d nw o

Date 0 /0/23

Biostatistics Fpidemiology
Statichicsnw oi ye

3 t it
ciance O conduchng Studi s to colect, orgonic, sufmorite,
onaly Ond dro cOnchukion frodato.
COSAD

Biostatis tics aveth 5)

d lpmonrt ond p lication▇ concepts Ond technią

-ues to iolo coa sci nc S
Sto ic hícal rns atpw cea▇ te

Nariable characterich othribute that can osSume dif. val es

Random Vaniold"volues op voriaboe is predicted out op Chancee
Raw material of
statistic Data - Values Hhnat the variolo e could assumen When the values obtained arise as a
result of chance factors so that they
cannot be predicted in advance --
Data Sct - collection Op data volues bio z boa known as observation or
measurements
Pata Value/Datum *— ndividudl aluesin each dała seh

Maior Brondnes: hta▇ ▇ Dt

Descrphi e Stohishic COSP

Collection, Oroawnization, SuimmarBation, prsentotion Op data .

scrio inga sttuatio n .)ao a 0e
No conclusion can be drawn

Measure of central tendency and measures of dispersion

bt

infential Statistican sete iaiert

eneroliting fron Samp s to Population) perforting estimations
ond hypołhesis tests.h y he etn eo

dct rmini0 elationshipsomong vorialołes) and making prdictions

ebtaai iasE

Populaton" consict al su ojccts búing Studied (N) N

Somp n grou Op Subjccts Sekcted fom a population. (n) H

Should represent the population

ashtbo
 Date a m

Casficotions O Vorialole a

GualitatYG Norialde n erical and can lbe Ordure|ronked

shoe size (wnreruphion
SuJogr s i s c t e - char o ▇ ▇ Onthe valucs it coh as ute. (1,2,3,4)
foot length Often include fraction and

cortimuous – do nok posss gopc/inltnupkion ( Aart

decimals

Gsdia oviabe - voriabo s that Con lo placed into distnct ouncgorits

hne loccOrding to sOme Cøo octen's e ticatbute
Zip code, room number

hobe

Con uols - mus mcaSud ) n SIners must b ron eel bbcCUS

▇ the ▇ f mcosuring ze)

Boundar 1tA D st bDot

1:Boundory haul▇ Onc ▇ ▇higher the Volueoyaist

2. Boundary Should ndin 5 (ovorage o oll volues)

Masur nent cales NOIR)

Nominal Le - mutually exdusive (nonoverloping), exhauachng cohcg

rics in which no order oF r nking Con lbe impsed
abDOn te dota :ntne n ba0t
p code

noming osNot ns(reli gion , tınicy r 9ender)

Ordinal Level - classify data Wnto cotcg 1es that Con e Tonted;
houe , prccise differenCes loehu en the onks
largel
amado not iSt.(agc, tshirt si ▇ rains warning,
medium,
all house (small,

Interval Level - onksdata ,on d prccše differevce loetwten coits

too
grack
i9,
cosudo ▇ houtoter,tacrc isno nconingul Bro
Ratio Level - posse all te charoctnittic Of intorval mcSASurcnent,
length
heigth
weigt
Ond tor xst true/mcanivnopul oro-(length)
area a-
calls
true roto exişt when fhe Samc vari loe is thecos d)
# of phone

en Ho dionnt memiotrs f the popaotion (noiahtwtich )

 Date O / 03 /23

Data Collechon Sampling Techniqueh

Saucce Data
TouhneyeOn s exrivheuts RSEE

sSuveYSe itmal records

Bosic neth s o Sampling
Prolooloiliyn moke use op clhare to pick sample

↳ Non- proon bility

ondom Samplingrd rii▇

lout seo ▇ mcthod (shoulooll ,conveniznt)

DayEt
samp es ar Scleetd by using chonce mcthools or rondom humiben
Systmoti c Sawmpl ng
numlbering) tochn sulojcct F the P Pulation ond then s ectvg rury
kth Sulojcct. ( , , 1 , 2o)
Straticied Sompling
Samples
within the
strata should divide the population into Groups (strata) ccording to Some chov c
by randomly
selected -tzrstic tat is io portonł to the Study, then sonping from ch .
Cu s ter Sompling
Intact group
p ulotion i divided Into qvoups Colled clustzrs (ie. g ogr phical
dokto) ; Hhen the vscorchor vondomly Select sOme op these clusturs
ond uss oll tmioers of the cticoted clusto oc the aubjccts op te es Sa

Stotstica Stsdics
Can also be done using Corvotional Study- veScorchhor mercly doservcs Ond tric to drow)
variables That cannot be Definite cause and
manipulated by the
researcher
effect cannot be
shown
Conclusions lbased on thesc ooservahons (no intonventiar)
xprimental Słtudy - rescordho monipulotes c e O the voriololcs
Ond tries to detrmine hou the mom pulation
in pl ences otaer varialolesunnatural setting
who knew thatthey were part of the study changed their ways.
-1924; subjects
owthornc Eet
 Date

Vorial les i Stokisfcal Stdi s

Inde ndl nt[explanatory)
Confounding lindlu nes depundot vor)
Du ndentcoutcome)

Mish es O Statistics
Suspuct SamP sno -DekachedSatistic Foulky Suv
Anbiquous Averog th (mplicd Co nechic Quesho C

Chnangino the Sul jeet Miskodin Grophsy -RecallBias.

