An Economic Analysis of Anemia Prevention during Infancy

MARCUS SHAKER, MD, MS, PAMELA JENKINS, MD, PHD, CHRISTINA ULLRICH, MD, MPH,
CARLO BRUGNARA, MD, BAO TRAM NGHIEM, BA,
AND HENRY BERNSTEIN, DO

Objective To compare the cost-benefit profile of reticulocyte hemoglobin content (CHr) with hemoglobin (Hb) alone and
Hb as a component of the complete blood count (CBC) for detection and treatment of iron deficiency in 9- to 12-month-old
infants.
Study design Cohort simulations were used to compare CHr with Hb from a societal perspective. Assumptions included a
9% prevalence of iron deficiency and testing characteristics/costs of CHr, Hb, and CBC (CHr <27.5 pg: sensitivity 83%,
specificity 72%, $11; Hb <11 g/dL: sensitivity 26%, specificity 95%, $5; CBC Hb<11g/dL, $15), as well as cost of iron therapy
($61 for established anemia). Sensitivity analyses were performed.
Results Under current market conditions, the incremental cost to diagnose and treat iron deficiency, compared with
diagnosing and treating anemia by Hb, was only $22 per patient screened ($440 per case of anemia prevented; number needed
to treat ⴝ 20). With a 10-year time horizon incorporating risks and costs of neurocognitive delays associated with untreated
iron deficiency, the cost of the CHr strategy was $280 per case of anemia prevented.
Conclusions CHr is an affordable strategy to prevent anemia in infants with possible iron deficiency. (J Pediatr 2009;154:44-9)

ron deficiency is the most common nutritional deficiency in the United States.1 Infants are at risk for iron deficiency from

I insufficient dietary iron, variable absorption, and rapid growth. Iron deficiency, with or without anemia, may impair mental
and motor development during infancy2-9 and result in impaired neurocognitive development, with lower behavioral,
cognitive, and motor scores during critical time periods.10
Hemoglobin (Hb) is the most commonly used screening test for anemia, but iron deficiency in infants may be present for
several months before it results in a decrease in the blood hemoglobin. Reductions in reticulocyte hemoglobin content (CHr)
reflect iron deficiency before anemia develops.10-12 A CHr of ⬍ 27.5 pg is a more accurate indicator of iron deficiency than
anemia, defined by a Hb ⬍ 11 g/dL, in healthy 9- to 12-month-old infants.13 Screening for iron deficiency with CHr can help
prevent iron deficiency anemia during infancy; however, the economic consequences of anemia prevention with CHr are
uncharacterized.
The objective of this analysis was to evaluate the cost-effectiveness of anemia
prevention during infancy with CHr screening. We performed economic modeling of
diagnostic strategies using CHr and Hb to diagnose iron deficiency. We sought to
determine the cost-effectiveness of CHr versus Hb testing and threshold costs for anemia From the Children’s Hospital at Dart-
prevention. mouth, Lebanon (M.S., P.J., H.B.), The Cen-
ter for the Evaluative Clinical Sciences (M.S.,
P.J.), Dartmouth College (B.N.), Hanover,
METHODS NH, the Dana Farber Cancer Institute
(C.U.), Children’s Hospital (C.B., H.B.), Bos-
Description of the Model ton, MA.
The authors have no conflicts of interest to
We used a computer-based mathematical model (TreeAge Pro 2005 Suite, Wil- declare.
liamstown, MA) to perform cohort simulations of competing diagnostic strategies used to Submitted for publication Dec 11, 2007;
evaluate iron deficiency (Figure 1). The model follows hypothetical cohorts of patients at last revision received Apr 28, 2008; ac-
cepted Jun 24, 2008.
risk for iron deficiency. In cohort simulation, transitions are experienced by the proportion
Reprint requests: Marcus Shaker, MD,
of persons in each state corresponding to transition probabilities. Monte Carlo micro- Dartmouth-Hitchcock Medical Center, De-
simulation allows individual patients from hypothetical cohorts to cycle through transi- partment of Pediatrics, One Medical Cen-
ter Dr, Lebanon, NH 03756. E-mail:
marcus.shaker@dartmouth.edu.
