentrations in each medium below the respective DL were set to one-half the DL. Average daily intake of copper (µg/d) from each medium was computed as the product of copper concentration and consumption of the sample medium. Aggregate daily intake was computed as sum of average daily intake of copper from those three media. Because the mass of the duplicate solid food and beverage samples was not obtained in cycle 1, the dataset was restricted to observations from cycles 2–6 for a maximum number of five observations per participant.

Sample weights were determined through consideration of the sampling design with appropriate weights reflecting differential probability of selection from the initial population for each stratum. Specific weights for each participant and cycle combination can be obtained from the authors. Descriptive statistics for copper concentration and intake in each medium, aggregate daily intake and food consumption rates were generated for overall results and for each cycle. The observed data exhibited positive skewness, while natural log-transformed values were approximately normally distributed. Therefore, statistical analyses were performed on natural log-transformed values. Following Scanlon et al. (9) , a repeated-measure mixed model was used to test for significant variability of log-transformed mean consumption rates, copper concentrations and average daily copper intake for each medium and aggregate daily intake among sampling cycles. In mixed model analyses, we treated cycle as a fixed effect and participants as a random effect. A two-way generalized linear model was used to test for significant interindividual variability for each exposure metric controlling for the effect of sampling cycle. Differences with P < 0.05 were considered significant.

The statistical analyses and simulation procedures described below were used to evaluate the ability of short-term measures of aggregate copper intake to represent long-term average intake for individuals and to represent the distribution of long-term average intake for the population. These analyses were restricted to individuals who participated in each of cycles 2 through 6 and, thus, provided a complete set of observations from each cycle. In this article, we define long-term average intake as the mean of the five measures of aggregate intake obtained over the year-long study.

Reliability is a concept used to describe the degree to which a randomly selected single measure of intake taken from a set of measures for an individual represents their long-term average intake. To estimate the reliability of a short-term measure of aggregate daily intake of copper for individuals, we computed the intraclass correlation coefficient of reliability (R). R is the ratio of between-person variance to the total variance observed in a repeated measure study (10) . R ranges from 0 to 1, with values near 0 indicating low reliability and values near 1 indicating high reliability. In this work, the temporal variability observed is a characterization of within-person variability, while the total variability is a combination of the temporal variability and the between-person variability.

In addition to computing R, simulation was used to describe empirically the accuracy of long-term intake estimates based upon short-term measures for individuals as well as for a population. Four simulations were performed in which one through four cycles of aggregate intake of copper were selected at random from each participant. In the first simulation, one measure of aggregate intake for each individual was selected at random and compared with the corresponding long-term average aggregate intake using methods described later. In the second simulation, two measures of aggregate intake were selected at random for each individual and their mean was compared with the corresponding long-term average aggregate intake. In the third and fourth simulations, three and four measures of aggregate intake were selected at random for each individual and the respective means were compared with the corresponding long-term average intake. In this way, we evaluated concordance between estimates of long-term average based on one through four observations for each person and the reference value of long-term average intake based on five repeated-measures of intake for each person.

Each simulation was comprised of numerous iterations, the exact number being the minimum number of iterations required such that each possible combination of aggregate intake measures for an individual was represented in the simulation. For example, the first simulation, in which one observation of aggregate intake was selected for each participant, consisted of five iterations because there were five observations of aggregate intake per person and there are only five ways (combinations) that one object can be selected from a set of five objects. The second and third simulations consisted of 10 iterations each because there are 10 combinations of pairs and triplets from a set of five observations. The final simulation consisted of five iterations because there are five combinations of quartets drawn from a set of five observations. For each iteration, the mean of the intake measures selected at random for each individual was taken as an estimate of their long-term average intake. In addition, for each iteration, the estimated population distribution of long-term average intake was summarized as the mean and SD of the individual-specific intake estimates. Pearson correlation analysis was used to describe the association between predicted and observed long-term average intake for individuals. Relative differences between predicted averages and observed averages were used to quantify the accuracy of the simulated intake for individuals. Relative differences between predicted and observed population means and SD were used to quantify the accuracy of the simulated intake at the population level.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The final dataset contained 263 observations from 68 participants who met the quality assurance requirements and participated in more than one cycle. The number of individuals, who provided all of the drinking water, 4-d composite duplicate beverage and solid food diet samples for cycles 2–6, respectively, was 59, 54, 59, 35 and 56. The number of observations collected per participant over the five cycles was as follows: 18 respondents provided a sample in all five cycles; 28 in four cycles; 17 in three cycles; and 5 in two cycles. Nearly 70% of the 68 individuals in the study population provided both the food checklist and duplicate portions in four or more cycles.

