help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Journal of Clinical Endocrinology & Metabolism , doi:10.1210/jc.2007-1279
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hyppönen, E.
Right arrow Articles by Gibb, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hyppönen, E.
Right arrow Articles by Gibb, I.
Related Collections
Right arrow Calcium and Bone Metabolism
Right arrow Metabolism
The Journal of Clinical Endocrinology & Metabolism Vol. 92, No. 12 4615-4622
Copyright © 2007 by The Endocrine Society

Serum 25-Hydroxyvitamin D Measurement in a Large Population Survey with Statistical Harmonization of Assay Variation to an International Standard

Elina Hyppönen, Steve Turner, Phillippa Cumberland, Christine Power and Ian Gibb

Centre for Paediatric Epidemiology and Biostatistics (E.H., P.C., C.P.), Institute of Child Health, London WC1N 1EH, United Kingdom; and Department of Clinical Biochemistry (S.T., I.G.), Newcastle upon Tyne Hospitals National Health Service Trust, Royal Victoria Infirmary, Newcastle upon Tyne NE1 4LP, United Kingdom

Address all correspondence and requests for reprints to: Dr. Elina Hyppönen, Centre for Pediatric Epidemiology and Biostatistics, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, United Kingdom. E-mail: e.hypponen{at}ich.ucl.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Subjects and Methods
 Results
 Discussion
 References
 
Context: An automated application of Immunodiagnostic Systems Limited (IDS) OCTEIA 25-hydroxyvitamin D [25(OH)D] enzyme immunoassay was developed for analyses of 25(OH)D in more than 7000 participants of the 1958 cohort. Variation between 25(OH)D assays hampers between-study comparisons and the definition of relevant cutoffs for hypovitaminosis D.

Objective: The objective of the study was to evaluate the importance of assay variation on the estimated prevalence of hypovitaminosis D and assess the use of statistical harmonization to overcome the observed differences.

Design: Agreement analyses were performed between two commercial 25(OH)D assays (IDS enzyme immunoassay and Diasorin RIA), with validation using performance data from Vitamin D External Quality Assessment Scheme (DEQAS).

Setting: The study was conducted in England, Scotland, and Wales.

Participants: Members of the 1958 British birth cohort participated in the study.

Main Outcome Measures: 25(OH)D was measured both by IDS and Diasorin RIA in 781 samples. Additional quality control data were obtained through participation in DEQAS (five distributions throughout the survey).

Results: Average 25(OH)D concentrations by IDS were –15.7 and –13.7 nmol/liter lower, compared with Diasorin or DEQAS mean, respectively (both P < 0.0001). Graphical examination demonstrated a dose-related bias between IDS with Diasorin and DEQAS mean, but log transformation removed the bias. After using the log difference between the measurements as an adjustment factor, there were no differences in average 25(OH)D concentrations (P ≥ 0.21 for comparison of IDS with Diasorin or DEQAS) and estimates for hypovitaminosis D obtained by IDS were similar to Diasorin.

Conclusions: Differences between assays have implications for public health messages about hypovitaminosis D. Harmonization of results with DEQAS enabled the use of previously determined cutoffs for hypovitaminosis D.


   Introduction
Top
 Abstract
 Introduction
Subjects and Methods
 Results
 Discussion
 References
 
HYPOVITAMINOSIS D has been associated with important short- and long-term health effects, including the risk of common chronic diseases such as diabetes, cardiovascular conditions, and cancer (1, 2). During the past decade, there have been discussions about possible epidemics of hypovitaminosis D in Western populations and even calls for the screening of vitamin D status as part of routine health care surveillance (1). The best indicator for vitamin D status is 25-hydroxyvitamin D [25(OH)D]. However, there are a number of method principles applied in the measurement of 25(OH)D that are associated with practical limitations, giving marked variation in estimation of 25(OH)D concentration by the different assay methods (3, 4, 5, 6, 7, 8, 9). Lack of method standardization is a major drawback in relating data from well-designed studies to each other as well as when defining cutoffs for hypovitaminosis D (4, 10).