ARR Absolute risk reduction NNS Number needed to screen 0022-3476/$ - see front matter
CBC Complete blood count NPV Negative predictive value Copyright © 2009 Mosby Inc. All rights
CHr Reticulocyte hemoglobin content WTP Willingness to pay reserved.
Hb Hemoglobin
10.1016/j.jpeds.2008.06.038

44
 Table I. Baseline assumptions and probabilities
Name Value
Costs*
CHr screening test15 $11
Hb screening test15 $5
Complete blood count15 $15
Treatment cost of anemia (3 month course)* $61
Treatment cost iron deficiency (3 month course)* $61
Treatment cost of iron deficiency for (1 month $20
course)*
Probabilities
CHr screening test ⬍ 27.5 pg13
Sensitivity 83%
Specificity 72%
Hb screening test ⬍ 11 g/dL13
Sensitivity 26%
Specificity 95%
Additional Assumptions in 10 year time horizon
Figure 1. Decision model: decision tree to evaluate competing diagnostic Second test probability (independent testing 60%
strategies for iron deficiency. characteristics, Hb test performed)17
Annual attributable risk of neurocognitive disability
requiring services
tional probabilities14 and was used to estimate variance from Untreated anemia7 10%
stochastic uncertainty (n ⫽ 1000 per cohort). Strategies were Treated anemia 0%
Annual (transient) charges of services† for $500
modeled over 3-month and 10-year time horizons. The ref-
neurocognitive delays, if required
erence population for this model was 9- to 12-month-old
*Local cost estimates.
infants. The prevalence of iron deficiency for the reference
†Costs of cognitive delay were derived from local charges18 published by the
population was estimated at 9%.1 Dartmouth-Hitchcock Medical Center (Lebanon, NH) of physician visits (CPT 99213
charge $118 per visit) and potential developmental assessment (CPT 99243 charge $335
per visit or 99244 charge $412 per visit).
Diagnostic Strategies
Definitions and Assumptions
We examined 3 potential approaches in the evaluation
Effectiveness was defined by the number of patients in
of iron deficiency. Diagnostic accuracy of each strategy and
whom early CHr detection and treatment of iron deficiency
response rates were extracted from published literature. An
prevented subsequent anemia. This was calculated with the
approach with CHr and subsequent treatment was modeled
absolute risk reduction (ARR) in missed diagnosis between Hb
against a screening approach for anemia with Hb alone or as
and CHr (negative predictive value [NPV]CHr ⫺ NPVHb). The
part of a complete blood count (CBC). Three-month courses number needed to screen (NNS) to prevent 1 case of anemia
of iron therapy were prescribed both for subjects with a CHr was calculated with the inverse of the ARR (1/ARR). Cost per
⬍27.5, as well as for established anemia. Diagnostic accuracy case of anemia prevented was then calculated by the product of
was modeled from published sensitivity and specificity of the the difference in per-patient screening cost (CHr ⫺ Hb) and the
CHr ⬍27.5 (sensitivity 83%, 95% CI 61%-95%; specificity NNS. The market costs of anemia diagnosis and treatment
72%, 95% CI 65% to 78%).13 Ullrich et al demonstrated that were calculated per correct diagnosis with the product of the
a CHr of less than 27.5 pg without anemia at initial screening per-patient screening cost (Hb) and the inverse of the positive
was associated with subsequent anemia on re-screening in the predictive value (PPV)Hb (1/PPVHb). Costs of repeated physi-
second year of life (risk ratio 9.1, 95% CI 1.04-78.9, P ⫽ cian visits and downstream costs, including developmental
.01).13 Published literature on accuracy of Hb less than 11 effects from delayed diagnosis of iron deficiency, were not
g/dL in the diagnosis of anemia in this age cohort was used included in the short-term model, but costs of subsequent
(sensitivity 26%; 95% CI 10%-48%; specificity 95%, 95% CI neurocognitive delay were considered in the long-term model.