이nuohR07 7 기

a 0 0다 0이 Do

(o 장 이ebda

대0 이f구 o 7
n공z 025M

t이 He n0 자 이

2a다r e o n

n (

2))
 -

Descriptive Statistics
organizing data: frequency distribution and graphs

Raw data
• unorganized data in raw form
Frequency distribution
• organization of raw data in table form, using classes and frequencies
Classes
• groups; qualitative/quantitative category
Frequency
* summation
• number of data values contained in a specific class S
Class Limits
• upper and lower limits of class; applicable only to quantitative data Total frequency should correspond to The
number of data values in raw data

two main types

categorical frequency distribution qualitative note: make sure that classes are exhaustive
• used for data that can be placed
in specific categories, such as
nominal or ordinal level data

grouped frequency distribution quantitative

note: in this distribution the values of 24 and 30 of
• when the range of the data is the first class is known as class limits. the lower class
large, the data must be grouped limit, 24, represents the smallest value that can be
included in the class and vice versa.
into classes that are more than
one unit in width
Class boundary
upper limit of 1st class + lower limit of next class
2
• used to separate classes so that there are no gaps in the frequency
distribution; ensures that frequency distribution has no gap
determine the increment first

Class width
• how large or wide a particular class; found by subtracting
the lower class limit of one class from the lower class limit
of the next class
General rules for constructing
Reasons for constructing a frequency distribution
grouped frequency distribution
1. there should be between 5 - 20 classes • to organize the data in A meaningful, intelligible
2. it is preferable (but not necessary) that the class width way
be an odd number to ensure that the class midpoint is • to enable the reader to determine the nature
the same place value as the data
midpoint - central number of the class
(upper + lower / 2)
or shape of the distribution
3. the classes must be mutually exclusive (no overlapping) • to facilitate computational procedures for
4. the classes must be continuous no
measures of average and spread
gap

5. the classes must be exhaustive • to enable the researcher to draw charts and
6. classes must be equal in width graphs for the presentation of data
• to enable the reader to make comparisons
except for open-ended distribution

among different data sets

steps in constructing gfd

Cumulative frequencies - a distribution that

shows the numbers of data less than or equal
1a. Find the highest and lowest value
2a. Find the range
Cumulative frequency
• as you go along it accumulates to a specific value
3a. Select number of classes desired
4a. Find class width (divide range by number of classes)
5a. determine class limits (choose lowest number)
6a. develop class boundaries and tally
 Graphical Presentation of frequency distribution
boundary)
Histogram Cyaxis:frequency;X axis:class boundary(
class
Cumulative frequency Graph (Ogive) (x axis upper
=

• a graph that displays the data by using contiguous • a graph that represents the cumulative
vertical bars of various heights to represent the
frequencies of the classs
frequencies for the classes in a frequency
distribution Logistics curve - peak ...
Absolute - actual number
Frequency polygon (x axis midpoint; y=frequency)
=

Relative - percentage

• a graph that displays the data using lines that

connect points plotted for the frequencies at the
midpoints of the classes ·
• frequencies are represented by the heights of
the points

Peak - Mode; highest frequency

statistic
Distribution Shapes Other types of graphs • a characteristic or measure obtained by using the
Bell shaped Bar Graphs data values from a sample
Uniform Pareto Chart parameter
J-shaped Time Series Graph • a characteristic or measure obtained by using all
Reverse J-Shaped Compound Time Series Graph the data values from a specific population
Right-Skewed Pie Graph
Left-Skewed Stem and Leaf Plots General Rounding Rule: do not round off in the middle
Bimodal
Misleading graphs - depends on intent
of calculations; applicable in manual computation
U-Shaped mean for ungrouped data
• raw data; add then divide by number of
Concepts elements
Measure of central tendency
• describes where the distribution may be mean for grouped data
X
“centered” • multiply frequency with midpoint f.xm iXm
=