We measured 25(OH)D from more than 7000 participants in the 1958 British birth cohort (11, 12). For this large population-based survey, we developed an automated application of Immunodiagnostic Systems Ltd. (IDS) enzyme immunoassay (EIA) with the prospect of minimizing hands-on time and interanalyst variation. To monitor and quantify the difference in 25(OH)D concentration measured by automated IDS with that obtained by the most commonly used method, namely Diasorin RIA (7), we carried out comparison analyses at the start of the study and on representative subsets of samples in monthly batches throughout a 19-month period of data collection. Further between-method comparisons were done through participation in the Vitamin D External Quality Assessment Scheme (DEQAS) (7, 9). In this paper we report how, using these evaluation data and a simple statistical procedure described by Bland and Altman (13), we transformed the values of 25(OH)D obtained by automated IDS to correspond with levels obtained by both Diasorin and the mean of DEQAS to facilitate interstudy comparisons and the use of previously determined cutoffs for hypovitaminosis D. We also demonstrate the public health implications of assay variation on the perceived problem with hypovitaminosis D.


   Subjects and Methods
Top
 Abstract
 Introduction
 Subjects and Methods
Results
 Discussion
 References
 
The 1958 British birth cohort covers all births in England, Wales, and Scotland during 1 wk in March 1958 (14, 15). From an original cohort of more than 17,000 births, survivors have been followed up into adulthood. Fieldwork for a biomedical survey was conducted from September 2002 to March 2004; hence, data were collected over the full seasonal range for the United Kingdom. Of 12,069 cohort members eligible for inclusion in this 45-yr study, 9,377 agreed to a clinical examination by a trained research nurse visiting the home. A venous blood sample was obtained from 88% (n = 8207) of participants. Blood samples, collected into Sarstedt evacuated tubes without anticoagulant, were sent by regular mail (without prior cooling) directly to the central laboratory. 25(OH)D was measured in 7591 participants. In this paper we present information from the initial 25(OH)D assay performance evaluation, based on 142 patient samples from a pilot trial (not participants in the 1958 cohort). We also present quality control data for 781 participants in the 1958 cohort, which were collected at the start and then at monthly intervals during the 19-month period of fieldwork (all participants were only included once). Participants gave consent for use of their blood samples in medical research studies of the causes, diagnosis, treatment, or outcome of disease. Ethical approval for the biomedical survey was obtained from the South East Multi-Centre Research Ethics Committee (United Kingdom).

Measurement of 25(OH)D

Assessments of 25(OH)D for all survey samples were done in one centralized laboratory, the Royal Victoria Infirmary (Newcastle upon Tyne Hospitals National Health Service Trust, Newcastle upon Tyne, UK). The IDS OCTEIA 25(OH)D EIA (IDS EIA) was used to quantify 25(OH)D in serum. Serum samples, calibrators, and controls were treated with 2% sodium hydroxide and then diluted with biotin-labeled 25(OH)D. These diluted samples were incubated in microtiter wells coated with a sheep anti-25(OH)D polyclonal antibody for 2 h at room temperature before aspiration and washing. Horseradish peroxidase-labeled avidin, which binds selectively to complexed biotin, was then added. After a further wash step, color develops as a result of using the chromogen substrate tetramethylbenzoate. Absorbance reading is then indirectly proportional to 25(OH)D concentration. A fully automated application of this kit was developed and evaluated in the central analytical laboratory on a BEP2000 programmable microtiter plate analyzer (Dade-Behring, Marburg, Germany). The automated method followed the manual protocol outlined above, the analyzer programming enabling automation of sample pretreatment and predilution, sample and reagent dispensing, temperature-controlled incubation, wash processes, and plate transports to and from the incubators and to and from the integrated photometric measurement system. The instrument has a maximum capacity of 4 x 96-well plates.

