1. An introduction to some commonly used
terms of significance for all clinicians
EPIDEMIOLOGICAL
STATISTICS
MODERATOR: Prof Kakkar and Prof. R M Kaushik
PRESENTER- Dr.Garima Aggarwal
2. Epidemiology – It is the study of the
rate/occurence of disease in a population
Incidence – The number of new cases occuring in
a defined population during a specified period of
time.
Prevalence – Refers to all current cases (OLD
and NEW) of a disease/ condition at a given point
/over a period of time in a given population
P = Incidence X Duration
3. Sensitivity – Ability of a test to identify
correctly all those who have the disease , that
is “TRUE POSITIVE”
ELISA for HIV is 99.5% sensitive
Specificity – It is defined as the ability of a
test to identify correctly those who do not
have the disease, that is “TRUE NEGATIVE”
ELISA for HIV is 98.5% specific
False negative-Patients who have the disease
are told that they do not have the disease.
False positive- Patients who do not have the
disease are told they have it.
4. Statistical Averages
MEAN – individual observations are added
together and then divided by the number of
observations
MEDIAN – data is first arranged in an
ascending or descending order of magnitude
and then the value of the middle observation is
located
MODE- most frequently occurring observation
in a series of observations
STANDARD DEVIATION-
5.
6. EPIDEMIOLOGICALSTUDY
OBSERVATIONAL EXPERIMENTAL
DESCRIPTIVE ANALYTICAL RCTs FIELD COMMUNITY
TRIALS TRIALS
7. Observational
studies-
CASE REPORT – clinical characteristic or outcome from a
single clinical subject
CROSS SECTIONAL STUDY – study based on a single
examination of a cross section of population at ONE
POINT IN TIME , where cross section is such that the
results can be projected on the entire study population
CASE CONTROL STUDY – study of a group of people
with the disease and compares them with a suitable
comparison group without the disease , i.e. CASES and
CONTROLS. Retrospective study.
8. COHORT STUDY – population group of those who
have been exposed to risk factor is identified and
followed over time and compared with a group not
exposed to the risk factor. Prospective study.
CASE CONTROL COHORT
CROSS SECTIONAL
9. Experimental
studies -
RANDOMISED CONTROLLED TRIALS – subjects in
the study are randomly allocated into “intervention”
and “control” groups to receive or not to receive an
experimental preventive or therapeutic procedure or
intervention.Most scientifically rigorous studies.
Select RAND Experiment
Select al V/S
population
suitable O- control Manipulation Blinding ASSESSMENT
sample MISE group
10. Statistical Analysis -
For observational studies
Relative Risk – Ratio of the incidence of the disease (or death) among exposed
group and the incidence among non exposed
Relative Risk = 1 = no association, >1 = positive association
Direct measure of the ‘strength’ of association between suspected cause and
effect
IMR in whites in the US is 8.9 per 1000 live births, and 18.0 in blacks. So the
Relative risk of Black v/s White population is 18/8.9 = 2.02. Therefore Black infants
are twice as likely to die in the first year of life.
Attributable Risk – It is the difference in incidence rates of disease (or death)
between an exposed group and non exposed group.
ATTRIBUTABLE RISK = (incidence of disease among exposed – incidence of
disease among non exposed) / incidence of disease among exposed x 100
Using above example, AR= 18.0-8.9 = 9.1, hence Of every 1000 black infants
there were 9.1 more deaths than were obsereved in 1000 white infants
11. Exposure to Risk factor CASES CONTROL
(Disease Present) (Disease Absent)
PRESENT a b
ABSENT c d
a +c b+ d
ODDS RATIO – looks at the increased odds of getting a disease
with exposure to a risk factor as opposed to getting the disease
without exposure.
OODS RATIO = a x d / b x c
SMOKING LUNG CANCER Without LUNG CA.
Smokers 33 55
Non Smokers 2 27
total
ODDS RATIO = 33 X 27 / 2 X 55 = 8.1
Smokers showed a risk of having Lung Cancer 8.1 times that of Non smokers.
12. Inferential statistics
CONFIDENCE INTERVAL – Confidence intervals are a way of
admitting that any measurement from a sample is only an
estimate of the population
A confidence interval specifies how far above or below a
sample based value , the population value lies within a given
range , from a possible high to a possible low.
We have 95% confidence intervals and 99% confidence
intervals.
If the confidence interval contains 1.0 it is not statistically
significant
13. What is the ‘p value’???
With scientific methods – we put forward a
research question eg. Smokers more likely to get
lung cancer!
Null hypothesis – says that all findings are a result
of chance or random factors i.e. smoking has no
real relation with lung cancer
Hypothesis testing – ‘p value’ – helps to interpret
output from a statistical test. It is the standard
against which we compare our results.
If p value < or = 0.05 - the results are statistically
significant, i.e. REJECT NULL HYPOTHESIS
If p value > 0.05 – statistically insignificant, i.e. DO
NOT REJECT NULL HYPOTHESIS
14. Statistical tests -
META-ANALYSIS- A statistical way of combining
results of many studies to produce one overall
conclusion.
Correlation coefficient – It indicates the degree to
which two measures are related
It ranges from -1.o to +1.0
Medical school grades and various factors affecting it.
Positive value – two variables go together in the same
direction. IQ has a positive corelation with medical
grades.
Negative value – presence of one variable is associated
with absence of another. Time spent on outdoor
activities negative correlation with grades.
15. t tests – used to compare MEANS of two
groups. Can be used for testing two groups only.
Paired t test – when comparing ‘before’ and
‘after’ results in the same group.
Unpaired t test – when comparing means of two
groups.
Chi square – can be used for any number of
groups.
Used for nominal data.