Introduction to

Biostatistics

Dr Noha Saleh
A. Professor of public health (biostatistics and epidemiology)
Consultant preventive medicine
Outlines:

• Statistics/Biostatistics/Medical statistics
• Role of biostatistics in the health care field
• Data types and data sources
Objectives:
1-Help participants to differentiate between statistics, biostatistics and
medical statistics
2-Introduce participants to the role of statistics in relation to human health
and public health
3-To create in them an awareness of the need to acquire an understanding
of statistical methods
4-Provide them with an understanding of the nature and types of data and
how the data are collected
 It derives from the Latin word status (manner of
standing/position).

In another sense it is a field of study concerned with :

• The collection, organization, summarization and

analysis of data, and The drawing of inferences
about a body of data when only a part of the data is
observed.
The tools of statistics are employed in many
fields as business, education, psychology,
agriculture, and economics….. When the data
being analyzed are derived from the biological
sciences and medicine we use the term biostatistics
 Why preventive medicine physicians
should learn biostatistics
 Evaluating the medical literature  to decide
whether they can believe the results presented in
the literature.
 Interpreting vital statistics (births, deaths..) how
they are measured? What they mean? How they are
used?  to diagnose and manage patients
effectively.
• Understanding epidemiologic problems 
prevalence of disease, variation by season or area
and influence of risk factors  to make
diagnoses and develop management plans.
• Interpreting information about drugs &
equipment  companies use graphs, charts &
results of studies comparing their products with
others in the market.
• Using diagnostic procedures  to know how
the test is sensitive to diagnose a disease
(Sensitivity) and also no disease (Specificity) in
a well person.
• Being informed  workers in the public health
field should be alert for statements that do not
seem quite right.
• Evaluating study protocols and articles  for
those who evaluate articles for publication
(to evaluate design & statistical methods).

 Participating in or directing research

projects
The General Sequence of Steps in a
Research Project planning  Design 
Execution (Data collection)  Data
processing  Data analysis Presentation
 Interpretation  Publication
 Sources of data

Primary Secondary
This is when data These are data collected and
collection is designed recorded for another
specifically for the study research study, and which are
available for use.
and the data are newly
collected.
Primary data
A- From records (Regular/Routine collection system):
 Hospital records.
Annual vital records  births, deaths.
Annual & monthly reports of WHO,
Ministry of Health.
B- Survey methods
1)Comprehensive survey (census) collection of data about
every individual in the society
 It takes great deal of money, effort and time
 Census is carried out every 10 years
(2) Sample survey
It is collection of data about a portion of
the population named the sample.
It requires less time, money and effort
than the comprehensive survey.
 A sample should be representative
C-Experiments  carried out in the
laboratory or field
Constant Variable
Observation which do A characteristic that takes
not vary from time to time on different values in
or from person to person different persons, places or
things  sex, age
 number of fingers,
number of eyes
The type of variable can be critically

important in determining method of

analysis that will be used.

 Categorical (Qualitative) Numerical (Quantitaive)
Variables which take the form of qualities or Variables expressed in the form of
names quantities
A- Two categories (Binary = Dichotomous=0/1): A- Continuous:
These often relate to the presence or absence of Obtained by measurement
some attribute Can take integer or fractional values
Male/female Can take any value between two fixed
Disease/No disease limits (upper/lower)
Married/Unmarried Age/Weight/Height/Distance/Blood
Diabetic/non diabetic pressure/temperature
Smoker/non smoker (ex-smoker?)
Hypertensive/normotensive
B- More than two categories: B- Discrete:
1= Nominal Obtained by enumeration or counting
Country of birth Take only integer values
Blood group A/B/AB/O Number of children
Married/Single/Divorced/Separated/Widow Number of beds
2= Ordinal: qualitative variables whose Pulse rate
categories can be put in a definite order Number of visits to the GP in a year
Non-smoker/ex-smoker/light smoker/heavy
smoker
Degree of pain: minimal-moderate-severe-
unbearable
Stages of breast cancer: I, II, III, IV (we cannot
say that stage IV is as twice as stage II) == to
distinguish from discrete
 Any quantitative variable can be transformed
to qualitative variable
Quantitative Variable Categorization Ordinal Two categories
Systolic blood pressure: >140 mm Hg Hypertensive Normal
90-140 Normal Abnormal
<90 Hypotensive
Blood glucose: >120 Hyperglycemia Normal
80-120 Normal Abnormal
<80 Hypoglycemia
Smoking >10/day Heavy Smoker
5-10 Moderate Non smoker
<5 Mild
Weight >80 Obese Obese
50-80 Average Not obese
<50 Underweight
 This may be defined as the assignment of numbers to objects or events according to
a set of rules.
• The Nominal Scale: As the name implies it consists of “naming” observations or
classifying them into various mutually exclusive and collectively exhaustive
categories. We can not quantify or order them.
• The Ordinal Scale: Whenever observations are not only different from category to
category but can be ranked according to some criterion, they are said to be
measured on an ordinal scale. Convalescing patients may be characterized as
unimproved, improved, and much improved. Individuals may be classified
according to socioeconomic status as low, medium, or high.
• The Interval Scale: The interval scale is a more sophisticated scale than the
nominal or ordinal in that with this scale not only is it possible to order
measurements, but also the distance between any two measurements is known.
We know, say, that the difference between a measurement of 20 and a
measurement of 30 is equal to the difference between measurements of 30 and 40.
• The Ratio Scale:The highest level of measurement is the ratio scale. This scale is
characterized by the fact that equality of ratios as well as equality of intervals may
be determined. Fundamental to the ratio scale is a true zero point. The
measurement of such familiar traits as height, weight, and length makes use of the
ratio scale.
• They are sometimes obtained by dividing two discrete variables but the
resultant rate or ratio is a continuous variable.
• Crowding index = family size/number of rooms.
• CBR = (LB/population) x 1000

• They are continuous and may be >100 as relative body weight (observed
body wt/desirable body wt) x 100 or may be negative as the percent
change.
• Occasionally the data in question are the relative positions of the members
of a group in some respect as ranking in sporting competition or
examination or when patients are given two or more treatments and asked to
express a preference.

• When it is not possible to take direct measurements it is often possible to

grade individuals in some way and each grade is given numerical value.

e.g. when the results of a test are scored as 0(negative),1(+), 2(++), 3(+++) or
classifying skin rash into 0 (mild), 1 (moderate), 2 (severe)
DESCRIPTIVE STATISTICAL
STATISTICS INFERENCE
Indicate whether each of the following variables is discrete
or continuous:
•the time it takes for you to get to school
•the number of Canadian couples who were married last
year
•the number of goals scored by a women’s hockey team
•the speed of a bicycle
•your age
•the number of subjects your school offered last year
•the length of time of a telephone call
•the annual income of an individual
•the number of employees at Statistics Canada
•the number of brothers and sisters you have
•the distance between your house and school
•the number of pages in a dictionary
Thank you