Automated assay characterization

Within-assay imprecision. Performance in the automated assay was assessed by determining imprecision using 20 replicates of quality control sera at each of three 25(OH)D concentrations (18.2, 88.2, and 121 nmol/liter) in a single assay.

Between-assay imprecision. This was initially assessed using three levels of quality control sera (IDS controls 1, 2, and 3) in separate assays over 3 wk (n = 13). Data were also accumulated on further batches of these materials by analyzing them in all subsequent assay batches throughout the full 19-month study period (n = 74).

Functional sensitivity. The sensitivity was determined as the 25(OH)D concentration at which the analytical coefficient of variation (CV) = 20% in a precision profile consisting of 6-fold replicate analyses at each of 12 kit standard and standard mixture concentrations in the range 2–438 nmol/liter.

Method linearity. This was verified using replicate analyses of three serum samples with high 25(OH)D levels in doubling-dilution series up to 1 in 16.

Sample stability. In the automated procedure, the stability was assessed by repeat analysis of three serum samples over five freeze-thaw cycles over 2 wk.

Minimum sample assay volume. Small sample analysis capability of the automated procedure was assessed by presenting a series of premeasured decreasing sample volumes (250 to 50 µl) for analysis and observing the first significant deviation from the baseline 25(OH)D concentration.

Recovery of 25(OH)D. A 12.5-nmol/liter stock solution of 25(OH)D (Sigma Chemicals, Poole, UK; product code H-4014) prepared in 98% HPLC grade ethanol was added to two serum samples with known endogenous 25(OH)D levels to increase expected concentrations by 25–250 nmol/liter.

Comparison between IDS EIA and Diasorin

Agreement was evaluated between IDS with the method most commonly used for epidemiological surveys, namely (manual) Diasorin I125-radiolabeled method. Comparison data were obtained at baseline (n = 142) and in monthly subset comparison batches throughout the period of fieldwork. Data from February and March (years 2003 and 2004) were combined; hence, there were 17 batches of 46 samples each (17 x 46 = 782, final n = 781 because one sample with clear measurement error was deleted). Baseline data were used in the initial performance evaluation only. To assess the possible effects of different calibrator value assignments made by the kit manufacturers, those from the Diasorin kit were analyzed in the automated IDS method on five separate occasions and recoveries quantified.

Participation in the DEQAS

DEQAS has more than 100 participating laboratories around the world (7, 9). Five samples containing varied concentration of 25(OH)D3 [occasionally 25(OH)D2] are sent as part of the scheme to all participating laboratories at three monthly intervals (7), and results were collated for within- and between-method comparisons. Quality assessment data for the present study were obtained on five occasions during the study period.

Statistical analysis

Agreement between 25(OH)D concentration determined by IDS with Diasorin and DEQAS mean of all methods was assessed using graphical plots as described by Bland and Altman (13) and relationships between methods quantified by Passing-Bablock regression (16). Lines of identity were drawn to indicate perfect agreement between the methods. Mean differences between the log-transformed values (log difference) and limits of agreement were calculated for both comparisons (1 and 2), and the average log difference was used as a correction factor to adjust for bias.

Formula

Formula

Formula
Harmonization factors obtained by comparison with both Diasorin (1) and DEQAS mean (2) were applied to the 781 samples measured by IDS for descriptive evaluation of monthly variation and hypovitaminosis D. Differences in average 25(OH)D concentrations between methods were compared using sign rank test. The proportion (with 95% confidence intervals) of values falling within three 25(OH)D thresholds (<25, <50, and <75 nmol/liter) were calculated for concentrations measured by Diasorin, IDS, and corrected IDS (1 and 2) to determine the influence of assay method and statistical correction on the estimated prevalence of hypovitaminosis D. Passing-Bablock comparisons were done by Microsoft Excel Analyse-it software (Analyse-it Software Ltd., Leeds, UK) and other statistical analyses using STATA (version 9; Stata Corp., College Station, TX).


