The Rule of Biostatistics In

Health Care Delivery

Dr. Hadeel Ahmed
)MBBS, MPH, Reg MD(
Definitions
• Statistics is the study of methods and
procedures for collection, classification,
summarization, presentation, analysis and
interpretation of data for making scientific
inferences from such data.
 Objectives of medical statistics
Detection of variation in characteristics among individuals for setting*
ranges of norms for population
Prediction of likely outcome of disease intervention programs in*
communities
Selection of appropriate forms of treatment for individual patients*
Public health administration and planning*
Planning, conducting, analyzing, interpreting and reporting of medical*
research
 Functions of Statistics

 Presentation of facts
 Comparison of facts
 Establishment of relationship
 Measure the effects
 Formulate policies in different fields
 Test a hypothesis
 Forecasting
 Functions of Statisticians

 Planning the statistical investigation

 Setting the objectives and define the scope of the investigation
 Collecting the data according to predetermined objectives
 Analyzing the collected data
 Presenting the data on a proper form
 Prepare the final report
 Do forecast on the basis of the data collected for the formulation of
policies and program design
Definitions
• Biostatistics is the application of statistical
methods to biology, medicine and public
health.
Statistics
• It deals with development and application of the
most appropriate methods for the:

Collection of data
Presentation of the collected data
Analysis and interpretation of the results
Making decisions on the basis of such analysis
 DATA VERSUS INFORMATION
Data is raw, unorganized facts that need to be processed

 When data is processed, organized, or presented in given

context so as to make it useful, it is called information
Statistical Data
• Data collection is the first step for statistical
studies.
• The data can be collected through primary source
or obtained from secondary sources:
Primary data is the data collected by investigator from
personal studies

Secondary data is the data obtained from some

secondary sources such as records, journals, magazines,
reports, newspapers
 Sources of data

Records Surveys Experiments

Comprehensive Sample
 Type of Statistics

Statistics

Descriptive Statistics Inferential Statistics

 Descriptive versus Inferential
statistics

Sample
Population refers to any well

defined group of individuals who

are being studied (university

students, all patients of a hospital

Population
.suffering from tuberculosis…..)
 Descriptive versus Inferential
statistics

Sample
Sample is a small group of a
population selected using a
suitable method of sampling

It can be regarded as
representative of the entire Population
population
 Descriptive statistics

Descriptive Sample
statistics
describing what is -
in
the sample data

Population
 Descriptive Statistics
Frequency
• Is the study of the
procedures which deal with 50
the collection, processing
and summarization of data 25
to make it more informative
and comprehensive. 0
20-29 30-39 40-49 50-59

• It involves presentation in Age Group

graphical and tabular
approaches.
 Inferential statistics
Sample
Sampling a population so
as to draw conclusions
about )make inferences(
the population of interest

Population
 Inferential Statistics (1)

Statistics that are

used to “infer” Population

something about Sample

the
Population based
on the data
collected from
the sample
drawn from the
population
 Inferential Statistics (1)

The sample should be

appropriately selected
to be able to infer from Population
the smaller sample to
Sample
the entire population.
 The sample must be
correctly drawn.
 Sampling is a critical
element of inference.
 Biostatistics

Descriptive Biostatistics Inferential Biostatistics

-
Collecting- Making Inférence-
,Hypothèses testing-
,Organizing-
Determining relationship-
Summeraizing- Making prédictions-
Presenting Data-
 Definitions
Data
• Are observations made on individuals.
Population:
• Aggregate of individuals of people, things, events with
common characteristic(s), e.g. university students,
hospitals, attacks of diarrhea ),
Variable:
• Is a quality, characteristic or a constituent of an
individual, that can be measured and is subject to
change.
 Types of data

Constant
Variables
 Variables & Constants
• Constants
Characteristics which does not vary or is fixed

• Variables
A characteristic of a person, object or
phenomenon that can have different values
 Data Vs variable
Variable
: Gender

Male – female – female – male – female …etc

Data
Variable - Examples

Variables Description
In yearsAge
1 = Male 2; = FemaleGender
1 = Present; 2 = AbsentMalaria
Weight in kg
Height in cm
 Variable - Examples

 Height 167cm
 Number of children 3
 Sex Male
 Ethnic group Chinese
 Severity of illness Moderate
 Status Alive
 Sources of Data

• Surveys/studies (households – health facilities)

• Records (registers of birth and death – patient’s record)

 Types of Data
Primary Data:
• Collected for the specific study (through survey or experiment)

Secondary Data:
• Collected for purposes other than the specific study ( from records
and reports)
 Classification of variables
Qualitative variables (categorical ):
Are those for which the value indicates
different groupings
1. Nominal:
• have distinct levels that have no inherent ordering
• Blood group (A, B, AB, O) – occupation (employee
worker self-employed – retired)
• When only two options (gender : male female ) it is
called binary or dichotomous
• Colour (red, blue, green)
 Classification of variables
2. Ordinal :
• Have levels that do follow a distinct ordering
• Stage of labour – socioeconomic status.
 Categorical variables, ordinal
Ordinal variables:
• These are grouped variables that are ordered or ranked in
increasing or decreasing order:

For example:
• High income (above 500 ,000 SD per month);
• Middle income (200,000-500,000 SD per month); and
• Low income (less than 200,000 SD per month).

