Contact details:

DR. JOY C. CHAVEZ

[email protected]
09985538526
COURSE OUTLINE
 PRE-TEST
1. The stages of a malignant disease (cancer) is recorded using the symbols 0, I, II, III, IV. We say that the
scale used is:
A. Alphanumeric
B. Numerical
C. Ordinal
D. Nominal

2. The fundamental statistical indicators are:

A. Mean
B. Median
C. Variance
D. Standard deviation
 PRE-TEST
3. Which of the following is NOT included of the the average of a series of numerical values?
A. The sum of the values divided by their number
B. Lower than the minimum value in the series
C. Lower than the maximum value in the series
D. An indicator of central tendency for the values of the series

4. The median of a series of numerical values is: A. Equal to the average

B. A graph or hart
C. A number
D. A frequency table
5. The median of a series of numerical values is:
A. A value for which half of the values are higher and half of the values are lower
B. The value located exactly midway between the minimum and maximum of the series C. The most
commonly encountered values among the series
D. A measure of the eccentricity of the series
INTRODUCTION TO
BIOSTATISTICS
DR. JOY C. CHAVEZ
 This session covers:

Origin and development of Biostatistics

Definition of Statistics and Biostatistics
Reasons to know about Biostatistics
Types of data
Graphical representation of a data
Frequency distribution of a data
“Statistics is the science which deals with
collection, classification and tabulation of
numerical facts as the basis for explanation,
description and comparison of
phenomenon”.
-Lovitt
Origin and development of statistics in
Medical Research
In 1929 a huge paper on application of statistics was
published in Physiology Journal by Dunn.
In 1937, 15 articles on statistical methods by Austin
Bradford Hill, were published in book form.
In 1948, a RCT of Streptomycin for pulmonary tb., was
published in which Bradford Hill has a key influence.
Then the growth of Statistics in Medicine from 1952 was
a 8-fold increase by 1982.
 C.R. Rao
Douglas Altman Ronald Fisher Karl Pearson

Gauss -
 “BIOSTATISICS”
(1) Statistics arising out of biological sciences, particularly from the
fields of Medicine and public health.
(2) The methods used in dealing with statistics in the fields of
medicine, biology and public health for planning, conducting and
analyzing data which arise in investigations of these branches.
Reasons to know about
biostatistics:
Medicine is becoming increasingly quantitative.
The planning, conduct and interpretation of much of medical research
are becoming increasingly reliant on the statistical methodology.
Statistics pervades the medical literature.
 Example: Evaluation of Penicillin
(treatment A) vs Penicillin &
Chloramphenicol (treatment B) for treating
bacterial pneumonia in children< 2 yrs.

What is the sample size needed to demonstrate the significance of one

group against other ?
Is treatment A is better than treatment B or vice versa ?
If so, how much better ?
What is the normal variation in clinical measurement ? (mild, moderate &
severe) ?
How reliable and valid is the measurement ? (clinical & radiological) ?
What is the magnitude and effect of laboratory and technical
error ?
How does one interpret abnormal values ?
 CLINIC
Documentation of medicalAL history of diseases.
Planning and conduct of clinical studies.
MEDICI
Evaluating the merits of different procedures.
In providing methods for definition of “normal” and “abnormal”.
NE
 PREVEN
To provide the magnitude ofTIVE
any health problem in the community.
To find out the basic factors underlying the ill-health.
MEDICI
To evaluate the health programs
community (success/failure).
which was introduced in the

To introduce and promote NEhealth legislation.

 WHAT
Planning DOES
Design
STATIST
Execution (Data
Data Processing
collection)

ICS
Data analysis
Presentation
COVER
Interpretation
Publication
 HOW A
Design of study “BIOSTAT
Sample size & power calculations ISTICIAN
Selection of sample and controls
Designing a questionnaire ” CAN
Data Management
HELP
Choice of descriptive statistics & graphs
?
Application of univariate and multivariate
statistical analysis techniques
INVESTIGATION
 SOURCES
Routinely Kept Records OF DATA
Surveys
Experiments
External Sources
 TYPES
OF
QUALITATIVE DATA
DATA
DISCRETE QUANTITATIVE
CONTINOUS QUANTITATIVE
 What is

biostatistics?
It is the science which 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