   Results
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
Discussion
 References
 
Assay performance and sample stability

Performance of the automated method during the initial evaluation and through the subsequent 19-month survey period is shown in Table 1Go. Analysis of the functional sensitivity of the automated IDS EIA method showed a cutoff of 5 nmol/liter at a CV of 20%, and assay linearity was confirmed up to 155 nmol/liter (data not shown). We examined sample stability over five freeze thaw cycles and found no significant differences in 25(OH)D levels measured after each cycle in separate assays (Table 2Go). The final analyzer configuration allowed results to be obtained on sample volumes down to 75 µl.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Within- and between-assay imprecision with IDS EIA using three levels of quality control sera

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Sample stability in repeated freeze/thaw cycles

 
Agreement between automated IDS EIA and Diasorin

During the initial method evaluation, 142 patient samples from a pilot trial were analyzed by both automated IDS and Diasorin methods. Passing and Bablock regression analysis of this data set gave the relationship: IDS EIA = 0.80 (95% confidence interval 0.7, 0.9) x Diasorin – 1.2 (–6.7, 3.7). In samples collected during the 19-month fieldwork period from 781 members of the 1958 British birth cohort, there was a linear association between 25(OH)D concentration obtained by the two methods; however, 25(OH)D concentrations measured by IDS were on average –15.7 nmol/liter (SE 0.4) lower, compared with those obtained by Diasorin (P < 0.0001, Fig. 1AGo). The difference between the methods was dose dependent, and it increased with 25(OH)D concentration (Fig. 1BGo). As suggested by Bland and Altman (13), data were log transformed to adjust for this bias, and Fig. 1CGo shows that the difference between the methods was constant on a log scale. In 15 samples (1.9%), 25(OH)D concentration measured by Diasorin was lower, compared with that measured by IDS, and as shown in Fig. 1CGo in the extreme of low 25(OH)D concentrations, there was a cluster of nine observations with a large positive difference. In seven of these nine observations, 25(OH)D measured by Diasorin was less than 10 nmol/liter, and IDS produced no values less than 10 nmol/liter. Analyses of Diasorin calibrator value assignments by IDS suggested some overestimation by IDS at very low values (12.5 nmol/liter or lower), whereas recovery at 30 nmol/liter or higher concentrations was underestimated (Table 3Go).


Figure 1
View larger version (15K):
[in this window]
[in a new window]

  		FIG. 1. Agreement in 25(OH)D concentration measured by IDS EIA and Diasorin. A, IDS against Diasorin. B, Bland and Altman (13 ) comparison with absolute values. Horizontal lines give average difference with limits of agreement indicated by dashed lines. C, Bland and Altman comparison, log transformed. Horizontal lines give average difference with limits of agreement indicated by dashed lines. D, Harmonized IDS against Diasorin.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Evaluating Diasorin standards using calibrators from IDS EIA

 
Correction of the proportional bias by log transformation demonstrated that on a logarithmic scale the difference in 25(OH)D concentration between the methods was constant throughout the value range. Hence, as suggested by Bland and Altman (13), 25(OH)D concentration measured by IDS was adjusted for this bias using the log difference between Diasorin and IDS as a correction factor (Fig. 1DGo). In the analysis to calculate the correction factor (1), we excluded the cluster of nine outliers (described above). As seen in Fig. 1DGo, after correction there was a good correlation between corrected IDS and Diasorin, with the line of identity passing through the middle of the data. There was no difference between average 25(OH)D of corrected IDS and Diasorin (difference 0.6 nmol/liter, SE 0.4, P = 0.12).

Comparison with DEQAS

During the initial method evaluation and then throughout the survey, internationally distributed DEQAS samples (five samples per distribution) were assayed by both the Diasorin method and the automated IDS method. During the period of participation (January 2003 to April 2004), the number of laboratories providing results for DEQAS varied between 82 and 97. Most of the participating laboratories (63–69%) used methods based on RIA, 9–12% used EIA, 12–17% used chemiluminescence, 3–7% used HPLC, and the remainder used other methods.

