M1: Introduction to Biostatistics

Statistics
- Study and use of theory and methods for
the analysis of data arising from random
processes or phenomena.
Biostatistics
- Branch of applied statistics directed
toward applications in the health sciences
and biology.
Process of Applying Statistics
Making hypotheses Design experiments and
observational studies
Gathering data Summary of data
Drawing Interferences Testing of hypotheses Lesson 2.1 - Non-Experimental Studies -
Application of Biostatistics
Population Based Studies
Community medicine Advanced biomedical - Referred as “Observational Study”
and public health technologies
1. Descriptive Study
Pharmacology Cancer research
Ecology Demography  What? Who? Where? When?
Population genetics and Physiology and  Person: Characteristic (age, sex, occupation) of
statistical genetics Anatomy the individuals affected by the outcome
Systems Biology Bioinformatics  Place: Geography (residence, work, hospital) of
Agriculture Genetics the affected individuals
 Time: Of events (diagnosis, reporting; testing)
Lesson 2 - Types of Epidemiological Study occurred
Example:
a) Incidence Study
Newly reported or registered disease cases
compared over time, place, or person Population
estimates or other population group totals used as
denominators.
2. Analytic Study
 This study aims to generate a hypothesis by
answering the "why?" and "how?" questions.
 The goal of this study is to measure the
association between exposure and outcome.
 Uses a comparison group.
Example:
b) Ecological Study
Rates are linked to the level of exposure to some
agent for the group as a whole.
 Lesson 2.2 II. Individual Based Studies 2) Non-randomized Study
A. Descriptive Study  Quasi-Experimental Research
Example:  Shares similarities with traditional
Case Reports are prepared for illustrating novel, experimental design or randomized
unusual, or atypical features identified in patients in controlled trial, however it specifically
medical practice, and they potentially generate new lacks the element of random assignment to
research questions. treatment or control.
B. Analytic Study  Field Trial
Examples:  Applies preventive interventions to
1. Cross-Sectional Study healthy individuals
 To learn about the characteristics of a  Community Trial
population at one point in time  Applies intervention to aggregative units
 Does not use a comparison group
2. Case-Control Study Aims of Experimental Studies:
 To study rare diseases a) To provide scientific proof of etiological
 To study multiple exposures that may be factors which may permit modification or
related to a single outcome control of disease
3. Cohort Study/ Longitudinal Study/ Follow-up b) To provide a method of measuring the
Study effectiveness and efficiency of health services
 Can be used to find multiple outcomes from for the prevention, control and treatment of
a single exposure disease and improve the health of the
 A cohort is a well-defined group of community.
individuals who share a common
characteristic or experience. Lesson 3 - Data Collection and Organization
Data
Lesson 2.3 - Experimental Studies o Observations of random variables made on
 Randomized Study the elements of a population or sample.
- Clinical trial and Control trial o Quantities / numbers or qualities/ attributes
1) Randomized Controlled Trial (RCT) measured or observed to be collected or
- Study in which people are allocated at analysed.
random (by chance alone) to receive one Types of Data
of several clinical interventions. Observational Data Experimental Data
 One of these interventions is the standard of Collection of data based Produced by
comparison or control. on what is seen, heard measurement, test
 The control may be a standard practice, a and inferred. method, experimental
placebo ("sugar pill"), or no intervention at all. design.
Researchers should not
Someone who takes part in a randomized interfere with the The researcher has
controlled trial (RCT) is called a participant or subjects or variables in control over some
subject. any way. variables.
 RCTs seek to measure and compare the
outcomes after the participants receive the
interventions. Because the outcomes are
measured, RCTs are quantitative studies.
 Lesson 4 - Methods of Data Presentation Scatterplot
Textual The data gathered are presented in
paragraph form.
Data is written and read.
It is a combination of texts and figures.
Tabular Method of presenting data using the
statistical table.
Graphical The most effective manner of
presenting data since it can be easily
understood. Examples are Pie, Barr,
Venn, Histogram, Line Diagram and  Used to examine a relationship between two
Epidemic curve non-sequential variables.
Parts of a Tabular Data  It is very good for examining data to get a 'feel'
for the patterns, and identify extreme or unusual
Table Heading Consists of table number and
values (outliers) for checking or further
title
investigation.
Stubs Categories which are found at
the left side of the body of the
Line Plot
table
Box Head Top of the column
Body Main part of the table
Footnotes Any statement or note inserted
Source Note Source of the statistics

 Where the x- axis represents some

sequential variable like time, or distance
along a transection (right, and below).
 In both cases there is an explicit (spatial or
temporal) relationship between adjacent
points along the x-axis, and the inclusion of
the line makes the pattern of this sequence
much clearer.

Double Y Plot
 A common mistake is forgetting to indicate the
table heading - this should have the Table
Number and Title.

Lesson 4.1 - Types of Graphical Presentation

Graphical Presentation
- Powerful tool in quickly and effectively  Have two different y-axes, allowing variables
relaying information to your audience. with different scales to be plotted.
- “You are NOT supposed to draw a  Primarily used in the same sorts of situations as
conclusion based on a graph you are just line plots, where you want to compare the
going to explain what is on the graph." pattern of change in two different types of
variable.
Bar Charts Histogram

 A plot that lets you discover, and show, the

 Probably the most widely used type of graph in underlying frequency distribution (shape) of a
science. set of continuous data.
 Usually fairly straightforward to produce, and Frequency Table
generally are either used to represent means
(and appropriate error bars), as in the graph
here (right), or counts of some sort, including
proportions or percentages.
 This chart can be presented horizontally or
vertically.
 Method of organizing raw data in a compact
Area Plot
form by displaying a series of scores in
ascending or descending order, together
with their frequencies.
Venn Diagram

 Combine features of line plots and stacked bar

charts.
 They have similar disadvantages to the latter,
though since such plots are generally used to  A diagram representing mathematical or logical
show trends in time or space it is often sets pictorially as circles or closed curves within
easier to interpret, as the patterns of an enclosing rectangle (the universal set),
expansion or contraction of areas on the common elements of the sets being represented
graph have a logical meaning. by the areas of overlap among the circles.
Pie Chart Epidemic Curve

 Pie diagrams are familiar to everyone, much

beloved of business graphics packages and  A visual display of the onset of illness among
media, but of relatively limited use for scientific cases associated with an outbreak.
figures.
 Lesson 5 - Roles of Quantitative Research in
Public Health
Survey Study To estimate the extent of the
disease in the population.
Surveillance Designed to monitor or detect
Study specific diseases.
Observational Investigate association between an
Study exposure and a disease outcome.
They rely on “natural” allocation
of individuals to exposed or non-
exposed groups.
Experimental Investigate the association
Study between an exposure, often
therapeutic treatment, and disease
outcome individuals are
“intentionally” placed into the
treatment groups by the
investigators.