 The methods used in dealing with statistics in the fields of medicine, biology and public
health.
 Statistics

Descriptive Statistics Experimental Statistics Inferential Statistics

Evaluation of the
Techniques for planning
Methods for processing, information generated by
and conducting
summarizing, presenting an experiment or through
experiments
and describing data observation
 Descriptive
statistics
◦ Summarizing and describing the data

◦ Uses numerical and graphical summaries to characterize sample

data
 DAT
 Definition:- A
A set of values recorded on one or more observational units. Data
are raw materials of statistics.
 Sources of data:-
Experiments
Surveys
records
 Types of data

Quantitative Qualitative
data(numerical) data(categorical
)

continuous Discrete Nominal Ordinal

 Quantitative data
•Discrete: Reflects a number obtained by counting—no decimal.
•Continuous: Reflects a measurement; the number of decimal places depends
on the precision of the measuring device.
• Ratio scale: Order and distance implied.
Differences can be compared; has a true zero. Ratios
can be compared. Examples: Height, weight, blood
pressure
• Interval scale: Order and distance implied.
Differences
can be compared; no true zero. Ratios cannot be
compared.
Example: Temperature in Celsius.
 Qualitative data
 Ordinal data
Difference and order are implied BUT, intervals are no longer
equivalent.
Example :- ranking syatem
 Nominal data
Only difference is implied.
Observations are classified into mutually exclusive categories.
Examples: Gender, ID numbers, pass/fail response
Measurement and
measurement scales
 QUALITATIVE
Nominal
Example: Sex ( M, F)
Exam result (P, F)
Blood Group (A,B, O or AB)
Color of Eyes (blue, green,
brown, black)
ORDINAL
Example:
Response to treatment
(poor, fair, good)
Severity of disease
(mild, moderate, severe)
Income status (low, middle,
high)
QUANTITATIVE (DISCRETE)

Example: The no. of family members

The no. of heart beats
The no. of admissions in a day

QUANTITATIVE (CONTINOUS)

Example: Height, Weight, Age, BP, Serum

Cholesterol and BMI
Discrete data -- Gaps between possible values

Number of Children

Continuous data -- Theoretically,

no gaps between possible values

Hb
 CONTINUOUS DATA

DISCRETE DATA

wt. (in Kg.) : under wt, normal & over wt.

Ht. (in cm.): short, medium & tall
Table 1 Distribution of blunt injured patients
according to hospital length of stay
hospital length of stay Number Percent
1 – 3 days 5891 43.3
4 – 7 days 3489 25.6
2 weeks 2449 18.0
3 weeks 813 6.0
1 month 417 3.1
More than 1 month 545 4.0
Total 14604 100.0
Mean = 7.85 SE = 0.10
 Scale of
measure
Qualitative variable:
A categorical variable
ment
Nominal (classificatory) scale
 - gender, marital status, race

Ordinal (ranking) scale

 - severity scale, good/better/best
 Scale of measurement
Quantitative variable:
A numerical variable: discrete; continuous

Interval scale :
Data is placed in meaningful intervals and order. The unit of
measurement are arbitrary.

- Temperature (37º C -- 36º C; 38º C-- 37º C are equal) and

No implication of ratio (30º C is not twice as hot as 15º C)
Ratio scale:
Data is presented in frequency distribution in logical order. A
meaningful ratio exists.

- Age, weight, height, pulse rate

- pulse rate of 120 is twice as fast as 60
- person with weight of 80kg is twice as heavy as
the one with weight of 40 kg.
 Scales of
color, city
Measure
Nominal – qualitative classification of equal value: gender, race,

Ordinal - qualitative classification which can be rank ordered:

socioeconomic status of families
Interval - Numerical or quantitative data: can be rank ordered and
sizes compared : temperature
Ratio - Quantitative interval data along with ratio: time, age.
 Methods of presentation of
data
1 Tabular presentation
2 Graphical presentation
Purpose: To display data so that they can be readily understood.

Principle: Tables and graphs should contain enough information to be self-

sufficient without reliance on material within the text of the document of
which they are a part.