There was a linear association between 25(OH)D measured by IDS and the average 25(OH)D of all laboratories participating in the DEQAS surveys (Fig. 2AGo). As in the comparison between IDS and Diasorin, concentrations by IDS were on average lower (difference –13.7 nmol/liter, SE 1.6, P < 0.0001), and there was a suggestion of dose-dependent bias, which demonstrated a greater negative difference with increasing 25(OH)D concentration (Fig. 2BGo). The difference between IDS and DEQAS mean was constant after log transformation (Fig. 2CGo), and dose-dependent bias was accommodated using the log difference between DEQAS mean and IDS as a harmonization factor (Fig. 2DGo). There was no significant difference in average 25(OH)D concentrations between IDS and DEQAS mean after harmonization (0.9 nmol/liter, SE 1.7, P = 0.58). Most of the laboratories participating in the DEQAS survey used Diasorin, and expectedly, there was good correlation between Diasorin and DEQAS mean with some variation around the line of identity but no evidence of systematic bias (Fig. 3Go).


Figure 2
View larger version (7K):
[in this window]
[in a new window]

  		FIG. 2. Agreement in 25(OH)D concentration measured by IDS EIA and mean of all laboratories participating in the DEQAS. A, IDS against DEQAS mean. B, Bland and Altman (13 ) comparison with absolute values. Horizontal lines give average difference with limits of agreement indicated by dashed lines. C, Bland and Altman comparison, log transformed. Horizontal lines give average difference with limits of agreement indicated by dashed lines. D, Harmonized IDS against DEQAS mean.

 

Figure 3
View larger version (6K):
[in this window]
[in a new window]

  		FIG. 3. Agreement in 25(OH)D concentration measured by Diasorin and mean of all laboratories participating in the DEQAS.

 
The influence of assay method on the prevalence of hypovitaminosis D

The pattern of monthly change 25(OH)D concentration was more consistent for IDS, compared with Diasorin (Fig. 4Go). Before statistical harmonization, the rates of hypovitaminosis D were higher (regardless of cutoff) when measured by IDS, compared with Diasorin. Harmonization of concentrations obtained by IDS according to either Diasorin or DEQAS mean resulted in identical prevalence rates, with only minor differences to those obtained by Diasorin (Table 4Go).


Figure 4
View larger version (27K):
[in this window]
[in a new window]

  		FIG. 4. Monthly variation in 25(OH)D concentration measured by Diasorin (A) and IDS EIA (B). 25(OH)D concentration measured by IDS is shown both for the absolute values (black) and harmonized by Diasorin (gray) and DEQAS mean (white) (n = 46 for all months, except for September 2003 n = 45).

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. Effect of assay and statistical modification on average 25(OH)D level and percentage classified as having hypovitaminosis D at three thresholds (n = 781)

 

   Discussion
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
References
 
In this paper we report the development and evaluation of an automated application of IDS EIA for determination of 25(OH)D in more than 7000 samples derived from the 1958 British birth cohort (12). The ease and overall consistency with which 25(OH)D can be measured with this automated EIA application made it ideally suited to large numbers of samples. We further confirmed the stability of 25(OH)D measured after repeated freeze-thaw cycles (17, 18). However, these findings highlight the urgent need for standardization of 25(OH)D assays. As we demonstrate, whereas showing relatively good general agreement, two assays may differ markedly in determining the proportion of population affected by hypovitaminosis D.

At all thresholds, the unmodified data suggested the prevalence of hypovitaminosis D measured by IDS EIA was markedly higher than estimates obtained by Diasorin. Harmonization of 25(OH)D levels measured by automated IDS either according to Diasorin or DEQAS mean corrected this imbalance, and the resulting prevalence rates were similar to those obtained by Diasorin. This comparison of methods demonstrates that if variation in 25(OH)D concentration by assay method is ignored, it may have crucial implications for both the comparison between observations from different (well designed) studies and the resulting public health messages. The underestimation of 25(OH)D concentrations reported in this study is likely to reflect a problem specific to IDS EIA, although issues in comparability have also been reported between other assays (4, 5, 6, 7, 8, 9, 19). Furthermore, we observed an important difference in calibrator value assignments between IDS and Diasorin at the extreme of low levels. To maintain the ranking of the concentrations measured by IDS intact, we applied the harmonization factor throughout the value range, including the low concentrations. After harmonization, the prevalence of 25(OH)D less than 25 nmol/liter (commonly used cutoff for vitamin D deficiency) was comparable, whether measured by IDS or Diasorin. However, the reported differences in calibrator value assignments between IDS and Diasorin suggests that direct comparisons should not be made at the more extreme of low concentrations. These findings clearly emphasize the need of method standardization at the level of kit manufacturers rather than data users.

