TOG9_3_189-194

10/7/07

5:39 AM

The Obstetrician & Gynaecologist

Page 189

10.1576/toag.9.3.189.27340 www.rcog.org.uk/togonline

2007;9:189194

Education

Education EMQs: a new

component of the MRCOG
Part 1 examination
Authors Suresh Duthie / Alison Fiander / Paul Hodges

Key content:
Extended Matching Questions (EMQs) are going to be introduced into the Part 1
MRCOG examination from September 2007.
One of the main aims is to direct the learning of aspects of biological sciences that
are relevant to the clinical practice of obstetrics and gynaecology.
The advantages of the EMQ format over the multiple-choice format include the
ability to test a candidates understanding of the relationship between facts.

Learning objectives:
To understand how the EMQ format can lead to better testing of knowledge of
basic and clinical sciences and increased examination validity.
To be aware of the additional discrimination EMQs have demonstrated in the first
Part 2 paper.
To be aware of the new bioscience and curriculum context of the MRCOG
examinations.

Ethical issues:
Improving the reliability and validity of the Part 1 MRCOG examination
contributes to setting standards in women's health.
Keywords EMQs / MCQs / Part 1 MRCOG
Please cite this article as: Duthie S, Fiander A, Hodges P. EMQs: a new component of the MRCOG Part 1 examination. The Obstetrician & Gynaecologist 2007;9:189194.

Author details
Suresh Duthie FRCOG
Consultant Obstetrician and
Gynaecologist
Blackpool, Fylde and Wyre Hospitals NHS
Trust, Blackpool Victoria Hospital, Whinney
Heys Road, Blackpool FY3 8NR, UK

Alison Fiander DM FRCOG

Chair of MRCOG Part 1 Subcommittee
and Chair of Obstetrics and
Gynaecology
Wales College of Medicine, Cardiff University,
Heath Park, Cardiff CF14 4XN, UK

2007 Royal College of Obstetricians and Gynaecologists

Paul Hodges PhD

Deputy to Head of Examinations
Department
Royal College of Obstetricians and
Gynaecologists, 27 Sussex Place, Regents
Park, London NW1 4RG, UK
Email: [email protected]
(corresponding author)

189

TOG9_3_189-194

Education

10/7/07

5:39 AM

Page 190

2007;9:189194

Introduction
What is the best way to prepare a doctor to become
a specialist obstetrician and gynaecologist in this
new century? Clinical and medical science is
growing faster than ever and investigations and
treatments derived from the new bioscience of the
1980s and 1990s are now having an impact on
clinical practice. Clinical consultants are being
asked to use these tools properly and to advise
women of their use. Senior clinicians need to be
able to engage with scientific innovations,
understand the opportunities and limitations of
new techniques and tools introduced into medicine
and to be confident when appraising new
technologies. This confidence can only follow from
an adequate grounding in clinical science.
Ultrasound, laparoscopy, in vitro fertilisation, fetal
blood sampling and prenatal diagnosis are now all
used in day-to-day clinical practice. Molecular
biology and genetics have clinical applications in
prenatal diagnosis, gynaecological oncology and
reproductive medicine. Advances in gene therapy
and stem cell science will, undoubtedly, be
translated into practice in the next decade. Clinical
epidemiology, statistics and an understanding of
clinical risk are other relatively new concepts that
are becoming increasingly deeply embedded within
clinical practice. In contrast, some subjects that the
doctor in training needs to know are unchanging,
such as embryology and the surgical anatomy of the
pelvis. Knowledge continues to be the single best
determinant of expertise in a subject1 and it
remains essential that a doctor in training
understands, assimilates and retains knowledge of
the basic and clinical sciences. The aim of the Part 1
MRCOG is, therefore, to direct the learning of
aspects of biological science that are relevant to the
clinical practice of obstetrics and gynaecology.
The new Core Curriculum allows the MRCOG
examinations to be blueprinted2 and maps the
Part 1 MRCOG as a component part of the overall
process of assessment for future consultants in the
specialty.3 The Part 1 and Part 2 MRCOG
examinations will, from September 2007, form a
single integrated and blueprinted examination of
relevant factual knowledge and clinical skills in the
specialty. Various preparations have been made for
this endpoint, including the ending of the
exemption scheme for Part 1.
The new syllabus will be clearer, better defined and
less broad than that currently in place. However, as
mentioned above, the syllabus has been updated to
include important new bioscience topics.4 This is so
throughout the syllabus and is made particularly
explicit in the Core Curriculum Matrix. Topic areas
included as distinct areas for the first time include:
virology, clinical trial design and analysis,
molecular and cell biology, genomics and
regulation of gene expression.
190

