Research Methodologies

Zikria, Ph.D.
Assistant Prof. Pharmacy Practice (BZU)
Re~Search
 Research is an ongoing process of hypothesis generation,
refutation, refinement, and corroboration
 Studious inquiry or examination; especially : investigation
or experimentation aimed at the discovery and interpretation
of facts, revision of accepted theories or laws in the light of
new facts, or practical application of such new or revised
theories or laws
 A detailed study of
a subject, especially in order to discover (new)
information or reach a (new) understanding
 Any gathering of data, information and facts for the
advancement of knowledge
“Research is a systematized effort to gain new
knowledge.” (V. Redman and A.V.H. Mory)

“Research is an honest exhaustive, intelligent

searching for facts and their meanings or implications
with reference to a given problem. The product or
findings of a given piece of research should be an
authentic, verifiable and contribution to knowledge in
the field studied.” (P.M. Cook)
Research?
Research simply seeks the answer of certain questions
which have not been answered so far and the answers
depend upon human efforts
Research?
Now the question arises

“Is the answer of the question in examination also

research”?

The answer is ‘no’, because the answers of these

questions are available (text-books, class-notes etc)

Research answers only those questions of which the

answers are not available in literature i.e., in human
knowledge
General characteristics of research
 Gathers new knowledge or data from primary or first –hand sources
 Places emphasis upon the discovery of general principles
 An exact systematic and accurate investigation
 Uses certain valid data gathering devices
 Logical and objective
 Researcher resists the temptation to seek only the data that support his/her
hypotheses
 Researcher eliminates personal feelings and preferences
 Endeavors to organize data in quantitative terms
 Patient and unhurried activity
 Researcher is willing to follow his/her procedures to the conclusions that
may be unpopular and bring social disapproval
 Research is carefully recorded and reported conclusions and
generalizations are arrived at carefully and cautiously
 Health Research
Health research → A systematic collection, analysis and
interpretation of health and health-related data to answer a
certain question or to solve a problem

Health systems research also known as health services

research or health policy and systems research
Kinds of research
 On the basis of objective of the research
Fundamental research
Applied research
 On the basis of approach of the research
Prospective research
Retrospective research
Cross sectional research
 On the basis of precision in research findings
Experimental research
Non experimental research
 On the basis of nature of findings
 Analytical research
 Descriptive research
 According to national science foundation
 Basic research
 Applied research
 Developmental research
 According to methodology
 Quantitative research
 Qualitative research
 Mix methodology
 Based on concept
 Conceptual research
 Empirical research
 Based on site of research
 Library research
 Field research
 Laboratory research
Health research triangle
 three operational interlinked categories of biomedical, health services and behavioral
research, the so-called health research triangle

Bio-medical Behavioral

Health services
Research Design
 The research design refers to the overall strategy or protocol that
the investigators choose to integrate the different components of
the study in a coherent and logical way, thereby, ensuring they
will effectively address the research problem and translate the
conceptual hypothesis into an operational one; it constitutes the
blueprint for the collection, measurement, and analysis of data.
Most problems in studies are due to poor design not poor analysis
When I came to practice I was looking for answers like everybody
else. For years I asked "what's the right answer?" Now I am
learning "What is the right question?“
Science is the holding of multiple working hypotheses
A study is only as good as its hypothesis
Admittedly, creative action can never be fully explained.
 Case report

Epidemiology

Descriptive Qualitative
Before-After
Quantitative
Case-Crossover

Case series
Study Case-Control
Designs Non-Experimental
Ecologic

Cohort study

Analytic Parallel
Cross-sectional
Blinded
RCT
Experimental Non Blinded

Non-RCT
15 Cross Over
 Types of Research Designs
 Cross-sectional
 Data are collected from the population on a single point or period only

 Retrospective
 Information usually relates to previous events

 Prevalence
 Cross-sectional studies collecting data relating to current situations or events to be occurring in a particular population

 Incidence
 Measure of the probability of occurrence of a given medical condition in a population within a specified period of
time.
 Observational
 Comprises of descriptive, often frequency analysis of cross-sectional data area

 Prospective
 Relating to future events

 Exploratory
 Commonly employ qualitative methodology and/or a mixture of approaches and methods

 Hypothesis generating
 Gives clues about important factors that may influence the behavior of individuals or success of service of explain a
phenomena
 Longitudinal
 Study follows up a sample of individuals or cases over a period of time
 Cohort
 A group of individuals is followed for a given period

 Experimental study
 To test a hypothesis (RCT)

 Quasi-experimental design
 If randomization is not possible (participants will necessarily be aware of the group they are in)

