Biostatistics

Dr. Ahmed Seif El-Din

Associate Professor of Pharmacology & Toxicology
Head of Pharmacology & Toxicology
Faculty of Pharmacy and Drug Technology, Egyptian Chinese
University
1
Outline:

• Definition of Biostatistics.
• Why Pharmacists Need to Know Statistics?
• Population & samples
• Types of Data / Variables.
• Types of biostatistics
• Types of data distribution

2
 Statistics
Definition
•it is a method for Collecting, Classifying, Summarizing, and
Analyzing Data that are subject to random variation

•Biostatistics is the branch of applied statistics directed toward

applications in the health sciences and biology

3
 CERTAIN & RANDOM events
CERTAIN events RANDOM events

• The certain event will ALWAYS occur every time

the necessary set of conditions is realized.
• The occurrence of the random event will
remain a PROBABILITY because the
known set of conditions does not reflect
the entire collection of reasons
necessary and sufficient for its
occurrence.
 CERTAIN & RANDOM events
CERTAIN events RANDOM events

• The certain event will ALWAYS occur every time the necessary set of conditions is realized. The
occurrence of the random event will remain a PROBABILITY because the known set of conditions does
not reflect the entire collection of reasons necessary and sufficient for its occurrence.
• This is mainly due to our variability and the reasons behind it are not
always known nor fully understood.

• In consequence, we have to understand and to benefit from the laws of

probability because the best we can do is to keep predicting the most
suitable ways to treat our patients.

Thoth Research Upgrading Support & Training (TRUST)

 RANDOM events

Variability is not a Chaos and has characteristic distributions

that we can understand, analyze and most importantly, that
we can predict.
8
9
10
 Steps of Biostatistics

1- Experimental design
2- Collection of data 3-Classification and
may be done by presentation of data:
questionnaires or present data in a suitable 4- Analysis of 5- Interpretation:
experiment. form (table, chart, graph) the data Using the obtained
& obtain as much information to reach a
information as possible conclusion or answer a
(mean, SD, SE) question.

11
 1- Experimental
design
Each measured biological parameter is liable to
variation when it’s measured again
A. Species variation → biological variation due to
species difference.

B. Individual variation → biological variation from

animal to animal in the same animal

C. Diurnal variation → biological variation from

day-time to night-time

D. Seasonal variation → biological variation from

season to another season.

12
 Methods to reduce biological
variability
1. Selection of Participants (Generalization)
They should be of the same species, age, weight, and sex
2. Replication (6/30 participant/group)
The effect of drug on a give parameter should be repeated several times and take the
average to reduce the error ( sample size → Variability).
3. Use of controls and reference standards to reduce the biological variability due to
changes in procedure, time or conditions.

-In screening, compare drug to control

A. Concurrent control: doing the experiment to both the control (receive the solvent)
and the new tested drug at the same time.
B. Placebo: A Pharmacologically inert compound identical to the active drug in
appearance and taste. It’s used incase of CNS drugs.
N.B. In bioequivalence, compare drug to reference standard.

13
 Methods to reduce biological
variability
3. Use of controls and reference standards (Continue)
C. Sham operation:
Manipulation similar to a specific operation without involving the organ in question.
E.g. in an experiment we study the effect of ovariectomy on depression, to remove
the ovaries (ovariectomy), the control is anesthesized, opened, but the ovaries are not
removed.
D. Litter mate control:
it means that each dose is given to a group composed of one animal from each litter
rather than to a group made of one single litter.
E.g. 3 mother rats gave birth to 3 small rats, and we want to test the three drugs;
Diamiacron, Amaryl, Galvus

Diamiacron
Amaryl
Galvus
14
 Methods to reduce biological
variability
4. Randomization:
it means the choice of samples without bias according to a certain random
digit table. The concept of "randomness" implies that every item being
considered has an equal chance of being selected as part of the study.
5. Bias Counteraction: Bias can be avoided by
A. Blind design → the patient doesn’t know the drug,
but the physician knows.
B. Double blind design → both the patient and the physician
don’t know the drug.
C. Cross over design: to be sure that the response obtained due to the real
effect of the drug not due to the animal variability, we apply cross over
design.
Conditions to apply this design: the studied dis must be chronic, and
the treatment must not be able to permanently alter the disease.

