Biomarkers of vitamin B-12 status in NHANES: a roundtable summary1–6
Elizabeth A Yetley, Christine M Pfeiffer, Karen W Phinney, Regan L Bailey, Sheena Blackmore, Jay L Bock,
Lawrence C Brody, Ralph Carmel, L Randy Curtin, Ramo ´n A Durazo-Arvizu, John H Eckfeldt, Ralph Green, Jesse F Gregory III,
Andrew N Hoofnagle, Donald W Jacobsen, Paul F Jacques, David A Lacher, Anne M Molloy, Joseph Massaro, James L Mills,
Ebba Nexo, Jeanne I Rader, Jacob Selhub, Christopher Sempos, Barry Shane, Sally Stabler, Patrick Stover, Tsunenobu Tamura,
Alison Tedstone, Susan J Thorpe, Paul M Coates, Clifford L Johnson, and Mary Frances Picciano
status biomarkers in NHANES took place in July 2010. NHANES
stopped measuring vitamin B-12–related biomarkers after 2006.
The roundtable reviewed 3 biomarkers of vitamin B-12 status used
in past NHANES—serum vitamin B-12, methylmalonic acid (MMA),
and total homocysteine (tHcy)—and discussed the potential utility
of measuring holotranscobalamin (holoTC) for future NHANES.
of the measurement procedures and reference methods and materials
that past NHANES used or that are available for future NHANES.
Roundtable members supported reinstating vitamin B-12 status mea-
sures in NHANES. They noted evolving concerns and uncertainties
regarding whether subclinical (mild, asymptomatic) vitamin B-12
deficiency is a public health concern. They identified the need for ev-
ical vitamin B-12 deficiency and adverse health outcomes as well as
circulating vitamin B-12 (serum vitamin B-12 or holoTC) and one
functional biomarker (MMA or tHcy) in NHANES. The inclusion
of both serum vitamin B-12 and plasma MMA, which have been
associated with cognitive dysfunction and anemia in NHANES
able, and National Institute of Standards and Technology reference
materials are available or in development for serum vitamin B-12 and
MMA.Am J Clin Nutr 2011;94(suppl):313S–21S.
The NHANES are a valuable source of data on the nutritional
and health status of the US population. The surveys require
highly accurate measurement procedures to ensure that their
prevalence estimates are comparable with commonly used
cutoffs of nutrient adequacy and safety, and that trends in these
estimates over timeareattributable to realchanges in nutritional
Currently, NHANES does not measure serum vitamin B-12 or
any other vitamin B-12–related biomarkers. However, NHANES
II (1976–1980), the second half of NHANES III (1991–1994),
and NHANES 1999–2006 measured serum vitamin B-12 concen-
trations. Furthermore, NHANES III (1988–1994) and NHANES
1999–2004 measured methylmalonic acid (MMA) concentrations,
and NHANES III (1991–1994) and NHANES 1999–2006 mea-
sured total homocysteine (tHcy) concentrations.
The National Center for Health Statistics of the Centers for
Disease Control and Prevention and the Office of Dietary Sup-
panel on July 15–16, 2010, in Rockville, MD, to review the mea-
surement of biomarkers of folate and vitamin B-12 status in
NHANES. In this article, we summarize the roundtable’s dis-
cussion, which focused on the following: 1) public health con-
siderations for including vitamin B-12–related biomarkers in
NHANES and 2) the quality of the vitamin B-12–related mea-
surement procedures and reference methods and materials that
1From the Office of Dietary Supplements, National Institutes of Health,
Bethesda, MD (EAY, RLB, CS, PMC, and MFP); the National Center for
Environmental Health, Centers for Disease Control and Prevention, Atlanta,
GA (CMP); the Analytical Chemistry Division, National Institute of Standards
and Technology, Gaithersburg, MD (KWP); the Haematology Department,
United Kingdom National External Quality Assessment Service, Sutton Cold-
field, United Kingdom (SB); the Department of Pathology, Stony Brook Uni-
versity Medical Center, Stony Brook, NY (JLB); Genome Technology Branch,
National Human Genome Research Institute, National Institutes of Health,
Bethesda, MD (LCB); the Department of Medicine, New York Methodist
Hospital, Brooklyn, NY (RC); Weill Medical College, Cornell University,
New York, NY (RC); the National Center for Health Statistics, Centers for
Disease Control and Prevention, Hyattsville, MD (LRC, DAL, and CLJ); the
Department of Preventive Medicine and Epidemiology, Loyola University
Chicago Stritch School of Medicine, Maywood, IL (RAD-A); the Department
of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis,
MN (JHE); the Department of Medical Pathology and Laboratory Medicine
and Medicine, University of California Davis, Sacramento, CA (RG); the Food
Science and Human Nutrition Department, University of Florida, Gainesville,
FL (JFG); the Department of Laboratory Medicine, University of Washington,
Seattle, WA (ANH); the Department of Molecular Medicine, Cleveland Clinic
Lerner College of Medicine, Case Western Reserve University, Cleveland, OH
(DWJ); the Department of Cell Biology, Lerner Research Institute, Cleveland
Clinic, Cleveland, OH (DWJ); Jean Mayer US Department of Agriculture
Human Nutrition Research Center on Aging, Tufts University, Boston, MA
(PFJ and JS); the Department of Clinical Medicine, School of Medicine,
Trinity College, Dublin, Ireland (AMM); Department of Biostatistics, Boston
University School of Public Health, Boston, MA (JM); Eunice Kennedy Shriver
National Institute of Child Health and Human Development, National Institutes
of Health, Bethesda, MD (JLM); the Department of Clinical Biochemistry,
Aarhus University Hospital, Aarhus, Denmark (EN); the Center for Food
Safety and Applied Nutrition, Food and Drug Administration, College Park,
Am J Clin Nutr 2011;94(suppl):313S–21S. Printed in USA. ? 2011 American Society for Nutrition
NHANES uses or that are available for future NHANES. Other
articles in this supplement issue summarize general background
information on biomarker measurement and use issues and the
roundtable’s review of folate status biomarkers (1–3). In an in-
troductory article, we describe how the individual articles in this
supplement fit within the overall roundtable process (4). We also
describe the relevance of the roundtable review and dialogue to
clinical and research settings outside the NHANES context.
