Validity Assessment of Nine Discriminant Functions Used

for the Differentiation between Iron Deficiency Anemia

and Thalassemia Minor
by Suad M. AlFadhli,a Anwar M. Al-Awadhi,a and Doa’a AlKhaldib
a
Department of Medical Laboratory Sciences, Faculty of Allied Health Sciences, Kuwait University, Kuwait
b

Downloaded from https://academic.oup.com/tropej/article-abstract/53/2/93/1660096 by guest on 16 August 2019

Hematology Unit, Sabah Hospital, Ministry of Health, Kuwait

Summary
Iron deficiency anemia (IDA) and thalassemia minor are two of the most common causes of microcytic
anemias worldwide. Because of similar red blood cell count parameters and blood picture, it was
imperative to develop other measures that would differentially and correctly diagnose these two
anemias. Several mathematical formulas and simple RBC indices have been introduced as simple, fast
and inexpensive means of providing differential diagnosis for IDA and thalassemia minor. The Objective
of this study was to apply and compare nine well-documented discriminant functions on a population of
153 confirmed cases of microcytic anemias (IDA n ¼ 56, b-thalassemia minor n ¼ 47 and a-thalassemia
n ¼ 50) and to measure validity using Youden’s Index. The results show that England and Fraser
(E & F) Index had the highest Youden’s Index value (98.2) in correctly differentiating between IDA and
a- and b-thalassemia minor, while Shine and Lal Index was found ineffective in differentiating between
microcytic anemias in our population. E & F Index showed with great sensitivity and specificity to be the
best discriminant function to differentiate between IDA and thalassemia minor cases.

Key words: iron deficiency anemia, thalassemia minor, discriminant functions.

Introduction undetected blood loss indicated by IDA can be

Microcytic anemia can be a result of insufficient
iron supply or defective hemoglobin synthesis.
The red blood cells appear smaller than normal on
a peripheral blood smear and the mean cell volume
(MCV), a measure of red blood cell size, is generally
<80 femtoliters (fL) (normal 80–100 fL). Iron
deficiency anemia (IDA) is the most prevalent
microcytic anemia in the world, while thalassemia
minor is prevalent in Mediterranean countries [1].
The differential diagnosis between IDA and
thalassemia minor is important, so that unnecessary
iron therapy or analysis for the source of the
Acknowledgements
We would like to express our thanks and gratitude to
the Department of Medical Laboratories at Sabah
Hospital for their assistance and to all subjects
participated in this study. We also thank Mr Mustafa
Jaffar for providing technical help.
Correspondence: Dr Suad M. AlFadhli, Department of
Medical Laboratory Sciences, Faculty of Allied Health
Sciences, Kuwait University, P. O. Box 31470, Sulaibekhat
90805, Kuwait. Tel.: þ965-9445422; Fax: þ965-483361;
E-mail <s.alfadhli@hsc.edu.kw>.
avoided in thalassemia minor. Routine screening
tests used for the differentiation between IDA and
thalassemia minor include; complete blood count
(CBC), serum iron, serum ferritin, total iron binding
capacity (TIBC), bone marrow iron stores, levels of
Hb A2, free erythrocyte protoporphyrin and zinc
protoporphyrin levels [1–6].
Despite their usefulness, these tests are often
expensive and time consuming. Routine electronic
red blood cell counts and indices driven from modern
blood cell analyzers can provide fast, cheap and
valuable clues as to whether IDA or thalassemia
minor may be present.
In this study we evaluate nine discriminant
functions used in the differentiation between IDA
and thalassemia minor by applying them on fully
diagnosed cases of microcytic hypochromic anemias.
Materials and Methods
Over a period of 6 months, 153 patients with
confirmed microcytic anemias were studied. CBC,
serum iron levels, Hb A2 and detection of Hb H
inclusion bodies were done to confirm diagnosis.
Fifty-six patients with Hb levels of <110 g dl
1,
MCV<72 fL and serum iron <5 mmol l
1 were

ß The Author [2006]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org 93
doi:10.1093/tropej/fml070 Advance Access Published on 13 December 2006