91%-98%).13 Market costs (current dollars) were estimated Progression from iron deficiency to anemia was assumed.
from a societal perspective with the laboratory fee schedule Costs were discounted at 3% per annum to reflect the average
from the Centers for Medicare and Medicaid Services15 (CHr annual consumer price index for all goods and services in the
⬍27.5 pg: $11; Hb ⬍11 g/dL: $5; CBC: $15). Local esti- long-term model.16
mates were used for the costs of iron therapy ($20.30 per
month). Baseline variables are shown in Table I. The cost of Sensitivity Analyses
each strategy was defined as the total costs for diagnosis and In each cohort baseline parameters were varied within a
treatment. clinically reasonable range to evaluate the preferred diagnostic

An Economic Analysis of Anemia Prevention during Infancy 45

strategy over a range of disease prevalence and treatment
durations.
Long-Term Model (10-Year Time-Horizon)
A Markov model was used to conduct additional anal-
yses with costs that might be associated with downstream
neurocognitive delays associated with untreated iron defi-
ciency.7 Markov modeling is a decision-analytic method that
incorporates a process derived from matrix algebra to describe
transitions experienced by a hypothetical cohort of patients
between defined health states over a linear time frame.14 It is
useful in clinical circumstances characterized by recurring
probabilistic risk. For these analyses the base case prevalence
and assumptions (initial 3-month treatment course for all iron
deficiency) of the short-term model were used together with
a second test for iron deficiency performed within 1 year of
the first screening test.
Neurocognitive delays associated with iron deficiency
anemia may persist for more than 10 years,7 so a 10-year
time-horizon model was designed to understand the impact
of the use of CHr as an initial screening test and establish a
comparison of costs between the use of CHr or Hb as the
initial test. It has been suggested that CHr may be a more
accurate screening test when used in 9- to 12-month-old
infants,13 but the incremental advantage of the use of this test
sequentially as compared with sequential Hb testing has
not been evaluated. As such, the first screening test was
either the CHr or the Hb test, and the second test per-
formed was the Hb.
Because 37% to 42% of at-risk patients may not receive
follow-up testing for anemia, a 60% rate of compliance with
follow-up testing was modeled.17 Lozoff et al7 found an
increased rate of long-term poor scholastic achievement and
use of special services in patients with treated chronic severe
iron deficiency during infancy: 26% of iron-deficient patients
repeated a grade compared with 12% non-iron-deficient chil-
dren (P ⫽ .04), and 21% of iron-deficient children were
referred for special services compared with 7% of non-iron-
deficient children (P ⫽ .02). Congruent with the attributable
rate of the neurocognitive deficit associated with iron defi-
ciency, an annual 10% rate of neurocognitive delays requiring
1 year of supplemental health care use (annual transient cost
$500) was assumed for patients with untreated iron defi-
ciency. Treatment of iron deficiency within the first 6 to 12
months was assumed to protect against this attributable in-
crease in disability costs. Annual disability costs were assumed
over a single year, with subsequent yearly costs modeled
independently at the annual risk of 10%. Costs of neurocog-
nitive delay were derived from local charges18 published by
the Dartmouth-Hitchcock Medical Center (Lebanon, New
Hampshire) of additional physician visits (CPT 99213
charge $118 per visit) and potential developmental assess-
ment (CPT 99243 charge $335 per visit or CPT 99244
charge $412 per visit); however, costs of delay may also
include additional services such as speech and language
therapy. Three-way sensitivity analyses were performed on
46 Shaker et al
these variables (rate of follow-up testing, rate of disability, and
cost of disability).
RESULTS
Short-Term Model, Base Case
CHr testing was the most accurate diagnostic approach
to exclude iron deficiency in this cohort (true-negative rate
98% vs 93%). The ARR between these strategies was 5%; for
every 20 patients screened by CHr, 1 case of iron deficiency
anemia not detected by Hb screening would be prevented.