 Feasibility study
 Small scale studies
 Aims to assess aspects of the efficacy or practicalities of an intervention
 Involves small number of selected settings and often focus on specific features of the service that would be
deemed essential for its success
 Case-control study
 Hypothesis generating or hypothesis testing
 Identification of causative factors or explanatory variables that lead to a particular outcome

 Case-study
 Focuses on a single case or a small number of cases in order to examine phenomena of interest

 Pilot Studies
 Small studies that are designed to test the methods, procedures, instruments and documentation
for a larger study to check the project feasibility
 Triangulation
 Studies combine different approaches, method and / or data within a single research study to provide
different perspectives on a set of issues related to the study aims and objectives.
Antimicrobial Resistance in Pakistan
Quantitative Research
Quantitative methods emphasize objective
measurements and the statistical, mathematical, or
numerical analysis of data collected through polls,
questionnaires, and surveys, or by manipulating pre-
existing statistical data using computational
techniques.
High proportion of published papers
Quick and cheap
Methodological rigour
Robustness
FINER- Feasible, Interesting, Novel, Ethical, Relevant
Research Objectives
Views, beliefs and attitudes
Generalization
Quantify
Casual relationship between variables
Proving or disproving a hypothesis
 Components of a Study Protocol
Element Purpose
Research questions What questions will the study address?
Descriptive
Relational – Variables
Causal – Inference
Hypothesis
Null – No relationship
Alternative – Significant Diff
Significance (background) Why are these questions important?
Design How is the study structured?
Time frame
Epidemiologic approach
Subjects Who are the subjects and how will the be selected?
Selection criteria
Sampling design
Variables What measurements will be made/recorded?
Predictor variables
Confounding variables
Outcome variables
Statistical issues How large is the study and how will it be analyzed?
Hypotheses
Sample size
Analytic approach
Study Design
 Cross-sectional on a single occasion
 detects point prevalence; relative conditions, feasible; quick; economic;
allows study of several diseases / exposures, useful for estimation of the
population burden, health planning and priority setting of health problems.
 Longitudinal
 powerful, no temporal ambiguity; several outcomes could be studied at
the same time; suitable for incidence estimation, expensive; time-
consuming; inefficient for rare diseases
 Cross-sectional survey for retrospective data
 Retrospective Cohort
 Repeated Cross-sectional
 Experimental
 planned, stronger evidence, valid results, controlled environments, less
common
 Non-experimental descriptive designs
 hypothesis generating
Populations
Doctors
Practitioners
Pharmacist
Physicians
Patients
Pharmacy Clients
Public
Policy Makers
Others
Settings
 Hospital
 Pharmacy
 Community
 Clinics
 Others
Administration of survey
 Written
Mail survey
Oral survey
Self-administered survey (supervised & unsupervised)
Group-administered survey
Interviewer-administered survey
Drop-off survey
 Oral
 Electronic
Sampling Procedures
 A sample is “a smaller (but hopefully representative) collection
of units from a population used to determine truths about that
population”
 Why sample?
 Resources (time, money) and workload
 Gives results with known accuracy that can be calculated mathematically
 The sampling process comprises several stages:
 Defining the population of concern
 Specifying a sampling frame, a set of items or events possible to measure
 Specifying a sampling method for selecting items or events from the
frame
 Determining the sample size
 Implementing the sampling plan
 Sampling and data collecting
 Reviewing the sampling process
Probability Samples
 Probability Samples: each member of the population has a
known non-zero/equal probability of being selected without
bias
 Random sampling- purest form, use software, draw directly from
the columns.

 Systematic sampling- Nth name selection technique, simple, cost

effectiveness.

 Stratified sampling- make stratum, internally homogeneous,

superior, reduces sampling error, used when one or more of the
stratums in the population have a low incidence relative to the other
stratums

 Cluster Sample- internally heterogeneous, City blocks either

political or geographical, Housing units, Hospitals

 Multistage Sampling
Nonprobability Samples
 Nonprobability Samples: members are selected from the
population in some nonrandom manner
 Convenience sampling- easily accessible, exploratory research for
inexpensive approximation, preliminary research

 Judgment/Purposive sampling- extension of convenience

sampling, confident that the chosen sample is truly representative of
the entire population

 Quota sampling- equivalent of stratified sampling, identify

representative, then convenience or judgment sampling is used to
select the required number of subjects from each stratum.

