Olga W. Souverein

Comparing four methods to estimate usual intake distributions

Background/Objectives:The aim of this paper was to compare methods to estimate usual intake distributions of nutrients and foods. As ‘true’ usual intake distributions are not known in practice, the comparison was carried out through a simulation study, as well as empirically, by application to data from the European Food Consumption Validation (EFCOVAL) Study in which two 24-h dietary recalls (24-HDRs) and food frequency data were collected. The methods being compared were the Iowa State University Method (ISU), National Cancer Institute Method (NCI), Multiple Source Method (MSM) and Statistical Program for Age-adjusted Dietary Assessment (SPADE).Subjects/Methods:Simulation data were constructed with varying numbers of subjects (n), different values for the Box–Cox transformation parameter (λBC) and different values for the ratio of the within- and between-person variance (rvar). All data were analyzed with the four different methods and the estimated usual mean intake and selected percentiles were obtained. Moreover, the 2-day within-person mean was estimated as an additional ‘method’. These five methods were compared in terms of the mean bias, which was calculated as the mean of the differences between the estimated value and the known true value. The application of data from the EFCOVAL Project included calculations of nutrients (that is, protein, potassium, protein density) and foods (that is, vegetables, fruit and fish).Results:Overall, the mean bias of the ISU, NCI, MSM and SPADE Methods was small. However, for all methods, the mean bias and the variation of the bias increased with smaller sample size, higher variance ratios and with more pronounced departures from normality. Serious mean bias (especially in the 95th percentile) was seen using the NCI Method when rvar=9, λBC=0 and n=1000. The ISU Method and MSM showed a somewhat higher s.d. of the bias compared with NCI and SPADE Methods, indicating a larger method uncertainty. Furthermore, whereas the ISU, NCI and SPADE Methods produced unimodal density functions by definition, MSM produced distributions with ‘peaks’, when sample size was small, because of the fact that the populations usual intake distribution was based on estimated individual usual intakes. The application to the EFCOVAL data showed that all estimates of the percentiles and mean were within 5% of each other for the three nutrients analyzed. For vegetables, fruit and fish, the differences were larger than that for nutrients, but overall the sample mean was estimated reasonably.Conclusions:The four methods that were compared seem to provide good estimates of the usual intake distribution of nutrients. Nevertheless, care needs to be taken when a nutrient has a high within-person variation or has a highly skewed distribution, and when the sample size is small. As the methods offer different features, practical reasons may exist to prefer one method over the other.

Effect of folate intake on health outcomes in pregnancy: a systematic review and meta-analysis on birth weight, placental weight and length of gestation

The beneficial effect of folic acid supplementation before and shortly after conception is well recognized, whereas the effect of supplementation during the second and third trimesters is controversial and poorly documented. Our aims were to systematically review randomized controlled trials (RCTs) investigating the effect of folate supplementation on birth weight, placental weight and length of gestation and to assess the dose–response relationship between folate intake (folic acid plus dietary folate) and health outcomes. The MEDLINE, EMBASE and Cochrane Library CENTRAL databases were searched from inception to February 2010 for RCTs in which folate intake and health outcomes in pregnancy were investigated. We calculated the overall intake-health regression coefficient (β^) by using random-effects meta-analysis on a loge-loge scale. Data of 10 studies from 8 RCTs were analyzed. We found significant dose–response relationship between folate intake and birth weight (P=0.001), the overall β^ was 0.03 (95% confidence interval (CI): 0.01, 0.05). This relationship indicated 2% increase in birth weight for every two-fold increase in folate intake. In contrast, we did not find any beneficial effect of folate supplementation on placental weight or on length of gestation. There is a paucity of well-conducted RCTs investigating the effect of folate supplementation on health outcomes in pregnancy. The dose–response methodology outlined in the present systematic review may be useful for designing clinical studies on folate supplementation and for developing recommendations for pregnant women.

Two non-consecutive 24 h recalls using EPIC-Soft software are sufficiently valid for comparing protein and potassium intake between five European centres – results from the European Food Consumption Validation (EFCOVAL) study

The use of two non-consecutive 24 h recalls using EPIC-Soft for standardised dietary monitoring in European countries has previously been proposed in the European Food Consumption Survey Method consortium. Whether this methodology is sufficiently valid to assess nutrient intake in a comparable way, among populations with different food patterns in Europe, is the subject of study in the European Food Consumption Validation consortium. The objective of the study was to compare the validity of usual protein and K intake estimated from two non-consecutive standardised 24 h recalls using EPIC-Soft between five selected centres in Europe. A total of 600 adults, aged 45-65 years, were recruited in Belgium, the Czech Republic, France, The Netherlands and Norway. From each participant, two 24 h recalls and two 24 h urines were collected. The mean and distribution of usual protein and K intake, as well as the ranking of intake, were compared with protein and K excretions within and between centres. Underestimation of protein (range 2-13%) and K (range 4-17%) intake was seen in all centres, except in the Czech Republic. We found a fair agreement between prevalences estimated based on the intake and excretion data at the lower end of the usual intake distribution (< 10% difference), but larger differences at other points. Protein and K intake was moderately correlated with excretion within the centres (ranges = 0·39-0·67 and 0·37-0·69, respectively). These were comparable across centres. In conclusion, two standardised 24 h recalls (EPIC-Soft) appear to be sufficiently valid for assessing and comparing the mean and distribution of protein and K intake across five centres in Europe as well as for ranking individuals.