B-12 assessment, clinical laboratory science, and biostatistics and
10 scientists from US government agencies that generate or use
NHANES data and develop reference methods and materials for
The roundtable reviewed 3 biomarkers of vitamin B-12 status
usedin NHANES:vitamin B-12,MMA, and tHcy.This article,in
among US population groups (5–7). The roundtable also consid-
ered holotranscobalamin’s (holoTC’s) possible utility as an in-
dicator of vitamin B-12 status for future NHANES (8, 9).
The roundtable reviewed vitamin B-12–related public health
and measurement quality issues in the context of the NHANES
mission and capabilities. The experts did not consider broader
research and clinical application issues for these biomarkers or
differences in themeasurementprocedures thatresearch, clinical,
and commercial laboratories use. The roundtable identified and
discussed the scientific issues involved in deciding whether to
ease of use, we used the more common terminology of “vitamin
B-12” in place of the more precise terminology of “cobalamin”
(9). When we used serum vitamin B-12 data in this article, we
converted the values to pmol/L (1.0 pg/mL or ng/L = 0.7378
pmol/L). For plasma MMA, we converted the values to nmol/L
(1.0 ng/mL or lg/L = 8.468 nmol/L).
MONITORING OF VITAMIN B-12 STATUS IN NHANES
The roundtable focused on 2 key vitamin B-12–related mon-
itoring issues: 1) the public health need for assessing vitamin
B-12 status and 2) the usefulness of several biomarkers in con-
ducting these assessments. Ralph Carmel outlined key issues for
the roundtable to consider (9). Regan Bailey provided data on
vitamin B-12 status assessments for adults in NHANES.
Public health need for monitoring of vitamin B-12 status
The roundtable considered whether the primary purpose of
deficiency characterized by megaloblastic anemia and/or neuro-
more common subclinical (mild, asymptomatic) vitamin B-12
deficiency, defined by abnormal concentrations of vitamin B-12–
related biomarkers [also referred to as SCCD or subclinical co-
balamin deficiency (9)]. Approximately 1–2% of the elderly US
population has clinical vitamin B-12 deficiency, and ’10–20%
has subclinical deficiency (10).
conditions such as pernicious anemia, the roundtable found the
question posed by Carmel as to whether monitoring the more
common subclinical deficiency has public health merit (9) to be
a helpful anchor point for their overall dialogue. The common
clinical deficiency lacks confirmatory evidence (9). Subclinical
food-bound cobalamin or unidentified causes) than does clinical
regain normal vitamin B-12 status or experience no further de-
terioration after several years, raising questions about the public
health concerns with subclinical deficiency (9–11).
Subclinical vitamin B-12 deficiency per se may adversely
affect cognitive or neurologic function. Results from some, but
not all, observational studies show associations between vitamin
B-12 deficiency and cognitive impairment (12–17). The round-
table noted, however, that no clinical trials have confirmed a
causal relation between subclinical vitamin B-12 deficiency and
adverse outcomes. Therefore, the roundtable agreed that the
public health implications of subclinical vitamin B-12 deficiency
is of concern and warrants further evaluation.
Biomarkers of vitamin B-12 status
The roundtable considered the usefulness of available bio-
MD (JIR); the Department of Nutritional Science and Toxicology, University
of California, Berkeley, CA (BS); the Department of Medicine/Division of
Hematology, University of Colorado Health Sciences Center, Aurora, CO
(SS); the Division of Nutritional Sciences, Cornell University, Ithaca, NY
(PS); the Department of Nutrition Sciences, University of Alabama at Bir-
mingham, Birmingham, AL (TT); Nutrition Science and Delivery, Depart-
ment of Health, London, United Kingdom (AT); and the National Institute
for Biological Standards and Control, Herts, United Kingdom (SJT).
2The authors dedicate this article and the roundtable to Mary Frances Pic-
ciano, Office of Dietary Supplements, National Institutes of Health, who
was the driving force behind this effort prior to her death in August 2010.
3Any opinions, findings, conclusions, or recommendations expressed in
this publication are those of the authors and do not necessarily reflect the
views of the Centers for Disease Control and Prevention (CDC), the NIH, the
US Department of Health and Human Services (DHHS), or the National
Institute of Standards and Technology (NIST) or any of the authors’ related
affiliations. Certain commercial equipment, instruments, or materials are iden-
tified in this article to specify adequately the experimental procedure. Such
identification does not imply recommendation or endorsement by the NIST, the
CDC, the NIH or the DHHS, nor does it imply that the materials or equipment
identified are necessarily the best available for the purpose.
4Supported by the National Center for Health Statistics, Centers for Dis-
ease Control and Prevention, and Office of Dietary Supplements, National
Institutes of Health.
5Address reprint requests to PM Coates, Office of Dietary Supplements,
National Institutes of Health, 6100 Executive Boulevard, Room 3B01, MSC
7517, Bethesda, MD 20892-7517. E-mail: firstname.lastname@example.org.
6Address correspondence to EAYetley (retired), Office of Dietary Supple-
ments, National Institutes of Health, 6100 Executive Boulevard, Room 3B01,
MSC 7517, Bethesda, MD 20892-7517. E-mail: email@example.com.
First published online May 25, 2011; doi: 10.3945/ajcn.111.013243.
YETLEY ET AL
usefulness depends on the status component being measured and
whether (and, if so, how) factors unrelated to vitamin B-12 status
affect these biomarkers. The roundtable also discussed whether
sample (the general US population) and for the NHANES mea-
Currently availablebiomarkers canbe categorizedas those that
to adequacy. MMA and tHcy, which are functional (metabolic)
of vitamin B-12 are inadequate; tHcy also accumulates with in-
adequate amounts of folate and, to a lesser degree, riboflavin and
vitamin B-6. These functional measures are useful for identifying
B-12 status. Information from both types of biomarkers can yield
from only one vitamin B-12 status category (9).
(8–10, 18–20). Impaired renal function alters MMA and tHcy and,
to a lesser extent, holoTC and serum vitamin B-12 concentrations.
Other factors, such as individual genetic variation, disease con-
ditions, prescription and other drug use, and pregnancy can also
affect the biomarkers. Measurement procedure errors and other
technological problems produce spuriously high or low biomarker
concentrations (9). These factors adversely affect the biomarkers’
sensitivity and specificity and result in false-positive or false-
negative classifications of vitamin B-12 deficiency.