Short-term and long-term average copper concentrations and intake.

Copper was present at detectable levels in all beverage and solid food samples and 99% of the drinking water samples. Distributions of copper concentrations in drinking water, beverage and solid food samples, consumption rates of these media and intake and aggregate daily intake contributed from those media are summarized in Table 1 . The distribution of copper concentrations in each medium was skewed right and ranged over two to three orders of magnitude. The average daily drinking water consumption rate based on all 263 observations was 1.04 L/d with a range of 0.04–4.10 L/d, while the mean weights of average daily duplicate beverage and solid food samples were 0.92 kg/d and 0.72 kg/d, respectively. As the product of concentration and consumption of sample medium, the distributions of intake of copper from the three media were skewed right and ranged over approximately three to four orders of magnitude. Aggregate daily intake computed as the sum of intake from the three media was also skewed right and ranged from 183 µg/d to 6632 µg/d with the mean 923 ± 690 µg/d (mean ± SD).

View larger version (36K): [in this window] [in a new window]   Figure 1. The distribution of annual average intake of copper from drinking water, beverage and solid food for each participant.

Descriptive statistics of intake of copper from drinking water, beverage and solid food and aggregate daily intake for each cycle are shown in Table 2 . Cycle-specific median intake from drinking water varied by a factor of four across cycles (8.14–33.4 µg/d), while those from beverage ranged from 54.5 to 59.3 µg/d and those from solid food ranged from 597 to 712 µg/d. In summary, intake through water exhibited substantial variability but only a small contribution to aggregate intake; beverage intake was relatively constant and accounted for < 10% of aggregate intake; and solid food accounted for the majority of the aggregate intake and most of the variability. In the repeated-measures mixed model, mean log-transformed intake from drinking water, solid food and aggregate daily intake varied significantly (P 0.0024) across cycles, while variation of mean intake from beverages was marginally significant (P = 0.0586). Temporal variability of aggregate daily intake of copper and the medium-specific intake was explored more fully by fitting the models to the 90 observations obtained from the 18 subjects who participated in all five sampling cycles, i.e., a balanced dataset. Analyses of copper intake among cycles in the balanced dataset were consistent with results from the full dataset, suggesting that the limited number of observations in cycle 5 did not bias the temporal analysis findings. The interindividual or between-person variability of intake was highly significant (P < 0.0001) for intake from each medium and for aggregate daily intake.

R was 0.50 for aggregate daily intake of copper. In the simulation method used to illustrate empirically the reliability of short-term measures of aggregate copper intake, the correlation coefficient between one randomly selected aggregate intake measure and observed annual average for individuals was 0.70, and 95% of the estimates were within 67.0% of the observed values (Table 3 ). The correlation coefficient increased to 0.96, and the 95% confidence interval of the relative difference between estimated and observed intake decreased to 16.8% when a sample of four observations was drawn for each respondent (Table 3) .

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The Continuing Survey of Food Intakes by Individuals 1994–1996 (CSFII) conducted by the Agricultural Research Service, U.S. Department of Agriculture is the most recent in a series of U.S. Department of Agriculture surveys designed to monitor food use and food consumption patterns in the U.S. population (12) . Based on 2 or 3-d food diaries and food composition tables from the CSFII, the mean copper intake from food and beverage for U.S. adults is estimated to be 1.2 mg/d (13) . Copper intake reported in the CSFII is 30% greater than the mean aggregate intake of short-term measures observed in our study and is equal to the 87th percentile of short-term aggregate intake in this population drawn from greater Baltimore, Maryland. In comparison to other U.S. nutrition surveys, our results are lower than the mean copper intake assessed by the Third National Health and Nutrition Examination Survey (female and male 31–50 y intake of 1.2 and 1.7 mg/d, respectively) and are comparable with intake estimates from the U.S. Food and Drug Administration Total Diet Study (female and male 31–50 y intake of 0.8 and 1.1 mg/d, respectively) (14) . Mean copper intake reported for other countries and assessed using duplicate diet methods are 1.7 mg/d in Bangkok, Thailand (15) , 1.1 mg/d in northeast Spain (16) and 1.0 mg/d for children in Duisburg, Germany (17) .