Average 25(OH)D concentrations obtained by Diasorin were very close to DEQAS mean, as would be expected, given that most laboratories used RIA-based methods (7). Diasorin has the drawback of being relatively labor intensive and showing variation in 25(OH)D level by the operator carrying out the assay (8, 17). In routine quality control evaluations, the laboratory that carried out the 25(OH)D analyses for our study achieves a CV of only 19.5% for Diasorin when assays are carried out by multiple operators, improving to 9.6% with a single operator. In this respect, the automated IDS performed better than single-operator Diasorin (CV ≤ 8.5% during the survey). The better consistency with IDS, compared with manual Diasorin, was also evident in the pattern of seasonal variation in 25(OH)D concentrations, which was markedly smoother when measured by the automated IDS. For the purposes of the present large survey, selection of an automated method clearly provided marked benefits because it would have been impractical to carry out the required large number of assays using a single operator.

Compared with the HPLC method, Diasorin has been in some studies reported to overestimate vitamin D3 (the naturally most abundant form) and underestimate the synthetic D2 (6, 7), although not all comparisons have shown these differences (20, 21). Underestimation of D2 is a limitation with IDS EIA because according to manufacturers the assay detects only 75% of this metabolite (22). The main source of vitamin D2 in the cohort is likely to be multivitamin supplements because single vitamin D products were not available over the counter in the United Kingdom during the survey period (12). In our study, only 11% of the cohort members used vitamin supplements containing vitamin D (12). Given the small proportion of supplement users and that supplements in the United Kingdom often contain D3 and not D2, underestimation in 25(OH)D concentrations due to failure to account fully for vitamin D2 intake is likely to be relatively small. This is supported by the clear association between 25(OH)D concentrations and use of vitamin D supplements observed in the current study (12).

Some further limitations need to be considered in relation to these findings. The DEQAS only describes the relative performance among the participating laboratories, whereas it does not provide evidence for how accurate any single method is. Furthermore, we did not repeat the analyses with a reference method such as the HPLC, but carried out a descriptive comparison with Diasorin RIA. The reason why we chose to compare primarily against Diasorin, was that it is the method that has been most commonly used in epidemiological surveys, including the Third National Health and Nutrition Examination Survey (23). The proposed thresholds for hypovitaminosis D are determined on the basis of previous studies using various assays (24). Harmonization of 25(OH)D concentrations according to DEQAS mean has been previously done as parts of efforts aiming to identify optimal 25(OH)D concentrations for bone health (25). The comparisons of IDS with Diasorin and DEQAS mean were made to ensure that results obtained by IDS EIA are comparable with other assays currently in use, to evaluate and quantify the differences, not to suggest superiority for one method over the other. The primary purpose for measuring 25(OH)D in the British 1958 birth cohort was to obtain information on the current situation with hypovitaminosis D in Britain (12), and as we show in this paper, the estimated prevalence rates of hypovitaminosis D using the harmonized IDS are comparable with the estimates that would have been obtained using Diasorin at all thresholds.