These classification errors are particularly important for sur-
veys such as NHANES that cannot conduct the follow-up di-
agnostic testing required for vitamin B-12 deficiency diagnosis
and treatment (9). For example, high MMA and tHcy concen-
trations occur across all serum vitamin B-12 concentrations,
although they are most prevalent in people with the lowest vi-
tamin B-12 concentrations (9). Clinical data show that ’2.5–
5.2% of patients with diagnosed vitamin B-12 deficiency have
serum vitamin B-12 concentrations above a commonly used cut-
off for low values (148 pmol/L), and an even higher proportion of
persons with abnormal MMA or tHcy concentrations have serum
vitamin B-12 concentrations .148 pmol/L (9). Approximately
20% of persons with low serum vitamin B-12 concentrations have
neither clinical nor metabolic signs of vitamin B-12 deficiency.
The roundtable agreed that because of problems with these bio-
markers’ sensitivity and specificity, measuring a circulating bio-
marker (vitamin B-12 or holoTC) and a functional biomarker
(MMA or tHcy) concurrently is preferable to measuring only
one biomarker in NHANES.
The roundtable also reviewed the relevance of common cutoffs
for the NHANES population. The vitamin B-12 cutoffs that sci-
entists and clinicians use generally are consistent with 2 or 3 SDs
from a reference range’s mean. A cutoff value consistent with 2
SDs is ’200 pmol/L, and a value of 148 pmol/L is consistent with
3 SDs (9, 18). A cutoff of 148 pmol/L misses 3–5% of patients
with clinical deficiency, whereas ’200 pmol/L identifies all pa-
tients but produces more false-positive results. Systematic bias
and spurious results affected some measurement procedures used
in the past to derive reference ranges (9). The cutoff used affects
NHANES prevalence estimates. Using cutoffs similar to those
used in the published literature, Regan Bailey (personal commu-
nication, 2011) found a prevalence (least-squared percentage 6
SE) of 2.8 6 0.2% for serum vitamin B-12 concentrations ,148
pmol/L, 10.5 6 0.4% for concentrations ,198 pmol/L, and 25.6 6
0.6% at ,258 pmol/L in an NHANES sample of people aged
?60 y with impaired but controlled renal function.
(9). Researchers have based MMA cutoff values on an approxi-
a percentile of a population-based reference range, or on fully
cobalamin-replete subjects (12, 18, 21). The cutoff value used
?60 y in whom impaired renal function was controlled, Regan
Bailey found that a cutoff .376 nmol/L resulted in a prevalence
of 2.4%, whereas a cutoff .271 nmol/L produced a prevalence
of 5.9% (RL Bailey, personal communication, 2011).
The roundtable discussed the urgent need to base cutoffs for
biomarker concentrations on adverse outcome occurrences. The
12 deficiency, which is the predominant type of vitamin B-12 de-
ficiency in NHANES populations because it is asymptomatic and
research has not confirmed a causal relation between subclinical
requires clinical trials and accurate and reliable measurement pro-
cedures. Cross-sectional studies, such as NHANES, and other ob-
helpgenerate hypotheses and examine relationsbetweenvariables.
MEASURES OF VITAMIN B-12–RELATED BIOMARKERS
The roundtable reviewed the quality and appropriateness of
Except for 2 special postsurvey projects at Tufts University that
used surplus sera to measure MMA and tHcy in the second phase
ofthe Centersfor Disease Controland Prevention measuredserum
vitamin B-12 and plasma MMA and tHcy in NHANES. Christine
Pfeiffer reviewed the NCEH laboratory procedures and results for
Measurement of serum vitamin B-12 in NHANES
NHANES used the Bio-Rad Quantaphase II radioimmuno-
assay (RIA) kit for (Bio-Rad Laboratories, Hercules, CA) all
serum vitamin B-12 measures except for the 1976–80 survey,
which used the Quantaphase I kit (Table 1) (24–28). In 2007, the
manufacturer discontinued the Bio-Rad RIA. The loss of this assay
kit and the finding that serum vitamin B-12 concentrations col-
lected from NHANES III (1991–1994) to NHANES 2006 had
not changed appreciably (1) resulted in a decision to drop mea-
sures of vitamin B-12 status after NHANES 2006.
pmol/L (29). Quality-control (QC) pools were stable and re-
producible across all surveys. For reference material 81/563 from
the National Institute for Biological Standards and Control
VITAMIN B-12–RELATED BIOMARKERS IN NHANES
materials.aspx), the NCEH obtained a mean (6SD) value of
234 6 14 pmol (n = 27) from 1994 to 2006 compared with an
NIBSC-certifiedconcentration of 236pmol/ampule. Over several
years in the College of American Pathologists (http://www.cap.
org/apps/cap.portal) proficiency testing program, the NCEH re-
sults were ’20% higher than the College of American Patholo-
gists all-laboratory trimmed mean; the Bio-Rad RIA had no peer
group using this measurement procedure with which to directly
Because the NCEH anticipated that NHANES might consider
reinstating vitamin B-12 measures at a later date, it evaluated the
Roche Elecsys 170 chemiluminescent measurement procedure
(Roche, Basel, Switzerland). The result was 373 pmol/ampule for
the NIBSC 03/178 reference material, with an assigned reference
value of 354 pmol/ampule. When the NCEH fortified a serum pool
with this reference material, complete recovery resulted (103%).
reference material (SRM) 3951 (30), which is under de-
velopment. The NISTwill use resultsfrom higher-order reference
measurement procedures to assign this SRM’s target values.
NCEH-measured mean (6SD) vitamin B-12 concentrations were
42.6 6 5.8 (CV: 13.7%), 114.4 6 6.0 (CV: 5.3%), and 283.1 6
7.3 pmol/L (CV: 2.6%) (n = 12) for this reference material. The
NCEH obtained within-day CVs (n = 10) of 0.9–2.1% for its low
and high (’175 and 825 pmol/L) QC pools. Between-day CVs
over 5 mo (n = 22) were 3.8–6.5% for 5 QC pools with concen-
trations of ’200–900 pmol/L. The manufacturer-specified limit of
detection for this serum vitamin B-12 measurement procedure
was 22 pmol/L.
samples it had previously measured with the Bio-Rad Quanta-
phase II RIA by using the Roche E-170 measurement procedure
to determine the procedures’ comparability. The median was 390
the Pearson correlation coefficient was 0.98 over a range of ’45–
1100 pmol/L. A log transformation removed the distribution’s
skewness with a Deming regression fit of log vitamin B-12Roche=
0.97 · log vitamin B-12Bio-Rad+ 0.10 pmol/L and showed a small
but significant constant and proportional bias (Figure 1A). The
Bland-Altman bias plot showed a mean relative bias of 10.7% for
the Roche procedure compared with the Bio-Rad procedure, with
95% limits of agreement from 25.6% to 27% (Figure 1B).