 Snowball sampling- friend of friend, special nonprobability method

used when the desired sample characteristic is rare

 Opportunistic Sampling
Sample Size
 Sample Size (n) is the number of individuals in a group under study.
 The larger the sample size, the greater the precision.
 For statisticians, an n > 30 is usually sufficient for the Central Limit
Theorem to hold so that normal theory approximations can be used
for measures such as the standard error of the mean.
 However, this sample size (n = 30) is unrelated to the clinicians’
objective of detecting biologically significant effects, which
determines the specific sample size needed for a specific study.
 Choosing a sample size is to be done by combination of logistical
and pragmatic considerations which include:
(a) The number of subjects who can be recruited in the given time
period within available resources, and
(b) The final figure calculated must represent the minimum number of
subjects required to procure a reliable answer to the research question.
“What size sample do I need?”
The answer to this question is influenced by a number
of factors
purpose of the study
population size
the risk of selecting a “bad” sample
allowable sampling error
type of outcome i.e. proportions, mean, rates
Confidence intervals and their corresponding multiplyin
factors, based on the normal distribution
Criteria
 Level of Precision or Sampling Error
This is the range in which the true value of the population is estimated
to be (i.e. ±5%). Sampling error is reduced when the study size
is increased and vice versa.
 Confidence or Risk Level/Interval
This is based on Central Limit Theorem, if a 95% confidence level is
selected, 95 out of 100 samples will have the true population value.
The larger the sample size, the narrower is the confidence interval.
The multiplying factor 1.96 is used when calculating the 95 percent
confidence interval.
 Degree of variability in the attributes being measured
The more heterogeneous a population, the larger the sample size
required to obtain a given level of precision. Proportion of 50%
indicates a greater level of variability than either 20% or 80%. 0.5 is
often used in determining a more conservative sample size.
Strategies for Determining Sample
Size
 Studies that are too small may fail to detect important effects on the
outcomes of interest, or may estimate those effects too imprecisely.
 If the sample size is larger than what is needed, the study will
become cumbersome and ethically prohibitive. Apart from this, the
study will become expensive, time consuming and will have no
added advantages.
 If the study is well designed with a desired sample size then the
standard error will be less and the power and precision will be
good. All statistical procedures become valid in this context.
 Inferential statistics has two parts: estimation of population
parameter in prevalence/descriptive studies and testing of
hypothesis for cohort/case control/clinical trials.
 There are several approaches to determining the sample size.
Using a Census for Small Populations
 One approach is to use the entire population as the sample.
 Although cost considerations make this impossible for large
populations, a census is attractive for small populations
(e.g., 200 or less).
 A census eliminates sampling error and provides data on all
the individuals in the population.
 In addition, some costs such as questionnaire design and
developing the sampling frame are “fixed,” that is, they will
be the same for samples of 50 or 200.
 Finally, virtually the entire population would have to be
sampled in small populations to achieve a desirable level of
precision.
Using a Sample Size of a Similar Study
Another approach is to use the same sample size as
those of studies similar to the one you plan.
Without reviewing the procedures employed in these
studies you may run the risk of repeating errors that
were made in determining the sample size for another
study.
However, a review of the literature in your discipline
can provide guidance about “typical” sample sizes that
are used.
Using Published Tables
These sample sizes reflect the number of obtained
responses and not necessarily the number of surveys
mailed or interviews planned
This number is often increased to compensate for
nonresponse.
Sample sizes in Table 2 presume that the attributes
being measured are distributed normally or nearly so.
If this assumption cannot be met, then the entire
population may need to be surveyed.
Equation 5 was used to calculate the sample sizes in
Table 1 and Table 2
Using Formulas to Calculate a Sample
Size
Although tables can provide a useful guide for
determining the sample size, you may need to calculate
the necessary sample size for a different combination
of levels of precision, confidence, and variability.
The fourth approach to determining sample size is the
application of one of several formulas
Formula For Calculating A Sample For
Proportions
For populations that are large, Cochran developed the
Equation 1 to yield a representative sample for
proportions.
n0 = sample size
Z2 = 1 – α equals the desired confidence level, e.g., 95%
e = desired level of precision
p = estimated proportion of an attribute that is present in
the population
q = 1-p
Z = found in statistical tables which contain

the area under the normal curve

Finite Population Correct ion For
Proportions
If the population is small then the sample size can be
reduced slightly.
This is because a given sample size provides
proportionately more information for a small
population than for a large population.
The sample size (n0) can be adjusted using Equation 3.
A Simplified Formula For Proportions
This formula was used to calculate the sample sizes in
Tables 2 and 3 and is shown below.
A 95% confidence level and P =0.5 are assumed for
Equation 5.
Formula For Sample Size For The Mean
 The use of tables and formulas to determine sample size in the
above discussion employed proportions that assume a
dichotomous response for the attributes being measured.
 There are two methods to determine sample size for variables that
are polytomous or continuous.
One method is to combine responses into two categories and then use
a sample size based on proportion (Smith, 1983).
The second method is to use the formula for the sample size for the
mean.
 The formula of the sample size for the mean is similar to that of
the proportion, except for the measure of variability.
 The formula for the mean employs σ2 instead of (p x q), as
shown in Equation 7.
The disadvantage of the sample size based on the
mean is that a “good” estimate of the population
variance is necessary.
Often, an estimate is not available.
Furthermore, the sample size can vary widely from
one attribute to another because each is likely to have a
different variance.
Because of these problems, the sample size for the
proportion is frequently preferred.
 Power of study
 It is necessary to consider the probability of obtaining a statistically
significant result in a trial, and this probability is called the power of the study.
 Thus a power of 80 percent to detect a difference of a specified size means
that if the study were to be conducted repeatedly, a statistically significant
result would be obtained four times out of five if the true difference was really
of the specified size.
 When designing a study, the objective is to ensure that the study size is large
enough to give high power if the true effect of the intervention is large
enough to be of practical importance.
 The power of a study depends on:
 The value of the true difference between the study groups; in other words, the true
effect of the intervention (effect size). The greater the effect, the higher the power to
detect the effect as statistically significant for a study of a given size.
 The study size; The larger the study size, higher is the power.
 The probability level at which a difference will be regarded as ‘statistically significant’.
 The power also depends on whether a one sided or two sided significance
test is to be performed and on the underlying variability of the data.
 Level of significance
Level of significance is the probability of rejecting
null hypothesis when it is true.
This is the most important factor in determination
of sample size.
Therefore, the level of significance must be fixed
before the testing of hypothesis, estimation and
sample size calculation.
In a standard situation, the probability can be
taken as 0.05 or 0.01.
Of late, researchers have also been taking the
probability up to 0.2 (20%).
 Event rate
If an event studied occurs more commonly among
the study population, naturally, the power can be
expected to be higher.
Even though the expected event rate can be found
out from previous studies, there is a possibility that
the event rate can be estimated wrongly because
of the background of the referred study, like
differences in place, time, population etc.
If the overall event rate falls to an unexpectedly
low level, the sample size must be re-estimated by
adopting the new (currently observed) event rate.
 Effect of compliance
Compliance is another factor that directly affects
the sample size. So, it should be calculated
correctly. The compliance adjustment formula is as
follows: Adjusted sample size n1 per group equals
n is the original sample size
c1, c2 are the average
compliance rates per group.
Other Considerations
 First, the above approaches to determining sample size have assumed
that a simple random sample is the sampling design.
 More complex designs, e.g., stratified random samples, must take into
account the variances of subpopulations, strata, or clusters before an
estimate of the variability in the population as a whole can be made.
 Another consideration with sample size is the number needed for the
data analysis.
 If descriptive statistics are to be used, e.g., mean, frequencies, then
nearly any sample size will suffice.
 On the other hand, a good size sample, e.g., 200-500, is needed for
multiple regression, analysis of covariance, or log-linear analysis,
which might be performed for more rigorous state impact evaluations.
 The sample size should be appropriate for the analysis that is planned.
 In addition, an adjustment in the sample size may be needed to
accommodate a comparative analysis of subgroups
 Minimum of 100 elements is needed for each major group or subgroup in
the sample and for each minor subgroup, a sample of 20 to 50 elements is
necessary.
 Similarly, 30 to 200 elements are sufficient when the attribute is present
20 to 80 percent of the time (i.e., the distribution approaches normality).
 On the other hand, skewed distributions can result in serious departures
from normality even for moderate size samples.
 Then a larger sample or a census is required.
 Finally, the sample size formulas provide the number of responses that
need to be obtained.
 Many researchers commonly add 10% to the sample size to compensate
for persons that the researcher is unable to contact.
 The sample size also is often increased by 30% to compensate for
nonresponse.
 Thus, the number of mailed surveys or planned interviews can be
substantially larger than the number required for a desired level of
confidence and precision.
Response Rates
Achieve a high response rate
Investigate the non-response bias
About 20 – 30% usually return a questionnaire
Follow up techniques could bring it up to about 50%
Covering letter, pre paid envelope, repeat mailing,
telephone calls, incentives
Still, response rates under 60 – 70% challenge the
integrity of the random sample
How the survey is distributed can affect the quality of
sampling
 Survey Instrument
 Questionnaire is used to collect information to meet study objectives.
 An instrument consisting of series of questions and/or statements about attitude/opinion
designed to elicit responses.
 Characteristics
 Acceptable
 Attractive
 Relevant
 Reasonable in length.

 Development
 Develop by yourself
 Develop by others
 Modifications/Additions

 Variables
 Population descriptors
 Components of the attributes of interest
 Validity
The validity of an instrument is the extent to which it
actually measures what it is designed to measure.
Accuracy
Testing the instrument
Types
Face Validity
 Prima facia
 Uncover obvious problems
 Co-researcher , Other researcher, