MRCS PART A

MRCS In Capsules

Dr. ABO OMAR

MRCS SCHOLARS ABO OMAR

Contents
Consents and Capacity
Audit
Types of Data
Study designs
Measures of Central Tendency
Statistical Hypothesis
Probability = P value
Normal Distribution Curve
Skewed Distribution Curve
STANDARD DEVIATION
Statistical error
Incidence and prevalence
Risk and Relative Risk
Absolute Risk Reduction
NNT
Odds and Odds Ratio
Sensitivity and Specificity
Predictive Values
Significance tests
Levels of evidence

MRCS SCHOLARS ABO OMAR

 Consent
Autonomy
- represents a patient's right to determine his / her own health-care decisions even if
the proposed therapy is "for his/her own good."
(doctor is not a god)

Consent
In general : occurs when one person voluntarily agrees to the proposal or desires of
another

In medical field : occurs when patient voluntarily agrees to the health service offered
by the doctor ( examination , investigations , treatment , procedures or operation)

Types of consent

A- Implied Consent
is not expressed by a person, but his permission and agreement is concluded from his
actions
Example : patient select one doctor from others to consult with him its an implied
consent that he agree for physical examination

B- Informed ( Expressed ) Consent

- Informed consent is a person’s voluntary decision about medical care that is made
with knowledge and understanding of the benefits and risks involved

- Battery is the legal term for failing to obtain informed consent before performing
a test or procedure on a patient (its a form of assault).

MRCS SCHOLARS ABO OMAR

Components of Informed Consent
1 - You must have the capacity (or ability) to make the decision.
2 - The medical provider must disclose information on the treatment, test, or
procedure in question, including the expected benefits and risks, and the likelihood
(or probability) that the benefits and risks will occur.
3 - You must comprehend the relevant information.
4 - You must voluntarily grant consent, without coercion or duress.

1- Decision-Making Capacity
To have decision-making capacity does not mean that you, as the patient, will always
make "good" decisions, or decisions that your doctor agrees with. Likewise, making a
"bad" decision does not mean that you, as a patient, are "incompetent" or do not
have decision-making capacity.

Capacity Key points include:

1. Understand and retain information
2. Patient believes the information to be true
3. Patient is able to weigh the information to make a decision
4. All patients must be assumed to have capacity

2- All Options must be described

you must fully inform the patient of the risks and benefits of each procedure prior to
undergoing the procedure
N.B.In addition to understanding the risks of the procedure, you must inform the
patient of what could happen if he does not choose therapy that you offer.
3- Consent is rquired for each specific procedure
4- Decision made when competent is valid when capacity is lost
5- Telephone consent is valid
5- The Person who will perform the procedure should obtain consent
6- Pregnant women can refuse therapy

MRCS SCHOLARS ABO OMAR

 Forms of Informed consent

Consent 1 = Consent the patient

competent adult 18 years or more = Who is able to consent for himself = patient
himself agree for investigations or treatment where consciousness is impaired
eg . 1 – patient with severe depression =
consent the patient and respect his refusal
2 – Down syndrome but understand all aspects of the operation or procedure =
consent the patient( down syndrome is not mentally retarted )

Consent 2 incompetent child

Adult( Parents ) consent on behalf of child = parents ( or legal guardians )agree for
investigations or treatment for a child or young person where consciousness is
impaired

Consent 3 procedure on a child or adult who are conscious ( during the procedure no
loss of consciousness = no general anasthesia = local only or no anasthesia ) = patient
himself or parentsagree for investigations or treatment

Consent 4 incompetent adult wholack capacityto provide informed consentLike

dementia = unable to consent for investigations or treatment = filled by healthcare
professional- two consultants

Gallick competent = 2 – 5 years below 18 years but patient understand the procedure
Can consent by agreement , but don’t accept his refusal
Has the right to agree but does not have the right to refuse

Consent by proxyfor adult when you cant wait to regain his capacity

Apply for courtwhen parents refuse a procedure for child which is necessary

MRCS SCHOLARS ABO OMAR

Bolam test considers whether a doctor'sdecision matches the opnion of aresponsible
body of doctors skilled in the same practice.