Measurement of plasma MMA in NHANES
The NCEH used gas chromatography/mass spectrometry (GC/
MS) to measure plasma MMA in NHANES (Table 2) (31–34),
resulting in long-term CVs (NHANES 1999–2004) of 3–8% for
low QC pools (’100–500 nmol/L), 2–4% for medium QC pools
(1000–2000 nmol/L), and 2.0–3.5% for high QC pools (’10,000
nmol/L). QC pools were stable and reproducible across all sur-
veys. As a participant in the Danish External Quality Assess-
ment program since 2000, the NCEH’s MMA results have
consistently been within the acceptance interval that the pro-
gram calculated from biological variation (621% of the pro-
gram mean). The NCEH’s results were, on average, 6% lower
than the program mean.
If NHANES reinstates MMA measures in future surveys, the
NCEH could use their GC/MS measurement procedure because
the NCEH has maintained proficiency in this procedure despite
its absence from recent NHANES. However, the NCEH is
considering using a liquid chromatography–mass spectrometry
(LC-MS/MS) procedure similar to the one that several labora-
tories haverecently used(35–37). The roundtable encouraged the
NCEH to use LC-MS/MS for future analyses.
Measurement of tHcy in NHANES
The measurement procedures, QC data, and survey compa-
are summarized in Table 3. The NCEH observed long-term
(NHANES 1999–2004) CVs of 3–5% for multiple QC pools
covering a tHcy concentration range of ’6–27 lmol/L (45). QC
pools were stable and reproducible in all surveys. The NCEH
Abbott AxSym (Abbott Diagnostics, Abbott Park, IL) results
were within 10% of the NIST SRM 1955 certified values, 9%
higher for the low level (’4 lmol/L), 1% higher for the medium
level (’9 lmol/L), and 6% lower for the high level (’18 lmol/L).
As a participant in the Danish external quality assurance program
in 2000–2008, the NCEH consistently obtained tHcy results within
the program’s acceptance interval on the basis of biological vari-
ation (617% of the program mean). The NCEH’s results were, on
average, 0.2% higher than the program mean.
Possible measurement of holoTC in NHANES
Ebba Nexo (8) described several procedures for assessing
holoTC, including a new enzyme-linked immunosorbent assay
procedure on the Abbott AxSym analyzer platform. This pro-
cedure is easier to use than previous ones and can measure
holoTC directly without sample pretreatment (46). In a com-
parison study in Europe, all available methods gave similar values
(47). Enzyme-linked immunosorbent assay measurement had a 4–
5% imprecision rate and a 5–20 pmol/L method-dependent de-
NIST REFERENCE METHODS AND MATERIALS
NIST reference methods and materials for use in NHANES to
achieve traceability of NHANES measurements to a higher-order
reference method and materials. Karen Phinney discussed the
development of reference methods and materials for vitamin
B-12, tHcy, and MMA at the NIST.
Vitamin B-12 biomarker measurement in NHANES1
SurveyMeasurement procedure Matrix Laboratory Population age
Bio-Rad QP I
Bio-Rad QP II
Bio-Rad QP II
Bio-Rad QP II
Bio-Rad QP II
Bio-Rad QP II
6 mo–74 y, subset
1QP, Quantaphase (Bio-Rad Laboratories, Hercules, CA); NCEH, Na-
tional Center for Environmental Health, Centers for Disease Control and
YETLEY ET AL
NIST methods and materials for vitamin B-12 and MMA
Currently, the NIST has no SRM with certified values for vi-
tamin B-12 or MMA, although it provides information values for
vitamin B-12 using RIA in SRM 1955 (48). The NIST is de-
veloping SRM 3951 (30) for serum vitamin B-12 with target pools
of 74 pmol/L (100 pg/mL), 148 pmol/L (200 pg/mL), and 332
pmol/L (450 pg/mL). The 332-pmol/L pool is “normal” serum,
of its naturally occurring vitamin B-12.
Unique challenges in developing a higher-order vitamin B-12
measurement procedure include the forms of vitamin B-12 to
measure. One challenge is whether converting all forms to cy-
anocobalamin and measuring this compound to determine total
serum vitamin B-12 best meets user needs. Also, do users need to
evaluate the different biologically relevant species? Another
challenge is the low concentrations that will be encountered in
some individuals with low vitamin B-12 status (’100 pmol/L or
less) of serum vitamin B-12. To account for potential analyte
losses during sample preparation, a stable-isotope–labeled in-
ternal standard should be used, but chemical synthesis of an
isotopically labeled form of vitamin B-12 is not currently fea-
sible. A labeled form of vitamin B-12 would be the ideal internal
standard because it should behave similarly to the unlabeled
endogenous form, including binding to the 2 circulating vitamin
B-12 binding proteins (haptocorrin and transcobolamin). Alter-
native options might be a labeled compound produced by bac-
teria or partial chemical synthesis incorporating isotopically
labeled precursor compounds.
The NIST has not found a good internal standard candidate
among other structurally related compounds. On the basis of
preliminary work using LC-MS/MS, the NIST has identified some
vitamin B-12 species that it could quantify. Remaining challenges
include obtaining the needed sensitivity, developing an appro-
priate quantitation mechanism, and obtaining sufficient precision
to certify concentration values in the reference materials. The
roundtable suggested developing a valid LC-MS/MS method to
measure total cobalamin after converting all species to cyanoco-
balamin. Characterization of vitamin B-12 species is a long-term
goal, but a need exists now for an SRM for total cobalamin.
The NIST is currently developing MMA reference methods
and materials. Phinney reported that GC/MS and LC-MS/MS
procedures using derivatized or underivatized compounds work
well. The NIST is developing new reference materials for SRM
(MMA concentration of ’170 nmol/L). The NIST’s SRM 1950
(50) materials are also appropriate (MMA of ’300 nmol/L).
However, SRM 3591 (30), which is under development, is not
appropriate (MMA of ’35 nmol/L).