The Obstetrician & Gynaecologist

Educational and evaluative

benefits of EMQs
The testing that supports this new curriculum and
syllabus must be thorough and efficient.
Throughout the process of standard setting, the
pass/fail threshold must be clearly exposed. Those
candidates with a score above the pass mark must
pass and those below must fail. The borderline
candidate must be treated fairly. But what is it that
we are actually trying to measure? It is vital to
measure a candidates ability to understand and
apply knowledge rather than simply to reward the
recall of isolated facts. This reflects the trend
towards integration of clinical sciences teaching
into integrated curricula within UK medical
schools. There is generally a move towards basic
medical sciences being taught by clinically qualified
staff and an overall reduction in the clinical sciences
content of medical schools curricula. Rote learning
without understanding is a potential pitfall of the
learning process. Another pitfall of true/false
multiple choice questions (MCQs) is that a testwise candidate might guess correctly too often and
gain a false pass.
The use of MCQs in postgraduate medical
examinations in the UK is well established.
Previously, the Part 1 MRCOG examination was
based entirely on assessment by MCQs. The format
has several advantages, including easy, wide subject
coverage and efficient use of testing time in a
consistent and reliable manner. However, the ability
of the MCQ format to probe a candidates actual
understanding of the subject can be limited and,
therefore, the College is going to introduce the
EMQ format as a component of the Part 1 MRCOG
examination from September 2007 onwards. The
advantages of the EMQ format over the MCQ
format include the ability not only to test the recall
of facts but to test a candidates understanding of
the relationship between facts, as the examples
provided demonstrate. Most EMQs will define their
clinical relevance quickly, which is not easy to
accomplish with the MCQ format.
The structure of the EMQs for the Part 1 MRCOG
examination will be identical to that used in the
Part 2, which has been described previously.5
Briefly, an EMQ is comprised of three parts:

a list of options
a lead-in statement
items (questions).
In the examination a single EMQ will consist of a
list of options with one to four items or questions,
avoiding overemphasis on a single subject. In
answering EMQs, candidates are advised not to
read through the list of options first to avoid being
wrongly cued by distractors among them. The leadin statement and then the items or questions should
2007 Royal College of Obstetricians and Gynaecologists

TOG9_3_189-194

10/7/07

5:39 AM

Page 191

The Obstetrician & Gynaecologist

2007;9:189194

be read carefully before selecting the correct answer

from the list of options.
The same guidelines will be followed as for Part 2
MRCOG EMQs. Although their less clinical nature
should make them somewhat shorter and simpler,
carefully reading the lead-in statement, the link
between the items (questions) and the list of
options before attempting to answer the questions
will remain vital.
Although the Part 2 EMQs have only been run in
trials and at one actual examination sitting so far
and there is, therefore, a lack of longitudinal data,
early data suggest that the format is performing

Education

Table 1

Part 2 MRCOG written examination September 2006

Discrimination
descriptor

Overall point
biserial figure
range

Percentage of MCQ
format questions
achieving

Percentage of
EMQ format
questions
achieving

Moderate to high
High
Very high

0.10.3
0.30.4
 0.4

50.4
4.4
1.3

57.5
27.5
10.0

Comparison of the discrimination

abilities of MCQ (T/F) and EMQ
question items (one Part 2 sitting
only)

very well. As it was hoped, based on the educational

literature,5 EMQs have so far proved to be excellent
discriminators. Candidates who performed well
overall in the Part 2 written examination generally
Syphilis is caused by Treponema pallidum

TRUE/FALSE

Figure 1

Example 1 (true/false MCQ)

Figure 2

Options

Example 2 (EMQ 1)
Maternal disease

Type

Risk of transplacental infection

Babesiosis

Protozoan

No

Babesiosis

Protozoan

Yes

Coccidioidomycosis

Fungus

No

Coccidioidomycosis

Fungus

Yes

Malaria

Bacterium

Yes

Malaria

Protozoan

No

Malaria

Protozoan

Yes

Malaria

Virus

Yes

Fever

Rickettsia

Yes

Q fever

Virus

Yes

Schistosomiasis

Helminth

Yes

Schistosomiasis

Protozoan

Yes

Syphilis

Protozoan

Yes

Syphilis

Spirochete

No

Syphilis

Spirochete

Yes

Syphilis

Virus

Yes

Tuberculosis

Bacterium

No

Tuberculosis

Bacterium

Yes

Tuberculosis

Fungus

No

Tuberculosis

Protozoan

No

Instructions: The options above refer to different maternal diseases caused by micro-organisms, the type of organism and
whether there is a risk of transplacental infection. Select the single correct profile for each of the micro-organisms in the items
below. Each option may be used once, more than once or not at all.
ITEM 1
Treponema pallidum
Answer
O. Syphilis; Spirochete; Yes
ITEM 2
Mycobacterium tuberculosis
Answer
R. Tuberculosis; Bacterium; Yes
ITEM 3
Plasmodium falciparum
Answer
G. Malaria; Protozoan; Yes