TABLE 1
Mathematical discriminant functions used to differentiate between iron deficiency anemia and thalassemia minor
Formula Equation
Mentzer Index (MI) MCV/RBC
2
Shine and Lal (S & L) Index (MCV)  MCH
England and Fraser (E & F) Index MCV
(Hb  5)
RDW
ka
2
Green and King (G & K) Index [(MCV)  RDW]/(Hb  100)
RDW Index (RDWI) MCV  RDW/RBC
Srivastava Index (SI) MCH/RBC
Ricerca Index (RI) RDW/RBC
a
In the counter used in this study, k was calculated to be 8.4.
diagnosed to have IDA. Forty-seven patients
were diagnosed to have b-thalassemia minor based
on low Hb and MCV levels, normal serum iron
and elevated levels of Hb A2 (>3.3%). The remaining
50 patients were diagnosed to have a-thalassemia
based on the detection of Hb H inclusion bodies
in peripheral blood smear stained with Brilliant
Crystal Blue stain.
CBC results were obtained using Beckman
Coulter HMX Automated Hematology Analyzer,
USA. Serum iron, TIBC and transferring saturation
were determined using BN ProSpecÕ System from
Dade Behring, Inc, Germany, while serum ferritin
was measured using Elecsys System from Roche
Diagnistics, Germany. Hb A2 levels were determined
using b-thal Hb A2 Quick Column kit from Helena
Biosciences, UK.
Patients, results were analyzed individually using
the nine discrimination functions in this study.
Sensitivity, specificity and Youden’s Index (a mea-
sure of validity) [(sensitivity þ specificity)
100] were
calculated for each function. Data were analyzed
using Microsoft Office Excel program 2003.
In addition to RBC count and red blood cell
distribution width (RDW), the following mathema-
tical functions were also included in our analysis:
Mentzer Index (MI), Shine and Lal (S & L) Index,
England and Fraser (E & F) Index, Green and King
(G & K) Index, RDW Index (RDWI), Srivastava
Index (SI) and Ricerca Index (RI) (Table 1).
Results
The means and standard deviations of CBC
parameters, serum iron and Hb A2 in the 153
patients investigated in this study are shown in
(Table 2).
94
S. M. ALFADHLI ET AL.
Cut off values Reference
TM < 13 7
IDA > 13
TM < 1530 8
IDA > 1530
TM < 0 9
IDA < 0
TM < 72 10
Downloaded from https://academic.oup.com/tropej/article-abstract/53/2/93/1660096 by guest on 16 August 2019
IDA > 72
TM < 220 11
IDA > 220
<4.4 TM 12
>4.4 IDA
<3.3 TM 13
>3.3 IDA
In all cases studied RBC count was found to be lower
than normal range in IDA (4.32  0.42  1012 l
1),
while values were within normal range in b- and
a-thalassemia (5.39  0.89 and 4.98  0.82, respec-
tively). Hemoglobin level was decreased in the three
anemias studied, however, the level of decrease in IDA
was greater than that in thalassemia. MCV was also
low in all cases although the level of decrement in
b-thalassemia (63.4  7.12 fL) cases was greater than
that found in IDA (67  5.9 fL) and a-thalassemia
(66.5  6.45 fL). As expected serum iron levels were low
in IDA (3.4  1.2 mmol l
1), while Hb A2 levels were
high in b-thalassemia minor (5.97  1.32%). All cases
of a-thalassemia showed Hb H inclusions confirming
the proposed diagnosis. IDA cases were also confirmed
by high TIBC levels with an average of 3.8  0.4 g l
1
(normal range 2.1–3.5 g l
1) and low transferring
saturation levels with an average of 3.7  1.3%
(normal range 20–40%). Ferritin (measured only in 28
cases of IDA) was low with an average of
4.8  1.8 ng ml
1 (normal range 13–400 ng ml
1).
The nine investigated discriminant functions were
applied on all microcytic anemia cases in this study,
including a-thalassemia cases (Table 3). E & F Index
identified 100% of b- and a-thalassemia cases and
almost 100% (98.2%) of IDA cases. Although, RBC
count perfectly identified (100%) cases of IDA, only
78.7 and 58% of b- and a-thalassemia cases
respectively, correctly fit the cut off value of this
index.
S & L Index had the lowest number of correctly
identified cases of IDA as only one case was correctly
identified from 56 diagnosed cases (1.7%), while
the RDW had the lowest number of correctly
identified cases of b- and a-thalassemia (6.3 and
12%, respectively). Almost all discriminant functions
(except for RDW) had higher percentages of
Journal of Tropical Pediatrics Vol. 53, No. 2
S. M. ALFADHLI ET AL.

5.97  1.32
2.53  0.59
correctly identified cases of b-thalassemia compared

2  0.4
2–3.3%
Hb A2
with a-thalassemia.
It is worth noting that thalassemia cases were
compared separately to IDA cases. The discriminant
functions are not intended to be used to differentiate

36.54  12.65
33.76  14.55
between b- and a-thalassemia.