•Tables and graphs share some common features, but for any specific situation,
one is likely to be more suitable than the other.
 Tabular
 Types of tables:-
Presentation
1.list table:- for qualitative data, count the number of observations
( frequencies) in each category.
A table consisting of two columns, the first giving an identification
of the
observational unit and the second giving the value of variable for
that unit.
Department Number of patients
Example : number of patients in each100
Medicine hospital department are
Surgery 88
ENT 54
Opthalmology 30
 Tabular
Presentation
2. Frequency distribution table:- for qualitative and quantitative
data
Simple frequency distribution table:-
Tabular
Presentation
complex frequency distribution table

Lung cancer
Total
Smoking positive negative

No. % No. % No. %

Smoker 15 65.2 8 34.8 23 100

Non smoker 5 13.5 32 86.5 37 100
Total 20 33.3 40 66.7 60 100
 Graphical presentation
For quantitative,
For qualitative,
continuous or
discrete or counted
measured data
data
 Histogram
 Bar diagram
 Frequency polygon
 Pie or sector
 Frequency curve
diagram
 Line chart
 Spot map
 Scattered or dot
diagram
 Bar
diagram
 It represent the measured value
(or %) by separated rectangles
of constant width and its lengths
proportional to the frequency
Conditi ons for Whi ch Patients w e r e r ef er red for treat ment

 Use:- discrete qualitative data B a c k and N e c k

A rthritis

 Types:- simple A nxie ty

multiple
Skin

Condition
D ig e s tive
Headache

component G ynecol ogi c

R espi ratory
Ci rculatory
General
Blood
E n doc r i n e
0 20 40 60 80 100 120
N u m b e r o f P a t i en t s
Bar
diagram
Multiple bar chart:- Each
observation has more than one value
represented, by a group of bars.

Component bar chart:-subdivision

of a single bar to indicate the
composition of the total divided into
sections according to their relative
proportion.
 Pie
diagram
Consist of a circle whose area
represents the total frequency
(100%) which is divided into
segments.
Each segment represents a
proportional composition of
the total frequency
 Histogra
m
it is very similar to the bar chart with
the difference that the rectangles or
bars are adherent (without gaps).
It is used for presenting continuous
quantitative data.
Each bar represents a class and its
height represents the frequency
(number of cases), its width represent
the class interval.
Frequency polygon

Derived from a histogram by

connecting the mid points of the tops
of the rectangles in the histogram.
The line connecting the centers of
histogram rectangles is called
frequency polygon.
We can draw polygon without
rectangles so we will get simpler form
of line graph
Scattered diagram
It is useful to represent the
relationship between two
numeric measurements.
Each observation being
represented by a point
corresponding to its value on
each axis
 Measures of central
Variable tendacy
usually has a point (center) around which the observed
values lie.
 The three most commonly used averages are:
• The arithmetic mean
• The Median
• The Mode
 Measures of central
1. Mean:-tendacy
◦ The arithmetic average of the variable x.

◦ It is the preferred measure for interval or ratio variables with relatively

symmetric observations.

◦ It has good sampling stability (e.g., it varies the least from sample to
sample), implying that it is better suited for making inferences about
population parameters.

◦ It is affected by extreme values

 Measures of central
tendacy
2. Median:-
The middle value (Q2, the 50th percentile) of the variable.

 It is appropriate for ordinal measures and for skewed interval or

ratio measures.

 It has low sampling stability.

The rank of median for is (n + 1)/2 if the number of observation is
odd and n/2 if the number is even
 Measures of central
3 Mode:-tendacy
◦ The most frequently occurring value in the data set.

◦ May not exist or may not be uniquely defined.

◦ It is the only measure of central tendency that can be used with

nominal variables, but it is also meaningful for quantitative
variables that are inherently discrete.