CHr screening resulted in treatment of 33% of the cohort,
and Hb resulted in treatment of 7% of infants. The CHr
strategy cost $31 per patient (95% CI $29.16-$32.84) and
$440 per case of anemia prevented. Diagnosis and treatment
of anemia established by Hb alone cost $9 (95% CI $8.05-
$9.95) per patient screened ($19 with a CBC). The incre-
mental cost of CHr to prevent anemia was $22 per patient
screened when compared with Hb. The current cost of treat-
ing established anemia was $25 per patient screened ($55 with
a CBC).
Short-Term Model, Sensitivity Analyses
The cost of anemia prevention remained below $740
per case prevented when higher incremental costs of CHr
were compared with Hb (incremental cost difference $6 to
$15 per test). Cost-effectiveness did not exceed this level
when higher costs of iron therapy (range $20.30 to $26.60 per
month), lower CHr sensitivities and specificities (ranges 73%
to 93%; 62% to 82%), and higher rates of spontaneous reso-
lution of iron deficiency19 (range 0% to 40%) were considered.
When sensitivity of CHr was lowered to 61%, the cost of
anemia prevention was $1008 per case prevented.
Table II includes cost-benefit results on the basis of
screening strategies at higher prevalence rates. As prevalence
increased, the advantage of CHr testing over the Hb strategy
for excluding iron deficiency became more pronounced. At a
disease prevalence of 11%, the true-negative rate of CHr was
97% versus 91% for Hb. The ARR between these strategies
was 6%, and the NNS was 16.7. CHr screening resulted in
treatment of 34% of the cohort, and Hb screening resulted in
treatment of 7% of infants. CHr cost $32 (95% CI $30.16-
$33.84) per patient screened and $383 per case of anemia
prevented. At a disease prevalence of 15%, the true negative
rate of CHr was 96% versus 88% for Hb. The ARR between
these strategies was 8% and the NNS was 12.5. CHr screen-
ing resulted in treatment of 36% of the cohort, and Hb
screening resulted in treatment of 8% of infants. CHr screen-
ing cost $288 per case of anemia prevented.
Because shorter durations of treatment may be appro-
priate for iron deficiency detected before anemia, we per-
formed sensitivity analyses with a 1-month treatment course
for iron deficiency. In this setting the CHr strategy cost $18
(95% CI $17.37-$18.63) per patient screened ($180 per case
of anemia prevented). When the cost per test of CHr ap-
The Journal of Pediatrics • January 2009
Table II. Cost-benefit results*
Short-term model Long-term model
95% Confidence 95% Confidence Proportion
Cost† intervals Cost† intervals treated Benefit‡ Cost-benefit§
Prevalence 5%
CHr with 3 mo Rx $30 $28.22-$31.78 $45 $37.88-$52.12 Short-term model
31% $700 (3 mo Rx)
CHr with 1 mo Rx储 $17 $16.41-$17.59 $32 $28.20-$35.80
$267 (1 mo Rx)
3%
Hb with 3 mo Rx $9 $8.09-$9.91 $28 $20.75-$35.25 Long-term model
6% $566 (3 mo Rx)
CBC with 3 mo Rx $19 $18.25-$19.74 $38 $33.21-$42.79
$133 (1 mo Rx)
Prevalence 9%
CHr with 3 mo Rx $31 $29.16-$32.84 $55 $46.80-$63.20 Short-term model
33% $440 (3 mo Rx)
CHr with 1 mo Rx储 $18 $17.37-$18.63 $41.50 $32.78-$50.22
$180 (1 mo Rx)
5%
Hb with 3 mo Rx $9 $8.05-$9.95 $41 $31.85-$50.15 Long-term model
7% $280 (3 mo Rx)
CBC with 3 mo Rx $19 $18.05-$19.95 $51 $41.66-$60.34
$10 (1 mo Rx)
Prevalence 11%
CHr with 3 mo Rx $32 $30.16-$33.84 $59 $50.78-$67.22 Short-term model
34% $383 (3 mo Rx)
CHr with 1 mo Rx储 $18 $17.43-$18.57 $46 $38.80-$53.20 $150 (1 mo Rx)
6% Long-term model
Hb with 3 mo Rx $9 $7.99-$10.01 $47 $37.22-$56.56
$200 (3 mo Rx)
7% ⫺$16 (1 mo Rx)
CBC with 3 mo Rx $19 $18.11-$19.89 $57 $46.48-$67.52
(cost saving)
Prevalence 15%
CHr with 3 mo Rx $33 $31.16-$34.84 $69 $60.07-$77.93 Short-term model
36% $288 (3 mo Rx)
CHr with 1 mo Rx储 $18 $16.73-$19.27 $54 $44.52-$63.48 $113 (1 mo Rx)
8% Long-term model
Hb with 3 mo Rx $10 $9.05-$10.95 $59 $48.39-$69.61
$125 (3 mo Rx)
8% ⫺$63 (1 mo Rx)
CBC with 3 mo Rx $20 $18.92-$21.08 $69 $57.65-$80.35
(cost saving)
*Disease prevalence 5%, 9%, 11%, and 15% with analysis for reticulocyte hemoglobin content (CHr) and anemia (Hb).