2007 Royal College of Obstetricians and Gynaecologists

191

TOG9_3_189-194

10/7/07

5:39 AM

Education

Page 192

2007;9:189194

The Obstetrician & Gynaecologist

Options
A

Embryological origin

Site of biosynthesis and secretion of

Arterial blood supply

Endoderm epithelium

Catecholamines

Branches from the coeliac axis of the aorta

Coelomic wall epithelium

Aldosterone, cortisol, dehydroepiandrosterone sulphate

Branches from the aorta, phrenic and renal arteries

Neural plate epithelium

Aldosterone, cortisol, dehydroepiandrosterone sulphate

Branches from the splenic artery

Neural crest

Catecholamines

Branches from the aorta, phrenic and renal arteries

Mesonephric duct

Aldosterone, cortisol, dehydroepiandrosterone sulphate

Branches from the aorta, phrenic and renal arteries

Neural crest

Catecholamines

Superior mesenteric artery

Paramesonephric duct

Aldosterone, cortisol, dehydroepiandrosterone sulphate

Branches from the aorta, phrenic and renal arteries

Endoderm epithelium

Cortisol

Branches from the aorta, phrenic and splenic arteries

Instructions: The list of options describes the profile of human organs in terms of embryological origin, the substances that are produced at that particular site and the
arterial blood supply. Select the most closely matching profile for each of the structures in the items below from the list of options. Each option may be used once, more
than once or not at all.
ITEM 1
Adrenal cortex
Answer
B
ITEM 2
Adrenal medulla
Answer
D

Figure 3

Example 3 (EMQ 2)

Figure 4

Example 4 (EMQ 3)

achieved high marks on the EMQ paper and

candidates who performed poorly overall received
low marks. At a more individual level, each
questions statistical measure of discrimination
Options
A

3-hydroxyisovaleric acid

17 -hydroxyprogesterone

17 -hydroxyprogesterone

17 -hydroxyprogesterone

Acetic acid

Adrenocorticotrophic hormone

Bilirubin

Ceramide trihexoside

Citrulline

Cortisol

K`

Cystine

Galactitol

Glutaric acid

Insulin

Methylmalonic acid

Succinylacetone

Instructions: The level of certain metabolites in amniotic

fluid changes significantly if the fetus has an inborn error
of metabolism. Select the single metabolite whose level in
amniotic fluid is altered by the inherited diseases in the
items below. Each option may be used once, more than
once or not at all.

(Pearson point biserial) was good. Indeed, every

EMQ had a positive point biserial, indicating that it
was discriminating in the correct way. Even taking
into account the inbuilt advantage of rather higher
weighting,6 Part 2 EMQs comfortably
outperformed the MCQs of the same examination
sitting with regard to discrimination (Table 1).
Because of the multiple marks awarded within the
Part 2 Short Answer Questions (SAQs) marking
schemes,7 a direct comparison is not feasible but
these early indications suggest that EMQs at least
match or slightly exceed them in terms of
discrimination, while being significantly more
reliable and reproducible. This is in line with the
educational literature.8 Also, it is hard to envisage
the utility of the SAQ format for Part 1.
Good EMQ writing techniques are now embedded
within the College and are based on established
research9 and guidelines.10 As such, there is no
reason to believe that the new Part 1 EMQs cannot
emulate the promising performance of their
clinically-based predecessors. This should produce
a more sophisticated and relevant examination that
is also fairer to borderline candidates, thus
improving the validity and reliability of the Part 1
examination.

ITEM 1
Congenital adrenal hyperplasia
Answer
B.