6–37 mmol/ l

3.4  1.2
In order to determine, the best function to be used

S.Fe
in differentiating IDA from b- and a-thalassemia,
sensitivity, specificity and Youden’s Index were
calculated (Table 4). In differentiating IDA from
b-thalassemia, the Youden’s Index from highest

Downloaded from https://academic.oup.com/tropej/article-abstract/53/2/93/1660096 by guest on 16 August 2019

to lowest was as follows: E & F > RBC > MI >

17.41  2.36
17.54  4.13
11.5–14.5%

17.06  2 SI > RI > RDWI > G & K > RDW > S & L. In differ-
RDW

entiating IDAs from a-thalassemia, the Youden’s

Hematological and biochemical parameters obtained from the screened patients.

Index from highest to lowest was as follows:

E & F > RBC > RDWI > RI > G & K >SI > MI >
RDW > S & L.
314.89  05.62
316.48  14.31
316.4  18.1
320–360 g/l
MCHC

Discussion
Several studies have derived discriminant functions
based on RBC indices that can be used to differ-
entiate IDA from thalassemia minor. The differential
Results are expressed as mean  SD

diagnosis of these two hypochromic microcytic

19.61  3.60
21.09  2.55

anemias is an everyday issue for physicians; accurate

21.4  2.9
27–31 pg
MCH

diagnosis is essential for useful treatment, prevention

of disease and cutting down expenses.
In this study we evaluated nine discriminant
TABLE 2

functions used in the differentiation analysis of 153

diagnosed cases of microcytic anemias. We attempted
63.40  7.12
66.50  6.45
67  5.9
81–99 fL

to include all well documented functions/indices in

MCV

this work in order to provide a wholesome analysis to

select the best formula to be used in differentiation
studies.
All investigated cases, confirmed to the set criteria
for each anemia [2, 14]. IDA cases were confirmed
0.42–0.52 l/l

0.29  0.03
0.34  0.05
0.33  0.06

with low RBC, Hb, MCV, serum iron, transferring

PCV

saturation, serum ferritin and high TIBC.

b-thalassemia cases were confirmed with low Hb
and MCV, high Hb A2 levels and normal serum iron.
a-thalassemia cases were conformed with the
106.51  15.36
104.84  19.98

presence of Hb H inclusions upon staining with

92.6  12.7
140–180 g/l

Brilliant crystal blue stain.

Hb

Although RDW, a measure of the degree of

variation in red cell size, has been reported to be a
good discrimination index to differentiate between
IDA and thalassemia minor [4, 5, 14, 15], our results
suggest that this index may be misleading as all
4.7–6.11012/l

4.32  0.42
5.39  0.89
4.98  0.82

thalassemia (a and b) cases had an equally elevated

RBC

level of RDW compared with IDA. Several studies

had also reported that RDW alone is not sufficiently
specific or sensitive enough to differentiate between
microcytic anemias, as elevated values were not
specific for IDA [16–19]. This may be explained by
56
47
50
N

the fact that an elevated RDW may be caused by

several factors like, erythrocytosis, presence of target
Anemia

cells and elevated reticulocyte counts [16, 18], all of

b-thals
a-thals

which may be present in thalassemia minor cases.

IDA

When validity analysis was applied on RDW,

Journal of Tropical Pediatrics Vol. 53, No. 2 95

 S. M. ALFADHLI ET AL.

TABLE 3
Cut-off values and number of cases correctly identified by the nine discriminant functions for the
three anemias investigated

Discriminant function Cut–off values IDA (n ¼ 56) b-thals (n ¼ 47) a-thals (n ¼ 50)

MI IDA > 13; Thals < 13 48 (85.7%) 35 (74.4%) 22 (44%)

S&L IDA > 1530; Thals < 1530 1 (1.7%) 46 (97.8%) 48 (96%)
SI IDA > 4.4; Thals < 4.4 39 (69.6%) 40 (85%) 30 (60%)
RI IDA > 3.3; Thals > 3.3 48 (85.7%) 31 (65.9) 26 (52%)
RDWI IDA > 220; Thals < 220 47 (83.9%) 31 (65.9%) 30 (60%)
G&K IDA > 72; Thals < 72 41 (73.2%) 35 (74.4%) 30 (60%)

Downloaded from https://academic.oup.com/tropej/article-abstract/53/2/93/1660096 by guest on 16 August 2019