In summary, there are marked differences in 25(OH)D concentrations between assays (3, 4, 5, 6, 7, 8, 9), which, as demonstrated in this study, may have important implications for the determination of hypovitaminosis D and the resulting public health messages (10). We used a statistical method (13) to overcome the problem of assay variation, and this enabled us to use previously suggested cutoffs for hypovitaminosis D (12). The role of DEQAS is crucial in providing an international reference against which harmonization factors can be constructed and in allowing a quantification of differences in 25(OH)D concentrations arising from different methodologies. However, as has been highlighted (4), a sustainable long-term solution to the problem of 25(OH)D assay heterogeneity will require method standardization.


   Acknowledgments
 
We thank Professor Reinhold Vieth (Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada) for advice on 25(OH)D assays and Mrs. Marie-Claude Fawcett (Royal Victoria Infirmary, Newcastle upon Tyne Hospitals National Health Service Trust, United Kingdom) for carrying out the 25(OH)D assays. DEQAS data were used with permission from the organizer.


   Footnotes
 
The vitamin D substudy was funded by The BUPA Foundation, and data collection for the 45-yr biomedical survey by the U.K. Medical Research Council (Grant G0000934). E.H. is Department of Health (United Kingdom) Public Health Career Scientist. Research at the Institute of Child Health and Great Ormond Street Hospital for Children National Health Service Trust benefits from R&D funding received from the National Health Service Executive.

Disclosure Statement: There are no conflicts of interest with this paper.

First Published Online August 28, 2007

Abbreviations: CV, Coefficient of variation; DEQAS, Vitamin D External Quality Assessment Scheme; EIA, enzyme immunoassay; 25(OH)D, 25-hydroxyvitamin D.

Received June 8, 2007.

Accepted August 22, 2007.