NIST methods and materials for tHcy
NIST SRM 1955 has certified values for mean (6expanded
uncertainty) tHcy concentrations of 3.98 6 0.18 lmol/L, 8.85 6
0.60 lmol/L, and 17.7 6 1.1 lmol/L (48, 51–53). Level 2 is a
naturally occurring serum material, level 1 is diluted, and level 3 is
fortified to achieve target concentrations. In a commutability study
with 20 fresh-frozen samples and a limited concentration range,
SRM 1955 was commutable across 14 commercial tHcy immuno-
and enzymatic assays (54). The Joint Committee for Traceability in
Laboratory Medicine database lists the NIST GC-MS and LC-MS/
MS methods as reference measurement procedures (55).
ROUNDTABLE DIALOGUE: VITAMIN B-12–RELATED
BIOMARKERS IN NHANES
The roundtable members were generally satisfied with the mea-
surement and reference material issues and therefore focused their
FIGURE 1. Crossover study (n = 280) comparing serum vitamin B-12
(B12) results using 2 different measurement procedures. A: Scatter plot with
Deming regression fit of log-transformed vitamin B-12 concentrations as
measured by the Roche E-170 procedure (Roche, Basel, Switzerland). com-
pared with the Bio-Rad Quantaphase II radioimmunoassay (Bio-Rad Labo-
ratories, Hercules, CA). The Deming equation was: log B12Roche= 0.97 ·
log B12Bio-Rad+ 0.10 pmol/L. The SE was 0.012 for the slope and 0.025 for
the intercept. B: Bland-Altman difference plot of vitamin B-12 concentra-
tions between the Roche E-170 procedure and the Bio-Rad Quantaphase II
radioimmunoassay. Each data point represents the relative difference be-
tween the 2 assays as a function of the mean between the 2 assays. The
mean relative bias for the Roche procedure compared with the Bio-Rad
procedure was 10.7%, and the 95% limits of agreement were from 25.6%
VITAMIN B-12–RELATED BIOMARKERS IN NHANES
discussion on issues relating to reinstating vitamin B-12 status bio-
markers in future NHANES.
Reinstatement of vitamin B-12 biomarkers in NHANES
Overall, the roundtable experts supported reinstating vitamin
B-12–related biomarker measurement in future NHANES and
discussed possible reasons to justify its inclusion in future
NHANES. The roundtable noted that interest in monitoring vi-
tamin B-12 status began with folic acid fortification of cereal
grains in the United States in 1996, because this fortification
might mask hematologic symptoms of vitamin B-12 deficiency,
resulting in diagnosis and treatment delays that would promote
the progression of mild vitamin B-12 insufficiency to more clin-
ically severe deficiency. However, before and after fortification,
median serum vitamin B-12 concentrations (343–356 pmol/L) and
(2.7–4.8%) did not change appreciably (29). Only 2% of the US
population had high MMA concentrations indicating inadequate
vitamin B-12 status (.370 nmol/L) in NHANES 1999–2000,
although prevalence in persons aged ?60 y was 7% (56).
The roundtable noted that the lack of change in serum vitamin
the assumption that mild vitamin B-12 deficiencyis verylikely to
progress to more severe clinical deficiency may be erroneous (9).
Some roundtable experts noted that the ratio of high folate con-
about potential adverse effects of folic acid fortification. Selhub
et al (15) observed anemia and cognitive impairment ’5 times
more often in the group with elevated folate status (.59 nmol/L)
or MMA .210 nmol/L) than in the group with normal vitamin
B-12 and folate status.
The roundtable noted that the evolving observational evidence
that suggests an association between subclinical vitamin B-12
deficiencies and adverse effects is of concern. However, pending
mild vitamin B-12 deficiency and adverse effects, whether mild
vitamin B-12 deficiencyimposes a public health risk is uncertain.
Beginning in 2011, NHANES will include 3 cognitive function
measures. Several experts thought that these new data would be
useful to explore previously observed associations with the more
questioned the relevance to vitamin B-12 deficiency of cognitive
function tests designed to detect Alzheimer disease or the con-
tinued need for more NHANES hypothesis-generating research,
given that several observational studies have already identified
associations between vitamin B-12 status and cognitive function.
need for population-based prevalence estimates as new data be-
come available from clinical trials. NHANES data could also
provide useful information on the quantitative relation between
intakes and biomarker concentrations, although malabsorption
problems with some vitamin B-12 insufficiencies would compli-
status by a blood test (eg, intrinsic factor antibody) in peoplewith
low vitamin B-12 status to distinguish patients with pernicious
The roundtable considered population-based changes beyond
the aging US population that could affect vitamin B-12 status and
would justify including vitamin B-12 biomarkers in future
NHANES. More people use proton pump inhibitors and hista-
mine 2 receptor antagonists (H2 blockers) that may decrease
vitamin B-12 absorption; these are now available over-the-counter
(57, 58). Other changes, such as the common use of antibiotics in
Methylmalonic acid measurements in NHANES1
SurveyMeasurement procedure Matrix LaboratoryPopulation age
30–39 y, ?65 y
1GC/MS, gas chromatography/mass spectrometry; Tufts University, Jean Mayer US Department of Agriculture
Human Nutrition Research Center on Aging, Boston, MA; NCEH, National Center for Environmental Health, Centers
for Disease Control and Prevention.
Total homocysteine measurements in NHANES1
1999–2000 Abbott FPIA IMx
2001 Abbott FPIA IMx
2002Abbot FPIA AxSYM
2003–2004 Abbott FPIA AxSYM Plasma
2005–2006 Abbott FPIA AxSYM Plasma
1HPLC-FD, HPLC with fluorescence detection; FPIA, fluorescence po-
larization immunoassay (AxSYM and IMx; Abbott Diagnostics, Abbott Park,
IL); NCEH, National Center for Environmental Health, Centers for Disease
Control and Prevention; Tufts, Tufts University, Jean Mayer US Department of
Agriculture Human Nutrition Research Center on Aging, Boston, MA.
YETLEY ET AL
children and young and middle-aged adults, could improve vi-
tamin B-12 status by decreasing Helicobacter pylori infections
that have statistical associations with decreased vitamin B-12
absorption. Increased adoption of vegan diets, the substitution of
chicken for beef, and decreased use of dairy products could
lower vitamin B-12 intakes. Several experts felt that as the use of
vitamin B-12–containing supplements (including some with doses
of 1000–2000 lg/d) and fortified foods increases, monitoring long-
term, high vitamin B-12 intakes would be useful to evaluate the
safety of chronic exposures to excessively high intakes in healthy
populations. Monitoring vitamin B-12 intakes could help assess
whether lifestyle changes are associated with changes in vita-
min B-12 status.