192

17 -hydroxyprogesterone

Examples
Example 1 (Figure 1) is a typical true/false MCQ
Part 1 question, which most candidates should be
able to answer successfully. Those who do not
know the answer have a 50:50 chance of guessing
2007 Royal College of Obstetricians and Gynaecologists

TOG9_3_189-194

10/7/07

5:39 AM

Page 193

The Obstetrician & Gynaecologist

the correct answer. In obstetric and

gynaecological practice, what is it that we expect a
candidate to know about Treponema pallidum,
other than being the cause of syphilis? There are
several facts that would help the doctor in
training prepare for a future career in the
specialty.
Example 2 (Figure 2) is a typical Part 1 EMQ on a
similar area. It has 20 options, a very clear lead-in
statement and three items. The lead-in statement
explains to the candidate that the list of options
refers to different maternal diseases caused by
micro-organisms. The second column in the list of
options shows different types of microbes and the
third column indicates whether or not there is a risk
of transplacental infection. The examination
candidate is asked to select the correct profile from
the list of options for each of the questions in the
worked example. Using this format, the
examination verifies that the candidate
understands that:

T. pallidum is the cause of syphilis

T. pallidum is a spirochete
there is a risk of transplacental infection.
There are, of course, many other facts that are also
relevant, including the mode of transmission,
clinical features (typical and atypical),
epidemiology, diagnosis and sensitivity to
penicillin. These can be incorporated into the test
profile. Equally, the profile connecting different
facts about another organism is tested in Item 2.
The candidate must avoid the distractors that are
present and select the option that describes
Mycobacterium tuberculosis as:

2007;9:189194

Options
A

Adrenocortical cancer and melanoma

Brain tumours, sarcoma of bone and soft tissue

Breast cancer and ovarian cancer

Breast cancer and retinoblastoma

Breast cancer, brain tumours, adrenocortical cancer,

leukaemia, sarcoma of bone and soft tissue

Breast cancer, brain tumours, endometrial cancer

Breast cancer, ovarian cancer, adrenocortical

cancer, acute myeloid leukaemia

Cancer of the rectum and cancer of the stomach

Cancer of the rectum, cancer of the endometrium,

cancer of the stomach and cancer of the bile duct

Colonic cancer and cancer of the stomach

Colorectal cancer and bronchopulmonary cancer

Colorectal cancer, cancer of the endometrium,

cancer of the stomach and cancer of the bile duct

Ovarian cancer

Ovarian cancer and retinoblastoma

Retinoblastoma

Retinoblastoma and adenocarcinoma of the cervix

Retinoblastoma and adrenocortical cancer

Retinoblastoma and cancer of the bladder

Retinoblastoma and cancer of the endometrium

Retinoblastoma and osteosarcoma

Education

Figure 5

Example 5 (EMQ 4)

Instructions: The list of options describes different

malignant conditions. The items below refer to
conditions that increase tumour susceptibility. Select
the single option that indicates which tumour(s) an
individual would be at increased risk of acquiring for
each of the conditions in the items below. Each option
may be used once, more than once or not at all.
ITEM 1
BRCA1
Answer
C.

Breast cancer and ovarian cancer

ITEM 2
BRCA2

the cause of tuberculosis

a bacterium
capable of transplacental infection.

Answer
C.

Breast cancer and ovarian cancer

A similar profile is tested for Plasmodium

falciparum in Item 3.

ITEM 3

Example 3 (Figure 3) uses a similar structure,

based on the knowledge point that the adrenal
cortex is derived from the embryonic coelomic
wall epithelium and the adrenal medulla is derived
from embryonic neural crest. It is important for a
candidate to realise that the adrenal gland is the
site of production of various hormones that are
central to human reproductive physiology and
adaptation to stress. As shown, this EMQ tests
understanding of:

Answer

the embryological origin of the components of

the adrenal gland
the hormones produced by the components of
the adrenal gland
the arterial blood supply of the adrenal gland.
2007 Royal College of Obstetricians and Gynaecologists

Hereditary retinoblastoma

T.

Retinoblastoma and osteosarcoma

While it is important not to make the question

resemble a verbose crossword puzzle, this format
can comprehensively cover the Part 1 syllabus, this
example encompassing anatomy, embryology and
biochemistry.
The use of information presented in tables or charts
tests a candidates understanding of the relationship
between facts. However, the EMQ format can also be
used to test the recall of isolated facts, as shown by
the next example (Figure 4). This is a particularly easy
193