Vitamin B12, Folate, Homocysteine, and Bone Health in Adults and Elderly People: A Systematic Review with Meta-Analyses

Elevated homocysteine levels and low vitamin B12 and folate levels have been associated with deteriorated bone health. This systematic literature review with dose-response meta-analyses summarizes the available scientific evidence on associations of vitamin B12, folate, and homocysteine status with fractures and bone mineral density (BMD). Twenty-seven eligible cross-sectional (n = 14) and prospective (n = 13) observational studies and one RCT were identified. Meta-analysis on four prospective studies including 7475 people showed a modest decrease in fracture risk of 4% per 50 pmol/L increase in vitamin B12 levels, which was borderline significant (RR = 0.96, 95% CI = 0.92 to 1.00). Meta-analysis of eight studies including 11511 people showed an increased fracture risk of 4% per μmol/L increase in homocysteine concentration (RR = 1.04, 95% CI = 1.02 to 1.07). We could not draw a conclusion regarding folate levels and fracture risk, as too few studies investigated this association. Meta-analyses regarding vitamin B12, folate and homocysteine levels, and BMD were possible in female populations only and showed no associations. Results from studies regarding BMD that could not be included in the meta-analyses were not univocal.

Biomarker-based evaluation of two 24-h recalls for comparing usual fish, fruit and vegetable intakes across European centers in the EFCOVAL Study

Background/Objectives:A standardized methodology is important to enable consistent monitoring of dietary intake across European countries. For this reason, we evaluated the comparability of the assessment of usual food intake collected with two non-consecutive computerized 24-h dietary recalls (24-HDRs) and a food propensity questionnaire (FPQ) among five European centers.Subjects/Methods:Two 24-HDRs using EPIC-Soft (the software developed to conduct 24-HDRs in the European Prospective Investigation into Cancer and Nutrition (EPIC) Study) were performed to determine fish, fruit and vegetable (FV) consumed by 600 adults in Belgium (BE), the Czech Republic (CZ), France (FR), the Netherlands (NL) and Norway (NO) in a validation study. An FPQ was used to identify non-consumers. Information from the 24-HDRs and FPQ were used to estimate individual usual food intake by the Multiple Source Method (MSM). Blood samples were drawn to determine fatty acids in phospholipids and serum carotenoids as biomarkers of fish, and FV intake, respectively.Results:The pooled correlation between usual fish intake and eicosapentaenoic acid plus docosahexaenoic acid in phospholipids was 0.19 in men and 0.31 in women (P for heterogeneity >0.50) and center-specific correlations ranged between 0.08 (CZ) and 0.28 (BE and NO) in men, and between 0.19 (BE) and 0.55 (FR) in women. For usual FV intake, the pooled correlation with serum carotenoids was 0.31 in men and 0.40 in women (P for heterogeneity >0.10); the center-specific correlations varied between 0.07 (NO) and 0.52 (FR) in men, and between 0.25 (NL) and 0.45 (NO) in women.Conclusion:Two standardized 24-HDRs using EPIC-Soft and an FPQ appeared to be appropriate to rank individuals according to their fish and FV intake in a comparable way among five European centers.

The Relationship between Zinc Intake and Serum/Plasma Zinc Concentration in Children: A Systematic Review and Dose-Response Meta-Analysis

Recommendations for zinc intake during childhood vary widely across Europe. The EURRECA project attempts to consolidate the basis for the definition of micronutrient requirements, taking into account relationships among intake, status and health outcomes, in order to harmonise these recommendations. Data on zinc intake and biomarkers of zinc status reported in randomised controlled trials (RCTs) can provide estimates of dose-response relationships which may be used for underpinning zinc reference values. This systematic review included all RCTs of apparently healthy children aged 1–17 years published by February 2010 which provided data on zinc intake and biomarkers of zinc status. An intake-status regression coefficient () was calculated for each individual study and calculated the overall pooled and SE () using random effects meta-analysis on a double log scale. The pooled dose-response relationship between zinc intake and zinc status indicated that a doubling of the zinc intake increased the serum/plasma zinc status by 9%. This evidence can be utilised, together with currently used balance studies and repletion/depletion studies, when setting zinc recommendations as a basis for nutrition policies.