Which biomarkers should future NHANES include?
The roundtable addressed which biomarkers to include in
NHANES if the program reinstates vitamin B-12 status moni-
toring. Given each biomarker’s sensitivity and specificity prob-
lems, the results of a single biomarker measurement would be
difficult to interpret. But if NHANES can measure only one
biomarker, some experts felt that this measure should reflect the
broad range of vitamin B-12 status in the US population. Vitamin
B-12 and holoTC increase continuously with increasing intakes,
at least until vitamin B-12 saturates the transport proteins, and
therefore reflect a broad range of intakes and status. Other ro-
undtable experts felt that the sensitivity of MMA and tHcy to
early inadequacies in vitamin B-12 status and marginal vitamin
B-12 intakes could support the use of either one as the sole
NHANES biomarker. Mean corpuscular volume to detect meg-
aloblastic anemia would not be useful because ,10% of patients
with vitamin B-12 deficiency have megaloblastic anemia, and
high mean corpuscular volume is more likely to reflect alcohol
abuse than vitamin B-12 deficiency (9).
However, the roundtable strongly urged that, given the sen-
sitivity and specificity problems of all vitamin B-12–related
biomarkers, future NHANES should concurrently measure at
least onebiomarker ofcirculating concentrations ofvitaminB-12
(vitamin B-12 or holoTC) and one biomarker of functional vi-
tamin B-12 status (MMA or tHcy). If a choice must be made
between serum vitamin B-12 and holoTC, holoTC has the ad-
vantage of not being affected by “false-positive” low vitamin B-12
concentrations such as those caused by genetically determined
haptocorrin deficiencies. These conditions result in low circu-
lating vitamin B-12 concentrations but have no adverse effect on
vitamin B-12 status because they do not affect holoTC, which
is the universal transport protein for vitamin B-12 uptake in all
cells. However, the roundtable also recognized that holoTC mea-
surement is new and would benefit from additional performance
studies to engender full confidence in the use of holoTC for
population-based surveys. Therefore, at this time, the roundtable
agreed that serum vitamin B-12 is preferable over holoTC as the
measure of circulating concentrations of vitamin B-12. MMA is
preferable to tHcy if a choice must be made between these bio-
with folate inadequacy, whereas both of these nutrient deficiencies
The roundtable generally agreed that, if NHANES reinstates
2 vitamin B-12–related biomarkers, serum vitamin B-12 and
plasma MMA would provide continuity with past surveys and
provide circulating and functional indicators of vitamin B-12
status. These variables have shown associations with anemia and
59). Reliable measurement procedures are available, and the NIST
is developing reference materials. The roundtable also suggested
that NHANES periodically measurevitamin B-12, holoTC, MMA,
and tHcy concurrently to provide valuable information on their
vary by biomarker or biomarker combinations.
In July 2010, a roundtable reviewed the measurement of vi-
tamin B-12 status biomarkers in NHANES. The roundtable was
uncertain whether subclinical vitamin B-12 deficiency is a public
health problem warranting population-based monitoring. How-
ever, associations between subclinical vitamin B-12 deficiency
and urgently need further evaluation in clinical trials. Clinical trials
must use accurate and reliable measurement procedures to de-
at-risk populations. The roundtable was generally satisfied with the
have used or that are available for use in future NHANES, as well
as the reference methods and materials available from, or under
development by, the NIST. The experts supported reinstating
vitamin B-12–related biomarkers in future NHANES and sug-
gested several public health and environmental reasons for doing
so. Finally, the roundtable agreed that future NHANES should
include at least 2 biomarkers—1 of circulating vitamin B-12
of vitamin B-12 insufficiency (preferably MMA).
We thank Anne Thurn and Claudia Faigen of the Office of Dietary Supple-
ments, National Institutes of Health; Kimberly Potter, Megan McNamee, and
Therese Gibson of ICF International; and Mike Bykowski of Consolidated
Solutions and Innovations for their outstanding logistical, organizational,
and follow-up support. We thank Deborah Berlyne for her expert and timely
technical editing services.
The authors’ responsibilities were as follows—MFP, CLJ, EAY, and PMC:
conceivedand sponsored the roundtableproject;MFP, CLJ,EAY, PMC,RLB,
DAL, AMM, JLM, CMP, KWP, CS, BS, and TT: served on the planning com-
in the roundtable discussions and read and approved the final manuscript. RG
and DWJ serve on the Scientific Advisory Boards of Emisphere Technologies
Diagnostics and also carried a consulting agreement with this firm. RC’s hos-
pital filed a patent application for his workon thegeneticsof transcobalamin I
deficiency. None of the other authorshad any personalor financial conflictsof
1. Yetley EA, Johnson CL. Folate and vitamin B-12 biomarkers in
NHANES: history of their measurement and use. Am J Clin Nutr 2011;
2. Bock JL, Eckfeldt JH. Advances in standardization of laboratory
measurement procedures: implications for measuring biomarkers of
VITAMIN B-12–RELATED BIOMARKERS IN NHANES
folate and vitamin B-12 status in NHANES. Am J Clin Nutr 2011;94
3. Yetley EA, Pfeiffer CM, Phinney KW, et al. Biomarkers of folate status
in NHANES: a roundtable summary. Am J Clin Nutr 2011;94(suppl):
4. Yetley EA, Coates PM, Johnson CL. Overview of a roundtable on
NHANES monitoring of biomarkers of folate and vitamin B-12 status:
measurement procedure issues. Am J Clin Nutr 2011;94(suppl):297S–
5. Green R, Miller JW. Vitamin B12deficiency is the dominant nutritional
cause of hyperhomocysteinemia in a folic acid-fortified population.
Clin Chem Lab Med 2005;43:1048–51.
6. Carmel R, Green R, Jacobsen DW, Rasmussen K, Florea M, Azen C.
Serum cobalamin, homocysteine, and methylmalonic acid concen-
trations in a multiethnic elderly population: ethnic and sex differences in
cobalamin and metabolite abnormalities. Am J Clin Nutr 1999;70:904–10.