15
 3- Classification & Presentation of data
1- Tabular presentation
A-Raw Data Table of survey of systolic BP for 100 normal
individual

When you mind the table, first its tedious to obtain

1- Min. value, max. value & the most repeated value thus chaos to obtain informative
data to describe it.
2- One thing you can do is to obtain mean value of these data (sum of values/no of
values)= 121.01 mmHg

16
 3- Classification & Presentation of data
1- Tabular presentation
B- Array Data Table of survey of systolic BP for 100 normal
Order the values in ascending order individual

*Advantage:
When you mind this table, it becomes neater and you can conclude the min.
value=107, max. value= 147 and the most repeated value (mode) = 125.
*Disadvantage:
The data are still very huge (i.e. 100 values) → Compress data and transform it to
frequency distribution table
17
 3- Classification & Presentation of data
1- Tabular presentation
C-Frequency Distribution Table of survey of systolic BP of 100
normal individual
It is a table showing the relation between a class interval and the frequency of occurrence
Range = largest observation – smallest observation
then Class interval = Range/No of categories
E.g. From an array table, the largest observation = 147, the smallest = 107 and we will divide
them into 8 categories. Class Frequency
Calculation: 105-109.9 2
-Range = 147-107 = 40 110-114.9 4
115-119.9 7
-Class interval = 40/8 = 5 units
120-124.9 17
therefore each class is formed of 5 125-129.9 40
units 130-134.9 15
135-139.9 8
140-144.9 4
1st class 105-109.9 145-149.9 3
2nd class 110-114.9 100
 3- Classification & Presentation of data
1- Tabular presentation
C- Frequency Distribution Table of survey of systolic BP of
100 normal individual
Class Frequency
105-109.9 2
110-114.9 4
115-119.9 7
120-124.9 17
125-129.9 40
130-134.9 15
135-139.9 8
140-144.9 4
145-149.9 3
100

Advantage: When you mind this table, its become more compressed and you can
conclude not the min. value only , but the values of BP concentrate around the class
125-129.9. Also, you can express the frequency by percentage (relative frequency %)
and give more clear view of the data.
19
 3- Classification & Presentation of data
1- Tabular presentation
D- Relative Frequency Distribution Table of survey of
systolic BP of 100 normal individual
In which the observations in the frequency table are expressed as % of the total
number.
E.g. if the total number of observations (N) = 100 & the frequency in a given class = 3,
then the relative frequency of this class = 3/100 X 100 = 3%
Class Frequency Relative frequency%
105-109.9 2 2%
110-114.9 4 4%
115-119.9 7 7%
120-124.9 17 17%
125-129.9 40 40%
130-134.9 15 15%
135-139.9 8 8%
140-144.9 4 4%
145-149.9 3 3%
100 100%

20
 3- Classification & Presentation of data
1- Tabular presentation
E- Cumulative Frequency Distribution Table of survey of
systolic BP of 100 normal individual
Blood Pressure Cumulative Frequency
< 105 0
<110 2
<115 6
<120 13
<125 30
<130 70
<135 85
<140 93
<145 97
<150 100

It shows the number of observations that are less than or more than a certain value
21
 3- Classification & Presentation of data
2- Graphical presentation
A- Bar Chart:
-Drawing rectangles whose heights are proportional to the
value of the measured phenomenon, while its width is
chosen to be suitable with the length of bar and number of
bars.
1- Simple bar chart for measurement of a single
phenomenon.
2-Multiple bar chart for more than one phenomenon e.g.
each cluster consists of a group of effects of the same drug
with different concentrations.

22
 3- Classification & Presentation of data
2- Graphical presentation

A- Bar Chart: The Stacked bar chart

23
 3- Classification & Presentation of data
2- Graphical presentation
B- Histogram & frequency curve
Is continuous bar charts corresponding to the data present in
frequency distribution table.
- A rectangle whose base equals to the class interval and whose
height is determined by the no of frequencies.

One limitation of the histogram is that it can represent only one variable at a time (like the pie
chart), and this can make comparisons between two histograms difficult, because, if you try to
plot more than one histogram on the same axes, invariably parts of one chart will overlap the
other.
24
 3- Classification & Presentation of data
2- Graphical presentation
C- The frequency polygon curve

This is done by plotting the center point of the classes on the x-axis against the
frequency values on the y-axis and hointing between these points using a ruler

25
 3- Classification & Presentation of data
2- Graphical presentation
D- The frequency distribution curve

Bell shaped graph

To draw bell shaped graph,
connect the midpoint of
each column by hand

26
27