Proceed without consent( considered as implied consent) = For all life saving in which
medical care is needed immediately to prevent serious or irreversible harm :
- ruptured AAA
- ruptured spleen or liver
- EDH, unstable patient
- massive intraabdominal hemorrhage
- massive intrathoracic hemorrhage.
N.B. :
- Not all emergencies are life saving (eg. Some fractures and wounds)
- Proceed without consent not applied to a patient with a pre existing DNR
(Do Not Resussitate) order

Audit and Research

Clinical Audit Versus Research

Research Aims to derive new knowledge which is potentially generalisable or
transferable and asks what is the right practice
clinical audit asks are we doing the right practice

Clinical Audit includes : structure, processes, and outcomes of care

Structure audit monitors the structure or setting in which patient care occurs ,
such as finance , nursing services, medical records and environment. This audit
assumes that a relationship exists between quality care and appropriate
structure. These above audits can occur retrospectively, concurrently and
prospectively.
e.g avalibility of a smoking cessation clinic in a locality.

MRCS SCHOLARS ABO OMAR

Process audits are used to measure the process of care or how the care was
carried out.process audit is task oriented and focus on whether or not practice
standards are being fulfilled . these audits assumed that a relationship exists
between the quality of the nurse and the quality of care provided
e.g waiting times for an appointment at a smoking cessation clinic Investigations
or imaging requsted

Outcome audit
outcomes are the end results of care ; the changes in the patients health status
and can be attributed to delivery of health care services. Outcome audits
determine what results if any occured as result of specific nursing intervention
for clients. e,g number of smokers who quite smoking

Types of audit
Financial audit A historically oriented, independent evaluation performed for the
purpose of attesting to the fairness, accuracy, and reliability of financial data

Operational audit A future-oriented, systematic, and independent evaluation of

organizational activities. Financial data may be used, but the primary sources of
evidence are the operational policies and achievements related to organizational
objectives. Internal controls and efficiencies may be evaluated during this type of
review.

Departmental review A current period analysis of administrative functions, to

evaluate the adequacy of controls, safeguarding of assets, efficient use of
resources, compliance with related laws, regulations and institutional policy and
integrity of financial information
Standards based audit involves defining widely agreed standards, collecting data
to measure current practice against those standards, and implementing any
changes deemed necessary

MRCS SCHOLARS ABO OMAR

Systems based audit : Evaluation of processes occurring within an institution.
Systems based audits are an integral part of the process of clinical governance.
Peer review
An assessment (evaluation) of creative work or performance by a clinical team in
the same field in order to maintain or enhance the quality of the work or
performance in that field with a view to improving clinical care.
Surgical audit - Data collection of all surgical cases, followed by ongoing review
and assessment of performance and outcomes. Related to peer review, but is
distinguished by aiming for inclusion of all cases carried out, rather than sampling
alone.

Types of Data
Quantitative data Qualitative data
deals with numbers and things you can measure deals with characteristics and descriptors
dimensions such as height, width, and length. that can't be easily measured, but can be
Temperature and humidity. Prices. Area and observed - such as smells, tastes, textures,
volume. attractiveness, and color
Types Types
Continuous = measurements Binary data place things in one of two
could be divided and reduced to finer and finer exclusive categories :
levels. right/wrong, true/false, or accept/reject.
For example, you can measure the height of
your kids at progressively more precise scales -
meters, centimeters, millimeters, and beyond -
so height is continuous data
Discrete = counts unordered or nominal data, assign items
is a count that can't be made more precise. to named categories that do not have order
For instance, number of children (or adults) in or natural value or rank.
your family is discrete data record the nationality of students in a class
As you can't have 2.5 kids, or 1.3 adults or coulers of clothes in a bag
Ordered or Ordinal data, assign items to
categories that have natural order, such as
"Short, Medium, or Tall."

MRCS SCHOLARS ABO OMAR

 Retrospective study Prospective study
data are collected from already existing there are no patients to begin with. We start
previous data assigning the patients to the study group and
collect the data as and when available

Study designs
Experimental studies Observational studies
intervention of the researcher subjects are observed
observation of what happens no action from the researcher
Types Types
Randomized Non-Randomized Cohort studies Case-Control studies
controlled trails controlled trails
(RCTs)

Observational study : we study and record various factors by observing the

population or sample. We do not intervene in the process.
case-control study : observational and retrospective = What happened
we study two existing groups with different outcome and compare the results,
usually for causal relationship. Measure odds ratio

cross-sectional studies : Observational = What is happening

we study a particular population at a given specific point of time.
Collects data from a group to assess frequency of a disease at a specific time
Measure prevelance , reference range , current health status
Can show risk factors but not causes
Descriptive studies , Most appropriate method to assess diagnostic tests