†Per-patient (mean).
‡Incremental rate of anemia prevention. This was calculated using the ARR in missed diagnosis between Hb and CHr (NPVCHr ⫺ NPVHb). Market value of anemia diagnosis by
Hb (or CBC) is $25 ($55), $23 ($49), and $19 ($42) per correct diagnosis (at prevalence rates of 9%, 11%, and 15%).
§Cost per case of anemia prevented (CHr versus Hb). This was calculated with the product of the difference in cost per patient (CHr ⫺ Hb) and the number needed to screen (1/ARR).
储Treatment with 1- and 3-month courses of iron therapy for CHr and Hb, respectively.

proached $3.50, the cost of anemia prevention fell below the
current market value of treatment (Figure 2).
Long-Term Model, Base Case
Incorporating downstream costs of disability over a 10
year time horizon was associated with lower incremental cost
of CHr screening ($55 per patient screened; 95% CI $46.80-
$63.20) when compared with Hb ($41 per patient screened
(95% CI $31.85-$50.15)). With a 10-year time horizon in-
corporating risks of neurocognitive delays, the cost of the
CHr strategy was $280 per case of anemia prevented.
Long-Term Model, Sensitivity Analyses
Sensitivity analysis demonstrated that CHr was a less
expensive initial screening method than Hb when the annual
cost of therapy for neurocognitive delay exceeded $1180.
Figure 3 depicts the 3-way sensitivity analysis of second test
probability, risk of disability, and cost of disability. In further
sensitivity analyses with the 10-year time horizon, the cost of
CHr was $10 per case of anemia prevented, when a short
treatment course (1 month) was prescribed for iron deficiency
detected by CHr and compared with a 3-month treatment
course for Hb.
DISCUSSION
Our analysis suggests that in healthy 9- to 12-month-
old infants, CHr is an affordable screening strategy to prevent
anemia, costing approximately $440 per case when a short-
term time horizon is used and $280 when a 10-year time
horizon in used. Because earlier diagnosis of iron deficiency
may allow shorter treatment courses than for established
anemia, the cost may actually be closer to $180 per case of

An Economic Analysis of Anemia Prevention during Infancy 47


Figure 2. Cost of anemia prevention: Incremental costs per case
prevented (CHr vs. Hb). The base analysis modeling 3-month iron
deficiency treatment is shown in red. The sensitivity analysis modeling
1-month of iron deficiency treatment is shown in blue. The black line
shows the current market costs per diagnosis of iron deficiency anemia
(9% prevalence rate of iron deficiency).
anemia prevented. When a 1-month treatment duration was
modeled and the cost per test of CHr approached $3.50, the
cost of anemia prevention fell below the current market value
of treatment.
Prevention of anemia is important because anemia can
lead to significant disability if untreated. Iron is necessary for
early neurocognitive developmental processes during infancy
and deficiency at this critical period can impair milestones and
individual potential.2-9 Although iron deficiency may resolve
spontaneously,17,19 the effect of undiagnosed and untreated
iron deficiency during critical periods of infant growth and
development place children at risk for permanent adverse
cognitive effects.3,6 Our base case analysis assumed anemia
progression for all CHr true-positive screening test results;
however, even when rates of spontaneous resolution of iron
deficiency were considered (up to 40%), costs did not exceed
$740 per case of anemia prevented.