The aggregate daily intake of copper observed in this investigation also can be compared with published levels of acceptable or safe intake. RDA is defined for essential nutrients at amounts sufficient to cover individual variations in requirements and provide a margin of safety above minimal requirements (18) . In early 2001, a new RDA for copper of 0.9 mg/d was published by the National Academy of Sciences (14) . In the NHEXAS-MD study, the mean aggregate daily intake of copper based on the short-term measures (i.e., 4-d composites) was 0.923 mg/d, with a median of 0.759 mg/d. Annual average aggregate daily intake of copper was < 0.9 mg/d for 60% of the NHEXAS-MD population. It is important to note that values similar to the RDA, such as safe and adequate daily intake, are not recommended for individual assessment but are instead intended to ensure adequate intake for all members of a population rather than for specific persons (11) .

Weighed food records over multiple days can provide solid data for nutrient assessment if the recording process does not influence food consumption choices (11) . The duplicate plates collected in this investigation are a form of weighed food record. The corresponding copper assays provide a more direct measure of copper intake than can be obtained from the combination of food diary or questionnaire data and food composition tables. Nevertheless, the duplicate plate methodology used here is prone to its own types of errors. First, as stated earlier, 9% of the respondents indicated that at least some food or beverage was not included in the duplicate plate composite samples. Second, good quality compliance with duplicate plate protocols has been reported for up to four consecutive days, but not much longer (19) . In our study, the average weight of duplicate solid food and beverage samples decreased by 16% and 13%, respectively, from cycles 2 through 6 (9 ,20) . Therefore, compliance with sampling protocols may have waned with each additional sampling cycle. Third, the absence of data from October through November could be a source of error due to seasonal effects on food availability and use (11) . We suspect that potential error resulting from these factors is relatively small, although the likelihood and magnitude cannot be quantified with a high degree of confidence.

The principal objective of this study was to determine whether there is temporal variation in aggregate daily intake of copper and the medium-specific intake. We found mean log-transformed aggregate daily intake varied significantly among five sampling periods over a year and absolute values varied within a range of 30%. Mean log-transformed copper intake from drinking water and solid food also varied significantly, while intake from beverage varied marginally significantly during this period. These results indicate that the timing of data collection should be considered in copper intake and risk assessments. Information on periodicity of temporal variation in aggregate copper intake would be useful; however, multiple years of data are required to perform such an evaluation. Although statistically significant, variability of the aggregate copper intake across cycles may not be substantive in the context of some nutrition and environmental health applications. Other errors or uncertainties in health risk assessments (e.g., outcome measurements and dose-response relationships) may be much larger. For example, only limited information, primarily from animal models, is available to determine a safe level of maximum copper intake (3) .

The temporal dimension of intake of substances in the environment is of interest because biological response to a nutritional deficiency or an environmental challenge can be associated with the duration and timing of the intake. The findings from repeated-measure studies have implications for epidemiology and quantitative risk assessment, tools used to evaluate likelihood of disease as a function of nutritive status or intake of environmental contaminants.

The objective of exposure assessment in epidemiology often is to categorize (e.g., the upper or lower quartile) or rank individuals by degree of intake or exposure to the factor hypothesized to be associated with likelihood of disease. If the intake is reasonably constant over time, then accurate determination of intake for an individual can be obtained from a single short-term intake measure. In the NHEXAS-MD study, we found moderate (r = 0.50) within-person variability of aggregate daily intake of copper over time, indicating that within-person and between-person variation contributed equally to total variance in the short-term measures of aggregate daily intake of copper. Results of the simulations summarized in Table 3 indicate that the mean of three short-term copper intake measures for an individual is strongly correlated (r = 0.90) with the observed long-term average intake. This information can be used in the design of future epidemiological studies of copper and health following methods described elsewhere (10) .