TOG9_3_189-194

10/7/07

5:39 AM

Education

Figure 6

Example of possible lists of options

Page 194

2007;9:189194

Nucleic acid base pairs

Types of receptors (cell membrane/nuclear)
Transport proteins and their ligands
Nutrients, such as amino acids
Lymphatic drainage of pelvic and abdominal structures
Branches of nerves and the innervation of structures
Branches of blood vessels and supply of organs/structures
Lengths of anatomical structures
Relationships between anatomical structures
Types of statistical tests and their relevance to clinical trials
Stages in the development of the gamete and human
embryo and the relationship to congenital abnormalities
Actions of pharmacological agents, side effects and
interactions
Shifts in the haemoglobin-oxygen dissociation curve
Immune effector cells and immune responses in health and
disease
Micro-organisms in obstetrics and gynaecology

The Obstetrician & Gynaecologist

should be noted that the quality of questions

provided by many commercial online revision
websites and courses is far below those used by the
College.
Figure 6 gives examples of possible lists of options,

some of which are combinations that could be used

in the construction of tables of options. Their
relevance or application to clinical practice is more
readily seen compared with MCQ answers.

Conclusion
The key points are:

Knowledge is the single best determinant of

expertise in a subject.
It is important to understand the relationship

Histological appearances of organs relevant to obstetrics

and gynaecology in health and disease
Hormones and their actions

question for the adequately prepared candidate.

However, there are a sufficient number of distractors
in the list of options to aid in discriminating between
a knowledgeable candidate and a test-wise
candidate, who is more dependent on guess work.
Functional distractors can be used in the list of
options in an EMQ to distract the borderline or
weak candidate who is not entirely sure of the
answer. A functional distractor can be based on a
real condition or a hypothetical (imaginary)
condition, as example 5 (Figure 5) demonstrates.
Item 1 asks about tumour susceptibility for BRCA1
mutation; most candidates would know that breast
cancer is a significant risk. However, five of the
options (C, D, E, F and G) contain breast cancer. The
candidate needs to be sure that option C is the
correct answer. Retinoblastoma is a rare condition,
but hereditary retinoblastoma increases the relative
risk of acquiring osteosarcoma by a factor of
approximately 2000. Therefore, the answer to Item 3
is T. Again, there are several functional distractors
for the candidate. Both the list of options and the
items can be varied in future as the scientific
community in the field of tumour susceptibility
acquires further knowledge.
Further examples, including some without
answers, are provided on the RCOG website.11 It

194

between different facts when studying the

scientific foundations of obstetrics and
gynaecology, i.e. the application of basic clinical
sciences to clinical practice.
EMQs are useful in testing the understanding of a
subject.
EMQs will complement the high reliability of the
MCQ format in the Part 1 MRCOG examination
whilst enhancing validity.
Good preparation and familiarity with the EMQ
and MCQ formats will be the key to passing the
examination.

References
1 Case SM, Swanson DB. Constructing Written Test Questions for the
Basic and Clinical Sciences. 3rd ed. Philadelphia: National Board of
Medical Examiners; 2001.
2 Hodges P. Blueprinting the RCOG examinations. The Obstetrician &
Gynaecologist 2007;9:5357. doi:10.1576/toag.9.1.053.27297
3 Royal College of Obstetricians and Gynaecologists. Specialist Training
Curriculum [www.rcog.org.uk/index.asp?PageID=1959].
4 Royal College of Obstetricians and Gynaecologists. Part One MRCOG
and Core Curriculum Matrix [www.rcog.org.uk/index.asp?pageID=1912].
5 Duthie S, Hodges P, Ramsay I, Reid W. EMQs: a new component of the
MRCOG Part 2 exam. The Obstetrician & Gynaecologist 2006;8:1815.
doi:10.1576/toag.8.3.181.27255
6 Schuwirth L, van derVleuten C. Merging views on assessment. Med
Educ 2004;38:120810.
7 Royal College of Obstetricians and Gynaecologists. A New Format for the
Part 2 MRCOG Written Examination
[www.rcog.org.uk/index.asp?PageID=1338].
8 Beullens J, Van Damme B, Jaspaert H, Janssen PJ. Are extendedmatching multiple-choice items appropriate for a final test in medical
education? Med Teacher 2002;24:3905.
doi:10.1080/0142159021000000843
9 Bhakta B, Tennant A, Horton M, Lawton G, Andrich D. Using item
response theory to explore the psychometric properties of extended
matching questions examination in undergraduate medical education.
BMC Med Educ 2005;5:9. doi:10.1186/1472-6920-59
10 Case SM, Swanson DB. Extended-matching items: a practical alternative
to free-response questions. Teach Learn Med 1993;5:10715.
11 Royal College of Obstetricians and Gynaecologists. A New Format for the
Part 1 MRCOG Examination [www.rcog.org.uk/index.asp?PageID=1862].

2007 Royal College of Obstetricians and Gynaecologists