• concepts of center: center of gravity (mean),

value in the middle (median), most typical value note: mean should have one decimal place higher
(mode) greek parameter
than the place value of the values
-

statistics
roman 41:52
average:
unimodal
Mean • one value occurs with greatest frequency
• arithmetic average; equal to the sum of total of
all values divided by the number of values bimodal
• affected by presence of outliers in the data • two value occurs with greatest frequency

median x
x M E multimodal
=

• middlemost value • having more modes; more than two values stand out
• obtained by sorting the values from lowest to
highest and getting the value in the middle Modal Class
• preferred to be used as a typical value than • class with highest mode or highest frequency
mean when distribution is skewed
• affected less than the mean
X Eim Midrange
=

• the sum of the lowest and highest value in the data

mode set, divided by 2; sensitive to outliers
• most frequently occurring value
• most descriptive when distributions are highly- Weighted Mean
peaked(leptokurtic), suggesting large • multiply each value by its corresponding weight and
concentration on a single value dividing the sum of the products by the sum of the
• some data sets may have more than one mode; it weights
is also possible not to have a mode
 Symmetric Measures of Position
• right-side is mirror image of the left side • measures that describe the position or location of
particular values along the cumulative distribution
Skewed (where is a value located along the distribution? ; is
• asymmetric distribution and describing where tapering it among the highest values or among the lowest
Skewed to the Right (positively skewed): right tail is longer; values or in the middle?)
more values concentrated on the left (more lower values) • they are sometime useful for determining cut-off
Skewed to the Left (negatively skewed): the left tail is points for certain categories
longer; more values concentrated on the right (more higher
values) Standard Score (z Score)
• number of standard deviation
Shape adjectives • obtained by subtracting the mean and value and
Kurtosis dividing the result by the standard deviation
• flatness or peakness of distribution table
Percentile
Flat (Platykurtic) • divide the data into 100 equal parts
• heavy tails
Decile
Highly-peaked (Leptokurtic) • divide the data into 10 equal parts
• fat tails
Quartile
Mesokurtic • position in fourths that a data value holds in the
• medium tails distribution

Measures of Variation/Dispersion Interquartile Range

• measure the spread or variability of the values from • the range of values bounded by the 25th and 75th
each other percentiles
• it gives information on the values of the 50% of the
Range data
• simplest measure of dispersion, used to get a quick
idea of the spread Outlier
• the difference between highest and lowest values • an extremely high or an extremely low data value
(waste of information) when compared with the rest of the data values

Variance Reasons for Outliers:

• average of the squared deviations of values from the • the data value may have resulted from a
mean measurement or observational error
• measured in square of the original units (makes it a • the data value may have resulted from a recording
problem for interpretation) error (typo)
• the data value may have been obtained from a
Standard Deviation subject that is not in the defined population
• square root of the variance • the data value might be a legitimate value that
• measured in the unit as that of the data values occured by chance (although the probability is
extremely small)
Coefficient of Variation
• ratio of the standard deviation to the mean Exploratory Data Analysis
• used to compare the measure of spread between sets • counterpart of traditional methods
of data that are measured in different units

Skvariance >

parameter statistics
0 S
Basic Epidemiological Concepts and Principles

Etiology and the Natural History of Disease

Etymology - Greek root
• epi - “upon” Stages of Disease Prevention Types of Response
Pre-disease Primary Prevention Health Promotion,
• demos - people “population”
Specific Protection
• logos - “study”
Latent Disease Secondary Prevention Pre-symptomatic,
diagnosis and
Epidemiology
treatment
• study of factors that determine the occurence
and distribution of disease in a population Symptomatic Disease Tertiary Prevention disability limitation
• study of diseases and how it spreads for early
symptomatic
diseases,
Epidemiologists rehabilitation for
• public health proffessionals who investigate late symptomatic
disease
patterns, causes of disease, and injury of
humans
Mechanisms and Causes of disease
Types of Epidemiology Biological Mechanism vs. Social, Behavioral, and Environmental
Classical Epidemiology Causes of a Disease
• population oriented
• studies the community origins of health problems Host Factors
• interested in discovering risk factors • responsible for the degree to which the individual is able
to adapt to the stressors produced by the agents
Clinical Epidemiology • Host Resistance:
• studies patients in health care settings - Genotype
• improve prevention , early detection, diagnosis - Nutritional Status
treatment, prognosis, and care of illness in - BMI
individual - Immune System
- Social Behavior
Syndromic Epidemiology
• looks for patterns of signs and symptoms that Agents of Diseases or Illness
indicate an origin in bioterrorism Categories:
Biologic Agents
Infectious-disease Epideiomology • allergens, infectious organisms, biological toxins, food
• study of the complex relationship among hosts Chemical Agents
and infectious agents • chemical toxins, dusts
Physical Agents
Chronic-disease Epidemiology • kinetic energy, radiation, heat, cold, noise
• addresses the etiology (origin), prevention, Social and Psychological Stressors
distribution, natural history, and treatment • anxiety, depression
outcomes of chronic health disorders
Environment Factors
Scientific Study of Disease • influences the probability and circumstances of contact
the scientific study of disease can be between the host and the agents
approached at the following four levels:
1. submolecular or molecular level (e.g. cell biology, Vectors
genetics, biochemistry, and immunology) • to be an effective transmitter of disease, a vector must
2. tissue or organ level (e.g. anatomic pathology) have a specific relationship to the agents, the environment,
3. level of individual patients (e.g. clinical medicine) and the host.
4. level of populations (e.g. epidemiology) • May include:
- insects, arachnids, mammals
- human groups
- inanimate objects
- part of the environment
Preventable Causes Ecoligical Issues in Epidemiology
• Extrinsic vs. Intrinsic Solution of Public Health Problems and Unintended
• In 1964, WHO estimated that the majority of Creation of New Problems
cancer cases were potentially preventable and were • a change in one part of a system can lead to an inevitable
caused by “extrinsic/environmental factors” change in other parts of the system
• Man-made or naturally occuring carcinogens, viral
infections Effects of Vaccination
Diptheria
Risk Factors • caused by bacteria,
“B” - Biological and Behavioral Factors • In 1990, major epidemic appeared in Russia despite
• Influenced by gender, age, weight vaccination among individuals
• By 1992, 72% of cases = older than 14 years old
“E” - Environmental Factors Smallpox
• patterns of disease, develop and test hypothesis • caused by Variola virus
about causal factors, and introduce methods to Poliomyelitis
prevent further cases of dieases • caused by Poliovirus
• transmitted by person to person through fecal-oral route
“I” - Immunologic Factors by vehicle (food)
• Smallpox - first infectious disease known to have • 2 Types of Vaccine : inactivated or killed-polio vaccine
been eradicated from the globe (IPV) ; live, attenuated oral-polio vaccine (OPV)
• Herd Immunity - when a vaccine diminishes an
immunized person’s ability to spread the disease, Effects of Sanitation
leading to reduced disease transmission 19th Century
• Immunodeficiency - genetic abnormalities; • Diarrhea - primary killer of children
infections; certain vaccines • Tuberculosis - leading cause of adult mortality
Sanitary Revolution
“N” - Nutritional Factors • reduced infant mortality; increased effective birth rate
• dietary variations play an important role in • but a cause of today’s worldwide population problem
producing difference in disease rates among Demographic Graph
populations • difference between birth rate and death rate that
develops when a country undergoes demographic transition
“G” - Genetic Factors
• Genetic Epidemiology - addresses the distribution 3. Vector Control and Land-use Pattern
of normal and abnormal genes in a given population 4. River Dam Construction and Patterns of Disease
• Heritability - distribution of genes relative to all
determinants of disease Synergism
• Genetic Screening - important in identifying • interaction or combination of factors that produce a
porblems in newborns and in determining greater effect than the sum of their idk huhuhuhuhuhuh
susceptibility rate

“S” - Services, Social Factors, and Spiritual Factors

• Medical care Services - may be beneficial to
health, but can also be dangerous
• Iatrogenic-disease - occurs when a disease is
induced inadvertently by treatment or during a
diagnostic procedure
• Social and Spiritual factors - personal beliefs,
religious faith; family support, social networks
Epidemiologic Surveillance and Epidemic Outbreak
Investigation

Epidemics
• disease outbreaks; occurrence of a disease at an
unusual or unexpected, elevated frequency

Surveillance
• process of collecting, analyzing, interpreting, and
reporting data on the incidence of death,
diseases, and injuries and the prevalence of
certain conditions
• knowledge of which is considered important for
promoting and safeguarding public health

Creating a Surveillance System

• clear objectives
• case definition
• intensity of planned surveillance (active vs.
passive)
• duration of surveillance (on-going vs. time-
limited) E Establish Diagnosis
Case Definition
E Epidemiologic
• types of analysis needed Establish
d is an Epidemic occurring
• plans on information dissemination
I Characterize epidemic by time, place, person
Develophypothesis
Epidemiologists study the patterns of disease based i test hypothesis

on: Initiate Control Measure

surveillance
• time Initiate specific follow-up

• geography
• characteristics of person involve

Functions of Disease Surveillance

1. Establishment if Baseline Data
• base line (or usual) rates and patterns of disease
van be known only if there is a regular reporting
and surveillance system
• continued surveillance allows epidemiologists to
outbreaks
detect deviations from the usual pattern of data entation of

2. Evaluation of Time Trends

• Secular (Long-Term) Trends
- evalutation of disease patterns over longer
periods of time
- is the trend caused by
changes in disease detection