In quantitative risk assessment, intake data are typically required for populations rather than for individuals. Emphasis is often placed on the central tendency or the tails of the intake for a population. The adverse health effects of chronic copper deficiency are well known; thus, information on the distribution of chronic copper intake for the U.S. population is of interest for evaluating public health and nutritional status. Short-term intake data ascertained through population-based surveys like the CSFII are a principal source of information for this type of evaluation. The degree to which a distribution of short-term (e.g., 2–4 d) estimates of copper intake represents the distribution of chronic copper intake is not well known. The summary of simulation results shown in Table 4 provides information relevant to this knowledge gap. The estimated population mean based on five iterations in which one short-term aggregate copper intake measure selected at random per person was within 20% of the mean of long-term average intakes based on five repeated-measures per person. However, the SD for distributions of single short-term intake measures was only in marginal agreement with the SD of long-term average copper intake. Thus, in this population, reliance on short-term intake measures as surrogates for long-term average intake could lead to erroneous conclusions about the number of people or fraction of the population at risk of health effects from copper deficiency or copper toxicity.

	ACKNOWLEDGMENTS

 
We thank Mary Ann Johnson of Department of Foods and Nutrition, the University of Georgia, for her contributions to this project.

	FOOTNOTES

 
¹ Presented in part orally at the 10th Annual Conference of the International Society of Exposure Analysis, Monterey, CA, October 26, 2000. The presentation title was Aggregate Consumption of Copper, Selenium and Nickel (3E-o3o in abstract book), by Y. Pang, D. MacIntosh and P. B. Ryan.

² Supported by the United States Environmental Protection Agency under Cooperative Agreement R822038-1, the U.S. Department of Agriculture Hatch Project Number GEO00843 and the University of Georgia Research Foundation.

⁴ Abbreviations used: CSFII, Continuing Survey of Food Intakes by Individuals; DL, detection limit; NHEXAS-MD, National Human Exposure Assessment Survey in Maryland; R, intraclass correlation coefficient of reliability.

Manuscript received February 5, 2001. Initial review completed March 8, 2001. Revision accepted April 23, 2001.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Swerts J., Benemariya H., Robberecht H., van Cauwenbergh R., Deelstra H. Daily dietary intake of copper and zinc by several population groups in Belgium: preliminary reports. J. Trace Elem. Electrolytes Health Dis. 1993;7:165-169[Medline]

2. Johnson M. A. Physiology of copper. Encyclopaedia of Food Science, Food Technology and Nutrition 1993:1243-1247 Academic Press London; San Diego;

3. World Health Organization Environmental Health Criteria 200, Copper 1998 World Health Organization Geneva, Switzerland.

4. Pettersson R., Rasmussen F. Daily intake of copper from drinking water among young children in Sweden. Environ. Health Perspect. 1999;107:441-446[Medline]

5. Echols S. L., MacIntosh D. L., Hammerstrom K. A., Ryan P. B. Temporal variability of microenvironmental time budgets in Maryland. J. Expo. Anal. Environ. Epidemiol. 1999;9:502-512[Medline]

6. Ryan P. B., Huet N., MacIntosh D. L. Longitudinal investigation of exposure to arsenic, cadmium, and lead in drinking water. Environ. Health Perspect. 2000;108:731-735[Medline]

7. Long S., Martin T. EPA method 200.8: determination of trace elements in waters and wastes by inductively coupled plasma–mass spectrometry. Methods for the Determination of Metals in Environmental Samples 1991 U.S. Environmental Protection Agency Cincinnati, OH.

8. Food and Drug Administration Standard Operating Procedure for Determination of Trace Elements in NHEXAS Food Composites by Inductively Coupled Plasma Mass Spectrometry 1997 U.S. Food and Drug Administration Washington, DC.

9. Scanlon K. A., MacIntosh D. L., Hammerstrom K. A., Ryan P. B. A longitudinal investigation of solid-food based dietary exposure to selected elements. J. Expo. Anal. Environ. Epidemiol. 1999;9:485-493[Medline]

10. Fleiss J. L. Reliability of measurement. The Design and Analysis of Clinical Experiments 1986:1-32 John Wiley & Sons New York, NY.