7. Ganji V, Kafai MR. Demographic, lifestyle, and health characteristics
and serum B vitamin status are determinants of plasma total homo-
cysteine concentration in the post-folic acid fortification period, 1999-
2004. J Nutr 2009;139:345–52.
8. Nexo E., Hoffmann-Lu ¨cke E. Holotranscobalamin, a marker of vitamin
B-12 status: analytical aspects and clinical utility. Am J Clin Nutr 2011;
9. Carmel R. Biomarkers of cobalamin (vitamin B-12) status in the epi-
demiologic setting: a critical overview of context, applications, and
performance characteristics of cobalamin, methylmalonic acid, and
holotranscobalamin II. Am J Clin Nutr 2011;94(suppl):348S–58S.
10. Carmel R, Sarrai M. Diagnosis and management of clinical and sub-
clinical cobalamin deficiency: advances and controversies. Curr Hematol
11. Hvas A-M, Ellegaard J, Nexo E. Increased plasma methylmalonic acid
level does not predict clinical manifestations of vitamin B12deficiency.
Arch Intern Med 2001;161:1534–41.
12. Morris MS, Jacques PF, Rosenberg IH, Selhub J. Folate and vitamin
B-12 status in relation to anemia, macrocytosis, and cognitive im-
pairment in older Americans in the age of folic acid fortification. Am J
Clin Nutr 2007;85:193–200.
13. Garrod MG, Green R, Allen LH, et al. Fraction of total plasma vitamin
B12 bound to transcobalamin correlates with cognitive function in el-
derly Latinos with depressive symptoms. Clin Chem 2008;54:1210–7.
14. Riggs KM, Spiro A III, Tucker K, Rush D. Relations of vitamin B-12,
vitamin B-6, folate and homocysteine to cognitive performance in the
Normative Aging Study. Am J Clin Nutr 1996;63:306–14.
15. Selhub J, Morris MS, Jacques PF, Rosenberg IH. Folate-vitamin B-12
interaction in relation to cognitive impairment, anemia, and biochemical
indicators of vitamin B-12 deficiency. Am J Clin Nutr 2009;89:
16. Miller JW, Garrod MG, Allen LH, Haan MN, Green R. Metabolic
evidence of vitamin B-12 deficiency, including high homocysteine and
methylmalonic acid and low holotranscobalamin, is more pronounced
in older adults with elevated plasma folate. Am J Clin Nutr 2009;90:
17. Clarke R, Sherliker P, Hin H, et al. Folate and vitamin B12status in
relation to cognitive impairment and anaemia in the setting of volun-
tary fortification in the UK. Br J Nutr 2008;100:1054–9.
18. Green R. Indicators for assessing folate and vitamin B12status and for
monitoring the efficacy of intervention strategies. Food Nutr Bull 2008;
19. Vogiatzoglou A, Oulhaj A, Smith AD, et al. Determinants of plasma
methylmalonic acid in a large population: implications for assessment
of vitamin B12status. Clin Chem 2009;55:2198–206.
20. Hvas A-M, Nexo E. Holotranscobalamin—a first choice assay for di-
agnosing early vitamin B deficiency? J Intern Med 2005;257:289–98.
21. Clarke R, Sherliker P, Hin H, et al. Detection of vitamin B12deficiency
in older people by measuring vitamin B12or the active fraction of
vitamin B12, holotranscobalamin. Clin Chem 2007;53:963–70.
22. Selhub J, Jacques PF, Rosenberg IH, et al. Serum total homocysteine
concentrations in the third National Health and Nutrition Examination
Survey (1991-1994): population reference ranges and contribution of vita-
min status to high serum concentrations. Ann Intern Med 1999;131:331–9.
23. Morris MS, Jacques PF, Rosenberg IH, Selhub J. Elevated serum
methylmalonic acid concentrations are common among elderly Ameri-
cans. J Nutr 2002;132:2799–803.
24. Gunter EW, Lewis BG, Koncikowski SM. Laboratory procedures used
for the third National Health and Nutrition Examination Survey
(NHANES III), 1988-1994. Atlanta, GA: Centers for Disease Control
and Prevention; Hyattsville, MD: National Center for Health Statistics,
1996. Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes3/
cdrom/nchs/manuals/labman.pdf (cited 4 December 2010).
25. Centers for Disease Control and Prevention. NHANES 1999-2000:
folate/vitamin B12, serum and whole blood, Bio-Rad Laboratories’
“Quantaphase II folate/vitamin B12” radioassay kit. Available from:
b12.pdf (cited 4 December 2010).
26. Centers for Disease Control and Prevention. NHANES 2001-2002:
folate/vitamin B12, serum and whole blood, Bio-Rad Laboratories
“Quantaphase II Folate/Vitamin B12” radioassay kit. Available from:
b12.pdf (cited 4 December 2010).
27. Centers for Disease Control and Prevention. NHANES 2003-2004:
folate/vitamin B12, serum and whole blood, Bio-Rad Laboratories’
“Quantaphase II Folate/Vitamin B12” Radioassay Kit. Available from:
20B12.pdf (cited 4 December 2010).
28. Centers for Disease Control and Prevention. NHANES 2005-2006:
folate/vitamin B12, serum and whole blood, Laboratories “Quanta-
phase II Folate/Vitamin B12” radioassay kit. Available from: http:
pdf (cited 4 December 2010).
29. Pfeiffer CM, Johnson CL, Jain RB, et al. Trends in blood folate and
vitamin B-12 concentrations in the United States, 1988–2004. Am J
Clin Nutr 2007;86:718–27.
30. National Institute of Standards and Technology. Development of
standard reference materials for vitamins B6 and B12 in human serum.
Available from: http://www.nist.gov/mml/analytical/organic/vitsb6andb12
inserum.cfm (cited 4 December 2010).
31. Centers for Disease Control and Prevention. NHANES 1999-2000:
methylmalonic acid (MMA), serum, GC/MS. Available from: http://
acid.pdf (cited 4 December 2010).
32. Centers for Disease Control and Prevention. NHANES 2001-2002:
methylmalonic acid (MMA), plasma or serum, GC/MS. Available
methylmalonic_acid.pdf (cited 4 December 2010).
33. Centers for Disease Control and Prevention. NHANES 2003-2004:
methylmalonic acid (MMA), plasma or serum, GC/MS. Available
mma.pdf (cited 4 December 2010).