E&F IDA > 0; Thals < 0 55 (98.2%) 47 (100%) 50 (100%)
RBC count IDA < 5; Thals > 5 56 (100%) 37 (78.7%) 29 (58%)
RDW IDA > 14; Thals < 14 55 (98.2%) 3 (6.3%) 6 (12%)

this index had one of the lowest Youden’s Index TABLE 4

values (Table 4). Sensitivity, specificity and Youden’s Index of each
Youden’s Index provides an appropriate discriminant function used in differentiation between
measure of validity of a particular question or IDA and b-thalassemia and IDA and a-thalassemia
technique [20]. This index depends on sensitivity
and specificity; it does not take into consideration the Discriminant Sensitivity Specificity Youden’s
positive and negative predictive values, as these function (%) (%) Index
values depend strongly on the underlying prevalence
MI IDA 85.7 74.4 60.1
of the disease. The use of positive and negative
b-thals 74.4 85.7
predictive values alone in validity analysis has been IDA 85.7 44 29.7
a source of contradicted results in previous reports a-thals 44 85.7
[21, 22]. The Youden’s Index introduces the least bias S&L IDA 1.7 97.8 0.5
to the measure of effect [20]. b-thals 97.8 1.7
Youden’s Index has been used previously IDA 1.7 96 2.3
to compare validity of discrimination indices in a-thals 96 1.7
differentiating between IDA and thalassemia SI IDA 69.6 85 54.9
b-thals 85 69.6
minor [23]. In a study done by Demir, et al. [23], IDA 69.6 60 29.9
it was reported that RBC count and RDWI are a-thals 60 69.6
the most reliable discrimination indices in differen- RI IDA 85.7 65.9 51.6
tiating between IDA and b-thalassemia trait. b-thals 65.9 85.7
We found that E & F Index to be the most valid IDA 85.7 52 37.7
discrimination index to correctly differentiate a–thals 52 85.7
between IDA and b-thalassemia trait (Youden’s RDWI IDA 83.9 65.9 49.8
b-thals 65.9 83.9
Index ¼ 98.2), the RBC count came second with
IDA 83.9 60 43.9
Youden’s Index of 78.8, while RDWI rated at a-thals 60 83.9
number 6 of the nine functions analyzed. G&K IDA 73.2 74.4 47.6
The difference in findings may be due to the small b-thals 74.4 73.2
number of patients in Demir’s study (n ¼ 63) IDA 73.2 60 33.2
compared with our study (n ¼ 153). Demir, et al. a-thals 60 73.2
also examine a pediatric population, while patients in E&F IDA 98.2 100 98.2
b-thals 100 98.2
this study were older than 16 years of age.
IDA 98.2 100 98.2
We took our work one step further and investi- a–thals 100 98.2
gated whether the same discriminant functions RBC IDA 100 78.7 78.7
can be applied to differentiate between IDA b-thals 78.7 100
and confirmed cases of a-thalassemia. Again IDA 100 58 58
Youden’s Index for E & F and RBC count rated a-thals 58 100
the highest of the formulas analyzed (98.2 and RDW IDA 98.2 6.3 4.5
78.7, respectively) suggesting that they were the b-thals 6.3 98.2
IDA 98.2 12 10.2
most reliable in differentiation between IDA and a-thals 12 98.2
a-thalassemia.