We did not include either the follow-up costs of anemia
management or the downstream costs of iron deficiency as-
sociated with cognitive impairment in the short-term model.
Because these costs would be expected to decrease the incre-
mental cost of anemia prevention, we used a Markov model to
estimate the downstream costs associated with iron defi-
ciency. Although long-term neurocognitive effects of iron
deficiency have been demonstrated, it is difficult to estimate
the actual risk and cost of these sequelae. On the basis of the
work of Lozoff et al7 discussed above, we assumed a 10%
annual risk of neurocognitive delay significant enough to
require some degree of additional health care use in infants
with untreated iron deficiency. The annual disability costs
modeled were quite low—$500 per year for children with
neurocognitive delays. We did not assume any disability for
patients who received follow-up testing within 1 year and
assumed a follow-up testing rate of 60%. This rate of fol-
48 Shaker et al
Figure 3. Sensitivity analysis on disability probability, cost, and second
test rate. Markov modeling was used to evaluate costs of CHr and Hb
screening over a 10-year time horizon. Areas in blue indicate situations in
which CHr is less expensive than Hb from the standpoint of cost per
patient screened.
low-up is consistent with that described by Bogen et al,17 who
found that 37% to 42% of children identified as anemic by
routine screening in an inner-city clinic did not return for
follow-up anemia testing within 6 months. To further eval-
uate the effect of changing these assumptions, sensitivity
analyses were performed.
CHr resulted in treatment of 33% of the cohort, as-
suming a 9% prevalence of iron deficiency. The consequences
of unnecessary treatment may include adverse effects of iron
therapy such as nausea, vomiting, constipation, diarrhea, and
dark stools. These unintended adverse effects are generally
mild and self-limited but may lead to increased medical cost
and use, as well as parental anxiety. The trade-off between
prevention of iron deficiency anemia and additional costs and
potential side-effects of therapy may be a patient preference–
sensitive decision, and further research may be needed to
better understand how best to involve patient values and
preferences in this decision.
The willingness to pay (WTP) threshold for anemia
prevention during infancy is uncharacterized; however, the
costs for prevention described in this analysis appear afford-
able when compared with benchmarks in other conditions.
The WTP thresholds to prevent an episode of otitis media
during infancy and an uncomplicated influenza infection dur-
ing infancy is $10020 and $175,21 respectively. Parents were
willing to pay $500 to reduce the risk of meningitis from 21
in 100 000 to 6 in 100 000.20 In a cohort of adult cancer
patients at risk for non-iron deficiency symptomatic anemia
the WTP threshold to prevent anemia ranged from $300 to
$875.22 Thresholds for anemia prevention during infancy fall
within these ranges, but differences in these populations limit
generalizability.
Given the importance of anemia prevention during
infancy, it is tempting to speculate on the use of CHr screen-
ing before 9 months of age. Such an approach may lead to
diagnosis and treatment of iron deficiency during a more
appropriate developmental window. Considering the low
rates of follow-up testing and adherence to iron therapy,17
The Journal of Pediatrics • January 2009
more accurate diagnosis of iron deficiency at earlier ages,
combined with shorter courses of treatment, may address
unmet needs in this population of children.
The cost of anemia prevention with a CHr threshold of
less than 27.5 pg in healthy 9- to 12-month-old infants
appears affordable. Characterization of the quality of life
implications of untreated iron deficiency and of WTP thresh-
olds in this context is warranted. The clinical introduction of
this practice should facilitate more rapid, affordable, and
efficient care of infants at risk for iron deficiency anemia.
REFERENCES
1. Looker AC, Dallman PR, Carroll MD, Gunter EW, Johnson CL. Prevalence of
iron deficiency in the United States. JAMA 1997;277:973-6.