34. Pfeiffer CM, Gunter EW. Automated assay for methylmalonic acid
(MMA) in plasma. 51st Annual Meeting of American Association of
Clinical Chemists, New Orleans, LA, July 1999. Clin Chem 1999;45
(suppl 6):A166 (abstr. 593).
35. Blom HJ, Rooij AV, Hogeveen M. A simple high-throughput method for
the determination of plasma methylmalonic acid by liquid chromatog-
raphy-tandem mass spectrometry. Clin Chem Lab Med 2007;45:645–50.
36. Lakso H-A, Appelblad P, Schneede J. Quantification of methylmalonic
acid in human plasma with hydrophilic interaction liquid chromatog-
raphy separation and mass spectrometric detection. Clin Chem 2008;
37. Fasching C, Singh J. Quantitation of methylmalonic acid in plasma
using liquid chromatography-tandem mass spectrometry. Methods Mol
38. Centers for Disease Control and Prevention. NHANES 1999-2000:
total homocysteine (tHcy), plasma, Abbott homocysteine (HCY) assay.
Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes_99_00/
lab06_met_homocysteine.pdf (cited 4 December 2010).
39. Centers for Disease Control and Prevention. NHANES 2001-2002:
total homocysteine (tHcy), plasma, “Abbott AxSYM homocysteine
(HCY) assay.” Available from: http://www.cdc.gov/nchs/data/nhanes/
nhanes_01_02/l06_b_met_homocysteine_Axsym.pdf (cited 4 Decem-
40. Centers for Disease Control and Prevention. NHANES 2001-2002:
total homocysteine (tHcy), plasma, Abbott IMX homocysteine (HCY)
assay. Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes_
01_02/l06_b_met_homocysteine_IMX.pdf (cited 4 December 2010).
41. Centers for Disease Control and Prevention. NHANES 2003-2004:
total homocysteine (tHcy), plasma, Abbot homocysteine (HCY) assay.
YETLEY ET AL
Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes_03_04/ Download full-text
l06_c_met_homocysteine.pdf (cited 4 December 2010).
42. Centers for Disease Control and Prevention. NHANES 2005-2006:
total homocysteine (tHcy), plasma, “Abbott Homocysteine (HCY)
assay.” Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes_
05_06/hcy_d_met.pdf (cited 4 December 2010).
43. Pfeiffer CM, Twite D, Shih J, Holets-McCormack SR, Gunter EW.
Method comparison for total plasma homocysteine between the Abbott
IMx analyzer and an HPLC assay with internal standardization. Clin
44. Pfeiffer CM, Caudill SP, Gunter EW, et al. Analysis of factors influ-
encing the comparison of homocysteine values between the third Na-
tional Health and Nutrition Examination Survey (NHANES) and
NHANES 1999+. J Nutr 2000;130:2850–4.
45. Pfeiffer CM, Osterloh JD, Kennedy-Stephenson J, et al. Trends in cir-
culating concentrations of total homocysteine among US adolescents and
adults: findings from the 1991-1994 and 1999-2004 National Health and
Nutrition Examination Surveys. Clin Chem 2008;54:801–13.
46. Brady J, Wilson L, McGregor L, Valente E, Orning L. Active B12:
a rapid, automated assay for holotranscobalamin on the Abbott Ax-
SYM analyzer. Clin Chem 2008;54:567–73.
47. Morkbak AL, Heimdal RM, Emmens K, et al. Evaluation of the technical
performance of novel holotranscobalamin (holoTC) assays in a multicenter
European demonstration project. Clin Chem Lab Med 2005;43:1058–64.
48. National Institute of Standards and Technology. Certificate of analysis,
standard reference material 1955, homocysteine and folate in frozen
human serum. April 2008. Available from: https://www-s.nist.gov/srmors/
d5a634a2a (cited 4 December 2010).
49. National Institute of Standards and Technology. Certificate of analysis,
standard reference material 909b, human serum. November 2003. Avail-
able from: http://ts.nist.gov/measurementservices/referencematerials/
archived_certificates/909b.Nov19.2003.pdf (cited 4 December 2010).
50. National Institute of Standards and Technology. Development of
a standard reference material for metabolites in plasma. Available
cfm (cited 4 December 2010).
51. Satterfield MB, Sniegoski LT, Welch MJ, Nelson BC. Comparison of
isotope dilution mass spectrometry methods for the determination of
total homocysteine in plasma and serum. Anal Chem 2003;75:4631–8.
52. Satterfield MB, Sniegoski LT, Sharpless KE, et al. Development of
a new standard reference material: SRM 1955 (homocysteine and fo-
late in human serum). Anal Bioanal Chem 2006;385:612–22.
53. Nelson BC, Pfeiffer CM, Sniegoski LT, Satterfield MB. Development
and evaluation of an isotope dilution LC/MS method for the deter-
mination of total homocysteine in human plasma. Anal Chem 2003;75:
54. Nelson BC, Pfeiffer CM, Zhang M, Duewer DL, Sharpless KE, Lippa
KA. Commutability of NIST SRM 1955 homocysteine and folate in
frozen human serum with selected total homocysteine immunoassays
and enzymatic assays. Clin Chim Acta 2008;395:99–105.
55. JCTLM/Joint Committee on Traceability in Laboratory Medicine.
Isotope dilution mass spectrometry methods for homocysteine in blood
serum. Se `vres Cedex, France: Bureau International des Poids et Me-
sures. Available from: http://www.bipm.org/jctlm/search.do?sortBy=
(cited 4 December 2010).
56. Pfeiffer CM, Caudill SP, Gunter EW, Osterloh J, Sampson EJ. Bio-
chemical indicators of B vitamin status in the US population after folic
acid fortification: results from the National Health and Nutrition Ex-
amination Survey 1999–2000. Am J Clin Nutr 2005;82:442–50.
57. Valuck RJ, Ruscin JM. A case-control study on adverse effects: H2
blocker or proton pump inhibitor use and risk of vitamin B12deficiency
in older adults. J Clin Epidemiol 2004;57:422–8.
58. Howden CW. Vitamin B12 levels during prolonged treatment with
proton pump inhibitors. J Clin Gastroenterol 2000;30:29–33.
59. Clarke R, Birks J, Nexo E, et al. Low vitamin B-12 status and risk of
cognitive decline in older adults. Am J Clin Nutr 2007;86:1384–91.
VITAMIN B-12–RELATED BIOMARKERS IN NHANES