Vitamin B12 Intake and Status and Cognitive Function in Elderly People

Current recommendations on vitamin B12 intake vary from 1.4 to 3.0 μg per day and are based on the amount needed for maintenance of hematologic status or on the amount needed to compensate obligatory losses. This systematic review evaluates whether the relation between vitamin B12 intake and cognitive function should be considered for underpinning vitamin B12 recommendations in the future. The authors summarized dose-response evidence from randomized controlled trials and prospective cohort studies on the relation of vitamin B12 intake and status with cognitive function in adults and elderly people. Two randomized controlled trials and 6 cohort studies showed no association or inconsistent associations between vitamin B12 intake and cognitive function. Random-effects meta-analysis showed that serum/plasma vitamin B12 (50 pmol/L) was not associated with risk of dementia (4 cohort studies), global cognition z scores (4 cohort studies), or memory z scores (4 cohort studies). Although dose-response evidence on sensitive markers of vitamin B12 status (methylmalonic acid and holotranscobalamin) was scarce, 4 of 5 cohort studies reported significant associations with risk of dementia, Alzheimers disease, or global cognition. Current evidence on the relation between vitamin B12 intake or status and cognitive function is not sufficient for consideration in the development of vitamin B12 recommendations. Further studies should consider the selection of sensitive markers of vitamin B12 status.

Folate Intake and Markers of Folate Status in Women of Reproductive Age, Pregnant and Lactating Women: A Meta-Analysis

Background. Pregnant and breastfeeding women are at risk for folate deficiency. Folate supplementation has been shown to be associated with enhanced markers of folate status. However, dose-response analyses for adult women are still lacking. Objective. To assess the dose-response relationship between total folate intake (folic acid plus dietary folate) and markers of folate status (plasma/serum folate, red blood cell folate, and plasma homocysteine); to evaluate potential differences between women in childbearing age, pregnant and lactating women. Methods. Electronic literature searches were carried out on three databases until February 2010. The overall pooled regression coefficient (β) and SE(β) were calculated using meta-analysis on a double-log scale. Results. The majority of data was based on nonpregnant, nonlactating women in childbearingage. The pooled estimate of the relationship between folate intake and serum/plasma folate was 0.56 (95% CI = 0.40–0.72, P < 0.00001); that is, the doubling of folate intake increases the folate level in serum/plasma by 47%. For red blood cell folate, the pooled-effect estimate was 0.30 (95% CI = 0.22–0.38, P < 0.00001), that is, +23% for doubling intake. For plasma-homocysteine it was –0.10 (95% = –0.17 to –0.04, P = 0.001), that is, –7% for doubling the intake. Associations tended to be weaker in pregnant and lactating women. Conclusion. Significant relationships between folate intake and serum/plasma folate, red blood cell folate, and plasma homocysteine were quantified. This dose-response methodology may be applied for setting requirements for women in childbearing age, as well as for pregnant and lactating women.

Transformations of summary statistics as input in meta-analysis for linear dose-response models on a logarithmic scale: a methodology developed within EURRECA

BackgroundTo derive micronutrient recommendations in a scientifically sound way, it is important to obtain and analyse all published information on the association between micronutrient intake and biochemical proxies for micronutrient status using a systematic approach. Therefore, it is important to incorporate information from randomized controlled trials as well as observational studies as both of these provide information on the association. However, original research papers present their data in various ways.MethodsThis paper presents a methodology to obtain an estimate of the dose–response curve, assuming a bivariate normal linear model on the logarithmic scale, incorporating a range of transformations of the original reported data.ResultsThe simulation study, conducted to validate the methodology, shows that there is no bias in the transformations. Furthermore, it is shown that when the original studies report the mean and standard deviation or the geometric mean and confidence interval the results are less variable compared to when the median with IQR or range is reported in the original study.ConclusionsThe presented methodology with transformations for various reported data provides a valid way to estimate the dose–response curve for micronutrient intake and status using both randomized controlled trials and observational studies.

Biomarker responses to folic acid intervention in healthy adults: a meta-analysis of randomized controlled trials

BACKGROUND The task of revising dietary folate recommendations for optimal health is complicated by a lack of data quantifying the biomarker response that reliably reflects a given folate intake. OBJECTIVE We conducted a dose-response meta-analysis in healthy adults to quantify the typical response of recognized folate biomarkers to a change in folic acid intake. DESIGN Electronic and bibliographic searches identified 19 randomized controlled trials that supplemented with folic acid and measured folate biomarkers before and after the intervention in apparently healthy adults aged ≥18 y. For each biomarker response, the regression coefficient (β) for individual studies and the overall pooled β were calculated by using random-effects meta-analysis. RESULTS Folate biomarkers (serum/plasma and red blood cell folate) increased in response to folic acid in a dose-response manner only up to an intake of 400 μg/d. Calculation of the overall pooled β for studies in the range of 50 to 400 μg/d indicated that a doubling of folic acid intake resulted in an increase in serum/plasma folate by 63% (71% for microbiological assay; 61% for nonmicrobiological assay) and red blood cell folate by 31% (irrespective of whether microbiological or other assay was used). Studies that used the microbiological assay indicated lower heterogeneity compared with studies using nonmicrobiological assays for determining serum/plasma (I(2) = 13.5% compared with I(2) = 77.2%) and red blood cell (I(2) = 45.9% compared with I(2) = 70.2%) folate. CONCLUSIONS Studies administering >400 μg folic acid/d show no dose-response relation and thus will not yield meaningful results for consideration when generating dietary folate recommendations. The calculated folate biomarker response to a given folic acid intake may be more robust with the use of a microbiological assay rather than alternative methods for blood folate measurement.