MRCS SCHOLARS ABO OMAR

longitudinal study , we study a population over a period of time for the same
variable factors/parameters repeatedly. One of the uses is to know the natural
history of the disease.
Cohort Study Observational & prospective = What will happen
is a type of longitudinal study where a group of patients is closely monitored
over a period of time.
Compares a group with a given risk factor to a group without to assess whether
risk factor increase likelihood of adisease.
Measure relative risk used for prognosis and aetiology
Most appropriate study for assessing prognosis
e.g smokers had a higher risk of developing COPD than nonsmokers.
Experimental studies = Randomised Control Trial = the best study
the gold standard When assessing efficacy of treatment or intervention.
Participants randomly allocated to intervention ( eg.ttt) or control (placebo)
Ideally the trials should be blinded, usually to the patient and those treating them
Cluster Randomised Controlled Trials
Groups are randomised rather than individuals
Avoids cross contamination amongst participants
Participants in any one cluster are more likely to respond in a similar fashion
Higher risk of unit of analysis error as these studies should be analysed as
clusters rather than on an individual basis. This leads to a higher false positive
rate.
assessing prognosis = Cohort
assessing diagnostic = Cross Sectional studies
assessing treatment = Randomised Control Trial

MRCS SCHOLARS ABO OMAR

 Measures of Central Tendency
Mean : average
the sum of all values, divided by their number

Median : middle value , the point which has half the values above, and half below.
= is the central observation when the observations are arranged in ascending or
descending order.
The observations should be either in ascending or descending order, if not already
arranged
Mode : is simply the most frequently occurring observation. ( 1,2,2 2,3,5)
If there are two numbers which occur in the same frequency, then there are two
modes.

Statistical Hypothesis
Null hypothesis : is a general statement or default position that there is no
relationship between two measured phenomena, or no association among groups
results purely
No difference between the two groups being compared or no effect of the
intervention
from chance , there are no differences
denoted by H0 - sample observations

Alternative hypothesis : an assumption of a population observations - and it can

be true or not
denoted by H1 - sample observations
influenced by non-random cause
Example: Suppose that we flip a coin 50 times :
H0 - half the flips would result in Heads and half in Tails = 0,50
H1 - the number of Heads and Tails would be different ≠0,50

MRCS SCHOLARS ABO OMAR

 P value = probability
P value is the probability of the null hypothesis being true = any observed
differences happened by chance and no any effect or difference
P value of less than 0.05 = statistically significant = only 5 % chance that the null
hypothesis is true
P value of less than 0.01 = highly statistically significant
P value of less than 0.001 = very highly statistically significant
what is this natural variation ?
For example, when you give the same drug to a group of patients, in the same
dose to treat the same condition, you will find that some patients improve
completely, some partially, and some not at all!
Also, there are many conditions which improve spontaneously. If you give a drug
during this period of spontaneous improvement (or natural healing), it is possible
that the improvement in the disease will be wrongly attributed to the efficacy of
the drug.
Placebo effect is an improvement or a result not attributable to the effects of the
drug given. Results may be because of psychological factors.

Normal Distribution Curve

The data which give a normal distribution curve called parametric data
also known as the Gaussian distribution or 'bell-shaped' distribution
Properties of normal distribution curve :
The central line is the mean value
symmetrical i.e. Mean = mode = median
The observations fall in the center around the mean symmetrically
Elliptical-shaped curve
68.3% of values lie within 1 SD of the mean
95.4% of values lie within 2 SD of the mean = If we take two standard deviations
(SDs) on either side of the mean, it covers 95 % of the area under the curve =
95 % of the observations fall under this area = 95 % confidence range
99.7% of values lie within 3 SD of the mean

MRCS SCHOLARS ABO OMAR

MRCS SCHOLARS ABO OMAR
 Skewed Distribution Curve
The majority of observations fall on to one side of the mean.
Time Bl. glucose
The mean, median, and mode do not coincide. (min) levels
The curve is not bilaterally symmetrical ( asymmetrical) 0 70
The data which give skewed distribution curve are called 15 110
nonparametric 30 160
Example : 45 180
If blood glucose levels are recorded after a meal 60 170
in a particular patient 75 168
90 160
If mean, median, and mode calculated on this data, we can
105 155
see that they do not coincide but differ significantly. 120 135
135 120
150 115
STANDARD DEVIATION
It expresses How much the values spread around (deviates from) the mean , only
used for normally distributed data , the larger the SD the more widely spread the data
A range of one SD above and below the mean
1 SD includes 68.2% of the values.
2 SD includes 95.4% of the data.
3 SD includes 99.7%.
Example :
group of patients had a normal distribution for weight. The mean weight of the
patients was 80 kg. For this group, the SD was calculated to be 5 kg.
1 SD below the average is 80 – 5 = 75 kg.
1 SD above the average is 80 + 5 = 85 kg.
So ,
68.2% of patients will weigh between 75 and 85 kg ( - 1 SD to + 1 SD )
95.4% will weigh between 70 and 90 kg ( - 2 SD to + 2 SD )
99.7% of patients will weigh between 65 and 95 kg ( - 3 SD to + 3 SD )