11. National Academy Press Dietary Reference Intakes: Applications in Dietary Assessment 2000 National Academy Press Washington, DC.

12. Casey P. H., Goolsby S. L., Lensing S. Y., Perloff B. P., Bogle M. L. The use of telephone interview methodology to obtain 24-hour dietary recalls. J. Am. Diet. Assoc. 1999;99:1406-1411[Medline]

13. U.S. Department of Agriculture Continuing Survey of Food Intakes by Individuals and the Diet and Health Knowledge Survey 1994–1996 1998 U.S. Department of Agriculture, Beltsville Human Nutrition Research Center/ARS Beltsville, MD.

14. Institute of Medicine Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc 2001:177-204 National Academy Press Washington, DC.

15. Songchitsomboon S., Komindr S., Piaseu N. Zinc and copper intake and sources in healthy adults living in Bangkok and surrounding districts. Biol. Trace Elem. Res. 1998;61:97-104[Medline]

16. Schuhmacher M., Domingo J. L., Llobet J. M., Corbella J. Dietary intake of copper, chromium and zinc in Tarragona Province, Spain. Sci. Total Environ. 1993;132:3-10[Medline]

17. Laryea M. D., Schnittert B., Kersting M., Wilhelm M., Lombeck I. Macronutrient, copper, and zinc intakes of young: German children as determined by duplicate food samples and diet records. Ann. Nutr. Metab. 1995;39:271-278[Medline]

18. National Academy of Sciences Recommended Dietary Allowances 10th ed. 1989:285 National Academy Press Washington, DC.

19. Thomas K. W., Sheldon L. S., Pellizzari E. D., Handy R. W., Roberds J. M., Berry M. R. Testing duplicate diet sample collection methods for measuring personal dietary exposures to chemical contaminants. J. Expo. Anal. Environ. Epidemiol. 1997;7:17-36[Medline]

20. MacIntosh D. L., Kabiru C., Scanlon K. A., Ryan P. B. Longitudinal investigation of exposure to arsenic, cadmium, chromium and lead via beverage consumption. J. Expo. Anal. Environ. Epidemiol. 2000;10:196-205[Medline]

This article has been cited by other articles:

				L. M. Klevay Homocysteine Is Not So Paradoxical Arterioscler. Thromb. Vasc. Biol., October 1, 2008; 28(10): e160 - e160. [Full Text] [PDF]

				Y. Jiang, C. Reynolds, C. Xiao, W. Feng, Z. Zhou, W. Rodriguez, S. C. Tyagi, J. W. Eaton, J. T. Saari, and Y. J. Kang Dietary copper supplementation reverses hypertrophic cardiomyopathy induced by chronic pressure overload in mice J. Exp. Med., March 19, 2007; 204(3): 657 - 666. [Abstract] [Full Text] [PDF]

				Y. Zhou, Y. Jiang, and Y. J. Kang A BRIEF COMMUNICATION: Copper Inhibition of Hydrogen Peroxide-Induced Hypertrophy in Embryonic Rat Cardiac H9c2 Cells Experimental Biology and Medicine, March 1, 2007; 232(3): 385 - 389. [Abstract] [Full Text] [PDF]

				J. T. Saari, P. G. Reeves, W. T. Johnson, and L. K. Johnson Pinto Beans Are a Source of Highly Bioavailable Copper in Rats J. Nutr., December 1, 2006; 136(12): 2999 - 3004. [Abstract] [Full Text] [PDF]

				L. M. Klevay Heart failure improvement from a supplement containing copper Eur. Heart J., January 1, 2006; 27(1): 117 - 117. [Full Text] [PDF]

				Y. Li, L. Wang, D. A. Schuschke, Z. Zhou, J. T. Saari, and Y. J. Kang Marginal Dietary Copper Restriction Induces Cardiomyopathy in Rats J. Nutr., September 1, 2005; 135(9): 2130 - 2136. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pang, Y. Articles by Ryan, P. B. Search for Related Content PubMed PubMed Citation Articles by Pang, Y. Articles by Ryan, P. B.