96 Journal of Tropical Pediatrics Vol. 53, No. 2

S. M. ALFADHLI ET AL.
Except for S & L Index and RDW, all other
functions had a lower Youden’s Index when IDA and
a-thalassemia were compared than when IDA and
b-thalassemia were compared (Youden’s Index for
E & F Index was the same in both comparisons).
This suggests, that it is probably more difficult to
differentiate between IDA and a-thalassemia than it
is to differentiate between IDA and b-thalassemia
using discriminant functions alone (also note close
MCV, MCH and MCHC values; Table 2). Therefore,
iron studies and Hb H detection must be incorpo-
rated for an accurate diagnosis.
S & L formula showed the lowest value for
Youden’s Index in both comparisons (0.5 and
2.3 in comparing IDA with b-thalassemia and
a-thalassemia, respectively) suggesting that this
formula is the least effective in discrimination
studies for microcytic anemias. A Youden’s Index
of a value close to zero suggests that the technique
used is no better than random. The ineffectiveness
of S & L formula has been reported previously [23].
In conclusion, although none of the functions used
showed absolute sensitivity (100%) in differentiation
IDA and thalassemia minor, we report the E & F
Index to be the most reliable discriminant function
to differentiate between IDA and thalassemia minor
(a or b), conversely, S & L formula appears
ineffective in our population. We also highlight the
use of the Youden’s Index as a reliable measure
of validity for the use of different functions in
differentiation between microcytic anemias.
References
1. Harmening DM, Clinical Hematology and
Fundamentals of Hemostasis 4th edn. Philadelphia:
F.A. Davis Company, 2002, pp. 104–87.
2. El-Agouza I, Abu Shahla A, Sirdash M. The effect of
iron deficiency anemia on the levels of hemoglobin
subtypes: possible consequences for clinical diagnosis.
Clin Lab Haem 2002;24:285–9.
3. Massey AC. Microcytic anemia. Differential diagnosis
and management of iron deficiency anemia. Med Clin
North Am 1992;73:549–66.
4. Novak RW. Red blood cell distribution width in
pediatric microcytic anemias. Pediatrics 1987;80:251–4.
5. Junca J, Flores A, Roy C, et al. Red cell distribution
width, free erythrocyte protoporphyrin, and
England–Fraser index in the differential diagnosis of
microcytosis due to iron deficiency or beta-thalassemia
trait. A study of 200 cases of microcytic anemia.
Hematol Pathol 1991;5:33–6.
6. van Tellingen A, Kuenen JC, de Kieviet W, et al.
Iron deficiency anemia in hospitalized patients: value
Journal of Tropical Pediatrics Vol. 53, No. 2
of various laboratory parameters. Differentiation
between IDA and ACD. Neth J Med 2001;59:270–9.
7. Mentzer WC. Differentiation of iron deficiency from
thalassemia trait. Lancet 1973;1:882–4.
8. Shine I, Lal S. A strategy to detect beta-thalassemia
minor. Lancet 1977;1:692–4.
9. England JM, Fraser P. Discrimination between
iron-deficiency and heterozygous thalassemia
syndromes in differential diagnosis of microcytosis.
Lancet 1979;1:145–8.
10. Green R, King R. A new red blood cell discriminant
incorporating volume dispersion for differentiating iron
Downloaded from https://academic.oup.com/tropej/article-abstract/53/2/93/1660096 by guest on 16 August 2019
deficiency anemia from thalassemia minor. Blood cells
1989;15:481–95.
11. Jayabose S, Giavanelli J, Levendoglu-Tugal O, et al.
Differentiating iron deficiency anemia from thalassemia
minor by using RDW-based index. J Pediatr Hematol
1999;21:314.
12. Srivastava PC, Bevington JM. Iron deficiency and-or
thalassemia trait. Lancet 1973;1:832–5.
13. Ricerca BM, Storti S, d’Onofrio G, et al.
Differentiation of iron deficiency from thalassemia
trait: a new approach. Haematologica 1987;75:409–13.
14. Clarke G, Higgins T. Laboratory investigation of
hemoglobinopathies and thalassemia: review and
update. Clin Chem 2000;46:1284–90.
15. Yermiahu T, Ben-Shalom M, Porath A, et al.
Quantitative determinations of microcytic–
hypochromic red blood cell population and
glycerol permeability in iron-deficiency anemia and
b-thalassemia minor. Ann Hematol 1999;78:468–71.
16. Roberts GT, El-Badawi SB. Red cell distribution
widths index in some hematologic diseases. Am J Clin
Pathol 1987;83:222–6.
17. Marsh WL, Jr, Bishop JW, Darcy TP. Evaluation of
red cell volume distribution width (RDW). Hematol
Pathol 1987;1:117–23.
18. Flynn MM, Reppun TS, Bhagavan NV. Limitations of
red blood cell distribution width (RDW) in evaluation
of microcytosis. Am J Clin Pathol 1986;85:445–9.
19. Burk M, Arenz J, Giagounidis AA, et al. Erythrocyte
indices as screening tests for the differentiation of
microcytic anemias. Eur J Med Res 1995;1:33–7.
20. Pekkanen J, Pearce N. Defining asthma in epidemio-
logical studies. Eur Respir J 1999;14:951–7.
21. Sirdash M, Bilto YY, El Jabour S, et al. Screening
secondary school students in the Gaza strip for
b-thalassemia trait 1998;20:279–83.
22. Laffery JD, Crowther MA, Ali MA, et al.
The evaluation of various mathematical RBC
indices and their efficacy in discriminating between
thalassemic and non-thalassemic microcytosis 1996;106:
201–5.
23. Demir A, Yarali N, Fisgin T, et al. Most reliable
indices in differentiation between thalassemia
trait and iron deficiency anemia. Pediatr Int 2002;44:
612–16.
97