MRCS SCHOLARS ABO OMAR

 Statistical Errors
Type 1 Error ( Alpha α ) Type 2 Error ( Beta β )
no effect but the test shows the effect some effect but the test fail to detect it
test rejects a true null hypothesis test fails to reject a false null hypothesis
Incorrect rejection of a null hypothesis Incorrect acceptance of a null hypothesis
( which is false ) ( which is false )
False positive False negative
equals the significance level of a test related to the power of a test
NB :
- The power of a test is the of the test to reject the null hypothesis when it is false
(thereby avoiding a type 2 error).
- Increasing the power of a test will reduce the probability of a type 2 error. Usually a
value of 0.8 is selected.

Example
FNAC is done on a group of patients to detect thyroid carcinoma
FNAC is reported as malignancy in 100 patients and
nonmalignant in another 100 patients.
When specimen is subjected to histopathology Examination after surgical excision,
the following results are obtained. So we have four possibilities.
Pre-operative FNAC correlation with post-operative histopathology (HPE) report
FNAC: malignancy FNAC: nonmalignant
HPE : malignant 95 7
HPE : nonmalignant 5 93
N = 100 N = 100
Considering HPE is the final and confirmatory in detecting malignancy
Sensitivity = 95 / 95 + 5 = 95 / 100 = 95 %
Specificity = 93 / 93 + 7 = 93 / 100 = 93 %
PPR = 95/95 + 7 = 95 / 102 = 0.93
NPR = 93/93+ 5 = 93 / 98 = 0.94

MRCS SCHOLARS ABO OMAR

True positive or sensitivity rate of the test or positive predictive value :
FNAC has detected malignancy in 95 % of cases.
True negative or specificity rate of the test or negative predictive value :
FNAC has correctly ruled out malignancy in 93 % of cases..
Sensitivity rate of a test is its ability to pick up the correct diagnosis.
Specificity rate of a test is its ability to rule out the diagnosis correctly.
False positive cases or type 1 error :
FNAC has wrongly detected malignancy in 5 % cases. These cases are reported as
malignancy in FNAC but these cases are actually nonmalignant.
False negativity cases or type 2 error :
FNAC has wrongly ruled out malignancy in 7 % cases. These cases were actually
malignant but FNAC has missed the diagnosis.
A false-negative test gives false assurance to the treating doctor and patient and
hence results in no or inadequate treatment.
A false-positive test produces tension , panic and may result in overtreatment.

MRCS SCHOLARS ABO OMAR

 Incidence and prevalence

Incidence
It is the number of new cases of a disease arising in a population in a given time
period (Per 100 population in 1 year).
Prevalence
It is the total number of all cases ( old and new) having the disease per 100
population at a given point of time.
Prevalence = (incidence) × (duration of condition)
In chronic diseases the prevalence is much greater than the incidence
In acute diseases the prevalence and incidence are similar.
In conditions such as the common cold incidence may be greater than prevalence

Risk and Relative Risk Ratio

Risk is the probability that an event will happen. It is calculated by dividing the
number of events by the number of people at risk.

Relative risk calculated when two groups are compared

It is the ratio of risk in the experimental group (treated or exposed or study group)
(experimental event rate, EER)
to risk in the control group (unexposed group ) (control event rate, CER)
= Incidence among Exposed/ Incidence among Non-Exposed
EER = rate at which events occur in the experimental group
CER = rate at which events occur in the control group

Absolute risk reduction (ARR ) Absolute risk increase (ARI)

is the difference between the event rate is the difference between the event rate
in the experimental group and that in the in the control group and that in the
control group experimental group
ARR = CER - EER ARI = EER - CER
Relative risk reduction (RRR) Relative risk increase (RRI)

MRCS SCHOLARS ABO OMAR

absolute risk reduction / control event absolute risk increase / control event rate
rate = CER – EER / CER = EER – CER / CER