Other examples are:

• Disability: no disability, partial disability, serious or total
disability
• Seriousness of a disease: severe, moderate, mild
 Classification of variables
Quantitative variables
Those for which the value has numerical meaning
1. Discrete :
• Discrete variables are variables that can take on only a
finite number of outcomes
• No possibility of fractions eg:
• Number of students
• Number of visits to a clinic (0, 1, 2, 3, 4, etc).
 Classification of variables
2. Continuous:
• Are values that can fall anywhere corresponding to
points on a line segment.
• Possibility of fractions
• Some examples are weight and diastolic
blood pressure
• Height in centimeters (2.5 cm or 2.546 cm or
2.543216 cm)
• Temperature in degrees Celsius (37.20C or
37.199990C etc.)
 Type of variables

• Severity of pain

• Weight of study subjects

• Number of babies delivered in a hospital

• Height of study subjects

• Gender, race of study subjects

• Numbers of boys, girls,

• Number of episodes of diarrheal diseases per years ,

Type of variables
Variables

Quantitative Qualitative
)Numerical data( (Categorical data)

-is un-measurable
Data represented-
- un-expressible in
numerically
magnitude or
e.g. weight and numerically
- it can be
height of study
subjects expressed in
qualities.

e.g. gender, race

of study subjects

Severity of pain
 Type of variables

variables

Quantitative
Qualitative

Discrete
The data can Continuous
only have The data can have
known fixed -any value within a
values and can defined range
e.g. Numbers of
be counted -any value in a
boys, girls,
continuous interval of
number of
measurement.
episodes of
e.g. weight, height age,
diarrheal diseases
B.P and blood glucose
per years ,number
level.
of babies
Type of variables
)variables

Quantitative Qualitative

Ordinal variables can be Nominal variables

rank ordered are used to classify
e.g. pain severity: mild, individuals into
moderate, severe…. distinct groups can
not be rank
ordered e.g. blood
group, gender….
 Types of variables
1. Dependent

2. Independent

3. Confounding

4. Universal (Background)
 Study
• In one study which was designed to determine the
maternal mortality. The aim of the study was to
identify the effect of maternal risk factors such as age,
parity, nutritional status, smoking, weight, occupation
on the prevalence of maternal mortality.

• What is the dependent variable?

• What is the independent variable?
 Answer
The dependent variable:
Prevalence of maternal mortality

The independent variables:

Age, parity, nutritional status, smoking, weight and occupation.
Example
 Data Presentation
Data can be presented in tables, graphs, charts, diagrams.
The presentation will depend on the type of data:
• Qualitative variables
• Frequency distribution tables
• Bar chart
• Pie chart

• Quantitative variables
• Frequency distribution tables
• Relative frequency tables
• Histogram chart
 Data Presentation
Qualitative variables
• Tables:
• Frequency distribution table
• Cross tabulation
• Graphs:
• Bar chart
• Pie chart
• Pictogram
 :Presentation of qualitative data

• 1.Tabulation form :
The basic rule for displaying qualitative data is to:
1.Classify them to categories
2. Then count the number of observations in
each category of the variable and
3. Present the number and percentage in a table.
 Data Presentation
Qualitative variables

• Frequency distribution

Method of Number of birth Percentage

delivery
Normal 478 79.7%

Forceps 65 10.8%

Caesarian 57 9.5%

Total 600 100.0%

• Cross tabulation:
• A frequency table involving at least two variables that have been
cross classified.

• Cross tabulation were obtained in diagnostic test evaluation study :

• Complex frequency distribution Table

 Complex frequency distribution Table
Table (III): Distribution of 20 lung cancer patients at the chest
department of Khartoum teaching hospital and 40 controls in May
2008 according to smoking

Lung cancer
Total
Smokin Cases Control
g No. % No. % No. %

Smoker 38.3
15 75% 8 20% 23
3
Non 61.6
smoker 5 25% 32 80% 37
7
Total 20 100 40 100 60 100
 :Presentation of qualitative data
• 2.Graphical form :
1. Bar chart
2. Pie chart
• Bar chart: is used with categories data ;
with nominal data, the bars may be in any order
that seen sensible to research, but with ordinal
data they should be arranged from lowest to highest.
The bars may be horizontal or vertical.
 Bar Chart
• Display frequency distribution
• Has columns with same width
• There are spaces between columns
• Bars may be horizontal or vertical

60

50

40
Number of
respondents 30
(% )
20

10

0
Malay Chinese Indian Others
Bar chart
 Pie Chart
• Shows frequency distribution
• Represented by the angles in different sectors of a circle
• The total angle 360o represents 100%
7%
Chinese
Malay
31% I ndian
22%
Others

40%
Proportion of ethnic groups in a
Pictogram