◦ Its sampling stability is very low

 Measure of
to be.
dispersion
Measures of variability depict how similar observations of a variable tend

Variability of a nominal or ordinal variable is rarely summarized

numerically.
The measure of dispersion describes the degree of variations or dispersion
of the data around its central values: (dispersion = variation = spread =
scatter).
Range - R
Standard Deviation - SD
Coefficient of Variation -COV
 Measure of
Range:- dispersion
It is the difference between the largest and smallest values.
It is the simplest measure of variation.
Disadvantage:- it is based only on two of the observations
and gives no idea of how the other observations are
arranged between these two.
 Measure of
dispersion
 Standard deviation:-
 Represents the average spread of the data around the mean.
 “Average deviation” from the mean.
 SD = √ Σ (mean – x)2 / n – 1
 Uses:-
1. It summarizes the deviations of a large distribution from mean in one figure used as
a
unit of variation.
2.Indicates whether the variation of difference of an individual from the mean is by
chance, i.e. natural or real due to some special reasons.
3. It also helps in finding the suitable size of sample for valid conclusions.
 Measure of
 Coefficient ofdispersion
variation:-
The coefficient of variation expresses the standard deviation as a
percentage of the sample mean.

C. V = SD / mean * 100

C.V is useful when, we are interested in the relative size of the

variability in the data.
 Inferential statistics

It refers to the process of selecting and

using a sample to draw inference about
population from which sample is drawn.

 Two forms of statistical inference

◦ Hypothesis testing
◦ Estimation
 Hypothesis Testing
During investigation there is assumption and presumption which
subsequently in study must be proved or disproved.
Hypothesis is a supposition made from observation. On the basis of
Hypothesis we collect data.
Two Hypothesis are made to draw inference from Sample value-
A. Null Hypothesis or hypothesis of no difference.
B. Alternative Hypothesis of significant difference.
The Null Hypothesis and the Alternative Hypothesis are chosen before
the sample is drawn.
 Hypothesis Testing
Null Hypothesis states that the observed difference is entirely due to sampling error,
that is - it has occurred purely by chance.
E.g:- There is no difference in the incidence of measles between vaccinated and
non- vaccinated children.
 Alternative Hypothesis of significant difference states that the sample result is
different
that is, greater or smaller than the hypothetical value of population.
A test of significance such as Z-test, t-test, chisquare test, is performed to accept
the Null Hypothesis or to reject it and accept the Alternative Hypothesis.
Error in Hypothesis
Testing
 P
The probability of committingvalue
Type 1 Error is called the P-value. Thus p-
value is the chance that the presence of difference is concluded when
actually there is none.
When the p value is between 0.05 and 0.01 the result is usually called
significant.
 When p value is less than 0.01, result is often called highly
significant.
When p value is less than 0.001 and 0.005, result is taken as very highly
significant.
 Confidence Interval :-
The interval within which a parameter value is expected to lie with a
certain confidence level as could be revealed by repeated samples is called
confidence interval.
A point estimate from a sample population may not reflect the true value
from the population so it is often helpful to provide a range.
Confidence Level : The degree of assurance for an interval to contain the
value of a parameter
(1-α).
 Test of significance
Test of significance is a formal procedure for comparing observed
data with a claim (also called a hypothesis) whose truth we want
to assess.
Parametric Non parametric
Based on specific distribution such  Not based on any particular parameter
as Gaussian such as mean
 Used when the underlying distribution
is far from Gaussian (applicable to
almost all levels of distribution) and
when the sample size is small
 Test of significance
Parametric test Nonparametric
 Student’s t- test(one sample,  Sign test(for paired data)
two sample, and paired)  Wilcoxon Signed-Rank test
 Z test  Wilcoxon Rank Sum test
 ANOVA F-test  Chi-square test
 Pearson’s correlation(r)  Spearman’s Rank
Correlation(p)
 ANOCOVA
 Kruskal-Wallis
test
Test of significance
 Samplin
Sample size:-This is the g
sub-population to be studied in order to
make an inference to a reference population(A broader population to
which the findings from a study are to be generalized)
Optimum sample size determination is required for the following
reasons:
1. To allow for appropriate analysis
2.To provide the desired level of accuracy
3.To allow validity of significance test
 Samplin
Disadvantagesgof inappropriate sample size
Small sample Large sample
1 Even a well conducted study may fail 1The study will be difficult and costly
to answer it research question
2. Time constraint
2. It may fail to detect important effect
or associations 3. Loss of accuracy.

3.It may associate this effect or

association imprecisely
 Samplin
Random error: error thatgoccur by chance. Sources are sample
variability, subject to subject differences & measurement errors. It
can be reduce by averaging, increase sample size, repeating the
experiment.

Systematic error: deviations not due to chance alone. Several

factors, e.g patient selection criteria may contribute. It can be reduce
by good study design and conduct of the experiment.
 Samplin
g a variable has the same value when
Precision: the degree to which
measured several times. It is a function of random error.
Accuracy: the degree to which a variable actually represent the true
value. It is function of systematic error.
Approaches for estimating sample size and performing power
analysis depend primarily on:
1. The study design
2. The main outcome measure of the study
 Samplin
g that could be used for
There are four procedures
calculating sample size:
1. Use of formulae
2. Ready made table
3. Nomograms
4. Computer software
 Samplin
 USE OF FORMULAE FORgSAMPLE SIZE:-
The appropriate sample size for population-based study is determined largely by
3 factors
1. The estimated prevalence of the variable of interest.
2. The desired level of confidence.
3. The acceptable margin of error
 For population >10,000.
n=Z2pq/d2
 USE OF READYMADE TABLE FOR SAMPLE SIZE
Such table that give ready made sample sizes are available for different designs &
situation
 Samplin
g
 USE OF COMPUTER SOFTWARE FOR SAMPLE SIZE:-
Epi-info
nQuerry
Power & precision
Sample
STATA
SPSS
 TERMS USED TO DESCRIBE THE MAGNITUDE OF AN EFFECT
Relative risk — The relative risk (or risk ratio) equals the incidence in exposed
individuals divided by the incidence in unexposed individuals. The relative risk
can be calculated from studies in which the proportion of patients exposed and
unexposed to a risk is known, such as a cohort study

Odds ratio — The odds ratio equals the odds that an individual with a specific
condition has been exposed to a risk factor divided by the odds that a control
has been exposed. The odds ratio is used in case-control studies
Absolute risk — The relative risk and odds ratio provide an
understanding of the magnitude of risk compared with a standard.
However, it is more often desirable to know information about
the absolute risk.
The "attributable risk" represents the difference in the rate of a
disease in an exposed, compared with a nonexposed, population. It
reflects the additional incidence of disease related to an exposure
taking into account the background rate of the disease
Number needed to treat — The benefit of an intervention can be expressed by the
"number needed to treat" (NNT). NNT is the reciprocal of the absolute risk
reduction (the absolute adverse event rate for placebo minus the absolute adverse
event rate for treated patients).
NNTs from different studies cannot be compared unless the methods used to determine
them are identical.
When the outcome is a harm rather than a benefit, a number needed to harm (NNH)
can be calculated similarly..
TERMS USED TO DESCRIBE THE QUALITY OF
MEASUREMENTS
Reliability — Reliability refers to the extent to which repeated
measurements of a relatively stable phenomenon fall closely to each other.
Several different types of reliability can be measured, such as inter- and
intraobserver reliability and test-retest reliability.
Validity — Validity refers to the extent to which an observation reflects
the "truth" of the phenomenon being measured
 MEASURES OF DIAGNOSTIC TEST PERFORMANCE
Sensitivity — The number of patients with a positive test who have a disease
divided by all patients who have the disease. A test with high sensitivity will not
miss many patients who have the disease (ie, few false negative results).
Specificity — The number of patients who have a negative test and do not have the
disease divided by the number of patients who do not have the disease. A test with
high specificity will infrequently identify patients as having a disease when they do
not (ie, few false positive results).
sensitivity and specificity are interdependent. Thus, for a given test, an increase
in sensitivity is accompanied by a decrease in specificity and vice versa
Predictive values — The positive predictive value of a test
represents the likelihood that a patient with a positive test has the
disease. Conversely, the negative predictive value represents the
likelihood that a patient who has a negative test is free of the disease
Accuracy — The performance of a diagnostic test is sometimes expressed
as accuracy, which refers to the number of true positives and true
negatives divided by the total number of observations
Experimental
statistics
 Case Study
Design
Often a description of a individual case’s condition or response to an
intervention.
◦ data may be qualitative, quantitative, or both
◦ Case series: observations of several similar cases are reported
 Strengths:-
◦ Enables understanding of the totality of an individual’s (or organization,
community) experience
◦ The in-depth examination of a situation or ‘case’ can lead to discovery
of relationships that were not obvious before
 Case Study
Design
◦ Useful for generating new hypotheses or for describing new phenomena
 Weaknesses:-
◦ No control group
◦ Prone to selection bias and confounding
◦ The interaction of environmental and personal characteristics make it weak
in internal validity
◦ Limited generalizability
 Cross-sectional Study
 Researcher studies a stratified group of subjects at one point in time
 Strengths:-
◦ Fast and inexpensive
◦ No loss to follow-up (no follow-up)
◦ Ideal for studying prevalence
◦ Data is useful for planning of health services and medical programs
 Weaknesses:-
◦ Difficult to establish a causal relationship from data collected in a cross-
sectional time-frame
◦ Not practical for studying rare phenomena
 Cohort
A group of individualsStudy
who do not yet have the outcome of interest
are followed together over time to see who develops the condition
May identify risk by comparing the incidence of specific outcomes
in exposed and not exposed participants
 Types of cohort study..
Prospective (concurrent)
Retrospective (historical)
Restricted (restricted exposures)
 Cohort
 Strengths:- Study
◦ Powerful strategy for defining incidence and investigating potential
causes of an outcome before it occurs
◦ Time sequence strengthens inference that the factor may cause the
outcome
 Weaknesses:-
◦ Expensive – many subjects must be studied to observe outcome of
interest
◦ Potential confounders: eg, cigarette smoking might confound the
association between exercise and CHD
 Case-Control Study
 Identify groups with or without the condition
Look backward in time to find differences in predictor variables that
may explain why the cases got the condition and the controls did not
MATCHING
• CHARACTERISTICS OFTEN USED
–age
–gender
–body mass index (weight / height2)
–smoking status
–marital status
 Case-Control Study
Case control

Exposed A B

Unexposed C D

Total A+C B+D

Odds ratio:- A*D/B*C

OR = 1 then exposure is NOT related to disease
OR>1 then exposure is POSITIVELY related to disease
OR<1 then exposure NEGATIVELY related to disease
 Case-Control Study
Strengths
◦ Useful for studying rare conditions
◦ Short duration & relatively inexpensive
◦ High yield of information from relatively few participants
◦ Useful for generating hypotheses
Weaknesses
◦ Increased susceptibility to bias:
◦ Separate sampling of cases and controls
◦ Retrospective measurement of predictor variables
◦ No way to estimate the excess risk of exposure
◦ Only one outcome can be studied
 Experimental study
It is a collection of research designs which use manipulation and controlled
testing to understand causal processes.
 Generally, one or more variables are manipulated to determine their effect
on a
dependent variable
 Characteristics or Features of Experimental Design:-
1. Manipulation(M)
2. Control(C)
3. Randomization(R)
 Experimental study
 Types of Experimental Designs:-
• True-Experimental (Simple):- M+R+C
• Quasi-Experimental:-M+R or C
• Pre-Experimental:- M, no R or C
True experimental design

POST-TEST-ONLY CONTROL DESIGN

PRETEST-POST-TEST-ONLY DESIGN
 True experimental design
RANDOMIZED BLOCK DESIGN CROSSOVER DESIGN
TYPES OF QUASI-
EXPERIMENTAL
DESIGN
NONRANDOMIZED CONTROL GROUP DESIGN

TIME-SERIES DESIGN
 TYPES OF PRE-
EXPERIMENTAL
RESEARCH
ONE-SHOT CASE DESIGN DESIGN

ONE-GROUP PRETEST-POSTTEST DESIGN

 Experimental study
 Strengths:-
Controls the influence of confounding variables, providing more conclusive
answers
Randomization eliminates bias due to pre-randomization confounding variables
Blinding the interventions eliminates bias due to unintended interventions
 Experimental study
 Weaknesses:-
Costly in time and money
Many research questions are not suitable for experimental designs
Usually reserved for more mature research questions that have already been examined
by descriptive studies
Experiments tend to restrict the scope and narrow the study question
 Meta
Statistically combinesAnalysis
results of existing research to estimate overall
size of relation between variables.
 Helps in
• Developing theory
• Identifying research needs,
• Establishing validity
 Can replace large-scale research studies
 Better than literature reviews