26.1 Statistical basis of Trials

Question 1

A recent meta-analysis of studies of the utility of the Mallampati score in the prediction of a difficult
airway found that it had a sensitivity of 60% and a specificity of 70%.

a) Outline what is meant by meta-analysis and the factors that ensure a high-quality conclusion from
the process. (10 marks)

Answer 1

Meta-analysis:
a quantitative* systematic review of
data from all available
primary studies that are similar in nature,
in order to reach a valid statistical
conclusion to a question.

Factors required for a high-quality conclusion:

> > Clearly defined question as the basis of the analysis.

> > Clear and reproducible methodology.

> > Comprehensive search of all available electronic databases based on appropriate search terms,
ensuring non-English studies are included.

> > Clear and valid criteria for inclusion or
exclusion of studies from the analysis,
ensuring only studies of sufficient quality included.

> > Consider publication bias to avoid
risk of over-representation of studies
with a positive outcome.

Question 2

b) Explain what is meant by sensitivity and specificity as applied to the interpretation of the Mallampati
data given above. (6 marks)

Answer 2

The sensitivity of a clinical test refers to the ability of the test to
correctly identify those patients with the disease or issue in question

The meta-analysis states that the
Mallampati test has a sensitivity of 60%,

which means that 60% of the people who have a difficult airway will be predicted as
difficult by the Mallampati test.

Sensitivity

=

True positives
__________________________
True positives + False negatives

The specificity of a clinical test refers
to the ability of the test to correctly
identify those patients without
the disease or issue in question.

The meta analysis states that the
Mallampati test has a specificity of 70%,

which means that 70% of the patients assessed
as having a straightforward airway
will indeed have a straightforward airway.

Specificity

=

True negatives
_____________________________
True negatives + False positives

Question 3

c) Rank the levels of scientific proof used to grade medical evidence. (4 marks)

Answer 3

1. Meta-analyses,
   randomised controlled trials (RCTs),
   or systematic reviews of RCTs.
2. Systematic reviews of,
   or individual,
   analytical non-RCTs such as
   case-control or cohort studies.
3. Non-analytic studies
   (for example, case reports,
   case series studies).
4. Expert opinion, formal consensus.

A grading code is then added after assessing the degree to which potential
sources of bias have been eliminated.

Question 4

Systematic review

Answer 4

*A systematic review is a qualitative review of all of the data of available
similar studies.