Absolute Risk Reduction The absolute risk reduction is the decrease in risk of a given activity
or treatment in relation to a control activity or treatment. It is the inverse of the number
needed to treat.

relative risk = 1 no difference in risk between the groups

both the groups have the same risk
relative risk > 1 the rate of an event is increased compared to controls
calculate the relative risk increase
relative risk < 1 the rate of an event is decreased compared to controls
calculate the relative risk reduction

Example 1: if we look at a trial comparing the use of paracetamol for back pain
compared to placebo we may get the following results
Total number of patients Achieved 50% pain relief
Paracetamol 100 60
Placebo 80 20
Experimental event rate, EER = 60 / 100 = 0.6
Control event rate, CER = 20 / 80 = 0.25
Relative risk = EER / CER = 0.6 / 0.25 = 2.4
RRI = (EER - CER) / CER = (0.6 - 0.25) / 0.25 = 1.4 = 140 %

Number needed to treat : how many patients would be need to receive a treatment
to prevent one event ( to cure one patient ). This is useful in determining the cost Vs
benefit of many treatments
The inverse / reciprocal of absolute risk reduction is the Number Needed
to Treat

MRCS SCHOLARS ABO OMAR

Example : One hundred women with vaginal candida were given an oral antifungal,
100 were given placebo. They were reviewed 3 days later. The results are
Given antifungal Given placebo
Improved Not improved Improved Not improved
80 20 60 40
ARR = improvement rate in the intervention group – improvement rate in
the control group
ARR = 80 % – 60 % = 20 % = 20 / 100 = 0.20
NNT = 1 / ARR = 1 / 0.20 = 100 / 20 = 5
So five women have to be treated for one to get benefit.
The incidence of candidiasis was reduced from 40% with placebo to 20% with
treatment (by half) , Thus, the RRR is 50%
Example : young men were treated with an expensive lipid-lowering agent. Five years
later the death rate from ischaemic heart disease (IHD) is recorded.
Given drug Given placebo
survived died survived died
998 = 99.8 % 2 = 0.2 % 996 = 99.6 % 4 = 0.4 %
ARR = improvement rate in the intervention group - improvement rate in
the control group = 99.8 % - 99.6 % = 0.2 %
NNT = 100 / ARR = 100 / 0.2 = 500
So 500 men have to be treated for 5 years for one to survive
The incidence of death from IHD is reduced from 0.4 % with placebo to 0.2 % with
treatment, Thus, the RRR is 50%

NB. For treatments, the lower the NNT is better than the higher
- NNT of 10 for treating a sore throat with expensive blundamycin is not attractive
- NNT of 10 for prevention of death from leukaemia with a non-toxic chemotherapy
agent is worthwhile
Expect NNTs for prophylaxis to be much larger
For example, an immunization may have an NNT in the thousands but still be well
worthwhile.

MRCS SCHOLARS ABO OMAR

 Odds and Odds Ratio
Odd is the ratio of an event happening to not happening
it is a way of comparing patients who already have a certain condition (cases) with
patients who do not (controls) (case–control study )
calculated by dividing the number of times an event happens by the number of
times it does not happen
Example : One boy is born for every two births
the odds of giving birth to a boy are 1:1 (or 50:50) = 1⁄1 = 1

Odds Ratio is the ratio of odds of the study group to odds of the control group.
Example : a new drug (Drug X) is being studied for side effects.
study group had mortality of 6 % , the odds are 6 : 94
control group had a mortality of 1 % , the odds are 1:99
odds ratio is odds of mortality of the study group divided by odds of mortality of the
control group.
Odds ratio = odds of the study group/odds of the control group
= 6/94 divided by 1/99 = 6.316

MRCS SCHOLARS ABO OMAR

 Sensitivity and Specificity
Disease positive Disease negative
Test positive TP ( A ) FP ( B )
Test negative FN ( C ) TN ( D )
Sensitivity Specificity
the ability of a test to correctly identify the ability of a test to correctly exclude a
disease = detect the true positive. disease = detect the true negative ,
the proportion of true +ve cases identified by the proportion of true -ve cases among all -
a test among all +ve cases
the probability that an individual with the
ve cases
the probability that an individual without the
disease is screened positive
disease is screened negative
True Positives/Number of people with True Negatives/Number of people without
disease disease
TP / TP + FN TN / TN + FP
A/A+C D/B+D
Positive Predictive Value Negative Predictive Value
The probability of having the disease if the test The probability of not having the disease if the test
result is +ve result is -ve
The propability of the condition to be confirmed The propability of the condition to be excluded
from all positive results from all negative results
proportion of those who have a positive test proportion of those who test negative who do
who actually have the disease not have the disease
depends upon the prevalence of the condition depends upon the prevalence of the condition
being tested for and the sensitivity of the test being tested for and the specificity of the test used
used
low PPV means that the pt has low low NPV means that the pt has high probability of
post-test probability of having the disease having a disease
high PPV means that the pt has high high NPV means that the pt has low probability of
post-test probability of having the disease having a disease
True Positives/All positive test results True Negatives/All Negative test results
TP / TP + FP TN / TN + FP
A/A+B D/C+D

MRCS SCHOLARS ABO OMAR

 Raising the cutoff point of a Lowering the cutoff point of a
diagnostic test diagnostic test
increase it's specificity increase it's sensitivity
increase PPV increase NPV
decrease it's sensitivity decrease it's specificity
decrease false +ve results decrease false -ve results

MRCS SCHOLARS ABO OMAR

 Significance tests
Quantitative
A – One group
1 – Simple t - test : one sample t-test = Compare the sample with population
The data are continuous variables (meaning, it can take infinite number of values
One thing in one group and compare with the normal or with the general
e.g blood cholesterol in obese pt
2 – r-test correlation : Two different things in one group
( to see the relarion between them )
e.g relation between Hgb level and No of delivery
e.g relation between Ca level and No of pregnancy

3 – Paired t-test = dependant t-test : One same value in one group

measured twice (before and after some interference or procedure)
eg in diabetic patient measure blood glucose level before and after insulin injection

B – Two groups
1 – Unparied t-test = independent t-test = two sample t-test : Two different groups ,
measure mean (one value ) in each and compare the mean of the two groups
The means in the two groups must be statistically independent
eg.measure Ca level in nullipara and multipara and compare between them
eg.measureBlood glucose level in diabetics and nondiabeticsand compare between
them

MRCS SCHOLARS ABO OMAR

Qualitative
Yes or no - positive or negative - mild , moderate or severe

1 - Chi-SQUARE (X2 test)

Compare two or more variable values
1. All values should be actual numbers ( not percentages )
2. You cannot use percentages or averages.
3. The value in all the cells should be five or more (For smaller sample sizes,
Fisher’s test is applied.)
4. The sample should have been randomly drawn.
5. Variable should be categorical

Fisher’s Exact Test

When the frequency is small, we cannot use chi test. In such cases we must use
Fisher’s exact test or simply Fisher’s test. This test is used in similar situations
where chi test is used but the frequency is small. It is already mentioned that chi test
is invalid if the value in each cell is less than 5 for a 2 × 2 table. In such cases Fisher’s
test is useful.
2 - Mann whitny test
Non gaussian data when compairing two unpaired groups
3 - Wilcoxon test
Non gaussian data when compairing two paired groups
4 - Bonferroni
multiple analyses of sub group data using multiple tests of non parametric
for correcting against multiple statistical analyses that might provide an erroneous
result (i.e. to correct against data dredging)

MRCS SCHOLARS ABO OMAR

 Level of evidence

Level of evidence Type of Study

1a Systematic reviews of randomized clinical trails (RCTs)
1b Individual RCTs
2a Systematic reviews of cohort studies
2b Individual cohort studies and low-quality RCTs
3a Systematic reviews of case-controlled studies
3b Individual case-controlled studies
4 Case series , poor-quality cohort and case -controlled studies
5 Expert opinion based on clinical experience

MRCS SCHOLARS ABO OMAR

MRCS SCHOLARS ABO OMAR
 Numerical data Categorical data
Number of 2 groups 3 or more groups 2 or more 2 or more
groups variables variables
(sample size is
small
Type of Parametric Non-parametric Parametric Non-
data parametric
Type of paired unpaired paired unpaired
samples
The Test Paired Unpaired Wilcoxan Mann ANOVA Kruskal- Chi-square Fisher exact
T test T test Ranksum Whitney (analysis of Wallis test test
Test Test variance) test

Paired : normally distributed ---> paired t test

not normally distributed ---> Wilcoxon Signed-Rank Test

Unpaired = two groups : normally distributed ---> Unpaired t test

not normally distributed ---> Mann-Whitney-U test
Unpaired = > two groups : normally distributed ---> ANOVA
not normally distributed ---> Kruskal Wallis test

GOOD LUCK

ABO OMAR

MRCS SCHOLARS ABO OMAR