2. Lozoff B, Brittenham GM, Viteri FE, Wolf AW, Urrutia JJ. Developmental
deficits in iron-deficient infants: effects of age and severity of iron lack. J Pediatr
1982;101:948-52.
3. Oski FA, Honig AS, Helu B, Howanitz P. Effect of iron therapy on behavior
performance in nonanemic, iron-deficient infants. Pediatrics 1983;71:877-80.
4. Walter T, Kovalskys J, Stekel A. Effect of mild iron deficiency on infant mental
development scores. J Pediatr 1983;102:519-22.
5. Halterman JS, Kaczorowski JM, Aligne CA, Auinger P, Szilagyi PG. Iron
deficiency and cognitive achievement among school-aged children and adolescents in the
United States. Pediatrics 2001;107:1381-6.
6. Lozoff B, Jimenez E, Wolf AW. Long-term developmental outcome of infants
with iron deficiency. N Engl J Med 1991;325:687-94.
7. Lozoff B, Jimenez E, Hagen J, Mollen E, Wolf AW. Poorer behavioral and
developmental outcome more than 10 years after treatment for iron deficiency in infancy.
Pediatrics 2000;105:E51.
8. Walter T, De Andraca I, Chadud P, Perales CG. Iron deficiency anemia: adverse
effects on infant psychomotor development. Pediatrics 1989;84:7-17.
An Economic Analysis of Anemia Prevention during Infancy
9. Lozoff B, De Andraca I, Castillo M, Smith JB, Walter T, Pino P. Behavioral and
developmental effects of preventing iron-deficiency anemia in healthy full-term infants.
Pediatrics 2003;112:846-54.
10. Goodnough LT, Skikne B, Brugnara C. Erythropoietin, iron, and erythropoiesis.
Blood 2000;96:823-33.
11. Major A, Mathez-Loic F, Rohling R, Gautschi K, Brugnara C. The effect of
intravenous iron on the reticulocyte response to recombinant human erythropoietin. Br J
Haematol 1997;98:292-4.
12. Brugnara C, Laufer MR, Friedman AJ, Bridges K, Platt O. Reticulocyte hemo-
globin content (CHr): early indicator of iron deficiency and response to therapy. Blood
1994;83:3100-1.
13. Ullrich C, Wu A, Armsby C, Rieber S, Wingerter S, Brugnara C, et al. Screening
healthy infants for iron deficiency using reticulocyte hemoglobin content. JAMA
2005;294:924-30.
14. Hunink MG. Decision Making in Health and Medicine: integrating evidence and
values. United Kingdom: University Press, Cambridge; 2004.
15. Centers for Medicare and Medicaid Services. www.cms.hhs.gov. Accessed
1/28/2007.
16. Bureau of Labor Statistics. http://www.bls.gov/. Last accessed 3/10/06.
17. Bogen DL, Krause JP, Serwint JR. Outcome of children identified as anemic by
routine screening in an inner-city clinic. Arch Pediatr Adolesc Med 2001;155:366-71.
18. Doctor’s Office and Emergency Department Visits. http://www.dhmc.org/
webpage.cfm?site_id⫽2&org_id⫽564&morg_id⫽0&sec_id⫽0&gsec_id⫽29698&item_
id⫽29698. Accessed 3/25/2008.
19. Biondich PG, Downs SM, Caroll AE, Laskey AL, Liu GC, Rosenman M, et al.
Shortcoming in infant iron deficiency screening methods. Pediatrics 2006;117:290-4.
20. Prosser LA, Ray GT, O’Brien M, Kleinman K, Santoli J, Lieu TA. Preferences
and willingness to pay for health states prevented by pneumococcal conjugate vaccine.
Pediatrics 2004;113:283-90.
21. Prosser LA, Bridges CB, Uyeki TM, Rego VH, Ray GT, Meltzer MI, et al.
Values for preventing influenza-related morbidity and vaccine adverse events in children.
Health Qual Life Outcomes 2005;3:18.
22. Ortega A, Dranitsaris G, Puodziunas AL. What are cancer patients willing to pay
for prophylactic epoetin alfa? A cost-benefit analysis. Cancer 1998;83:2588-96.
49