   References
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
 References
 

  1. Holick MF 2004 Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. Am J Clin Nutr 80(Suppl 6):1678S–1688S
  2. Zittermann A, Schleithoff SS, Koerfer R 2005 Putting cardiovascular disease and vitamin D insufficiency into perspective. Br J Nutr 94:483–492[CrossRef][Medline]
  3. Zerwekh JE 2004 The measurement of vitamin D: analytical aspects. Ann Clin Biochem 41(Pt 4):272–281
  4. Binkley N, Krueger D, Cowgill CS, Plum L, Lake E, Hansen KE, DeLuca HF, Drezner MK 2004 Assay variation confounds the diagnosis of hypovitaminosis D: a call for standardization. J Clin Endocrinol Metab 89:3152–3157[Abstract/Free Full Text]
  5. Hollis BW 2004 The determination of circulating 25-hydroxyvitamin D: no easy task. J Clin Endocrinol Metab 89:3149–3151 (Editorial)
  6. Glendenning P, Taranto M, Noble JM, Musk AA, Hammond C, Goldswain PR, Fraser WD, Vasikaran SD 2006 Current assays overestimate 25-hydroxyvitamin D(3) and underestimate 25-hydroxyvitamin D(2) compared with HPLC: need for assay-specific decision limits and metabolite-specific assays. Ann Clin Biochem 43(Pt 1):23–30
  7. Carter GD, Carter R, Jones J, Berry J 2004 How accurate are assays for 25-hydroxyvitamin D? Data from the international vitamin D external quality assessment scheme. Clin Chem 50:2195–2197[Free Full Text]
  8. Hollis BW, Horst RL 2007 The assessment of circulating 25(OH)D and 1,25(OH)(2)D: where we are and where we are going. J Steroid Biochem Mol Biol 103:473–476[CrossRef][Medline]
  9. Carter GD, Jones JC, Berry JL 2007 The anomalous behaviour of exogenous 25-hydroxyvitamin D in competitive binding assays. J Steroid Biochem Mol Biol 103:480–482[CrossRef][Medline]
  10. Lips P 2007 Vitamin D status and nutrition in Europe and Asia. J Steroid Biochem Mol Biol 103:620–625[CrossRef][Medline]
  11. Hyppönen E, Power C 2006 Vitamin D status and glucose homeostasis in the 1958 British birth cohort: the role of obesity. Diabetes Care 29:2244–2246[Abstract/Free Full Text]
  12. Hyppönen E, Power C 2007 Hypovitaminosis D in British adults at age 45y: nationwide cohort study on dietary and lifestyle predictors. Am J Clin Nutr 85:860–868[Abstract/Free Full Text]
  13. Bland JM, Altman DG 1999 Measuring agreement in method comparison studies. Stat Methods Med Res 8:135–160[Abstract/Free Full Text]
  14. Butler NR, Bonham DG 1963 Perinatal mortality. Edinburgh, UK: Livingstone
  15. Power C, Elliott J 2006 Cohort profile: 1958 British birth cohort (National Child Development Study). Int J Epidemiol 35:34–41[Free Full Text]
  16. Passing H, Bablok W 1983 A new biometrical procedure for testing the equality of measurements from two different analytical methods. J Clin Chem Clin Biochem 21:709–720[Medline]
  17. Ersfeld DL, Rao DS, Body JJ, Sackrison Jr JL, Miller AB, Parikh N, Eskridge TL, Polinske A, Olson GT, MacFarlane GD 2004 Analytical and clinical validation of the 25 OH vitamin D assay for the LIAISON automated analyzer. Clin Biochem 37:867–874[CrossRef][Medline]
  18. Antoniucci DM, Black DM, Sellmeyer DE 2005 Serum 25-hydroxyvitamin D is unaffected by multiple freeze-thaw cycles. Clin Chem 51:258–261[Free Full Text]
  19. Ibrahim F, Parmentier C, Boudou P 2007 Divergence in classification of 25-hydroxyvitamin D status with respect to immunoassays. Clin Chem 53:363–364[Free Full Text]
  20. Hollis BW 2000 Comparison of commercially available (125)I-based RIA methods for the determination of circulating 25-hydroxyvitamin D. Clin Chem 46:1657–1661[Abstract/Free Full Text]
  21. Maunsell Z, Wright DJ, Rainbow SJ 2005 Routine isotope-dilution liquid chromatography-tandem mass spectrometry assay for simultaneous measurement of the 25-hydroxy metabolites of vitamins D2 and D3. Clin Chem 51:1683–1690[Abstract/Free Full Text]
  22. IDS OCTEIA kit literature 2002 Ref Document AC-48PL-A, draft 1. Boldon, Tyne-and-Wear, UK: IDS, Ltd.
  23. Looker AC, Dawson-Hughes B, Calvo MS, Gunter EW, Sahyoun NR 2002 Serum 25-hydroxyvitamin D status of adolescents and adults in two seasonal subpopulations from NHANES III. Bone 30:771–777[Medline]
  24. Bischoff-Ferrari HA, Giovannucci E, Willett WC, Dietrich T, Dawson-Hughes B 2006 Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health outcomes. Am J Clin Nutr 84:18–28[Abstract/Free Full Text]
  25. Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ, Vieth R 2005 Estimates of optimal vitamin D status. Osteoporos Int 16:713–716[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Ann Rheum DisHome page
K Atherton, D J Berry, T Parsons, G J Macfarlane, C Power, and E Hypponen
Vitamin D and chronic widespread pain in a white middle-aged British population: evidence from a cross-sectional population survey
Ann Rheum Dis, June 1, 2009; 68(6): 817 - 822.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
E. Orwoll, C. M. Nielson, L. M. Marshall, L. Lambert, K. F. Holton, A. R. Hoffman, E. Barrett-Connor, J. M. Shikany, T. Dam, J. A. Cauley, et al.
Vitamin D Deficiency in Older Men
J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1214 - 1222.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
E. Hypponen, B. J. Boucher, D. J. Berry, and C. Power
25-Hydroxyvitamin D, IGF-1, and Metabolic Syndrome at 45 Years of Age: A Cross-Sectional Study in the 1958 British Birth Cohort
Diabetes, February 1, 2008; 57(2): 298 - 305.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hyppönen, E.
Right arrow Articles by Gibb, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hyppönen, E.
Right arrow Articles by Gibb, I.
Related Collections
Right arrow Calcium and Bone Metabolism
Right arrow Metabolism

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals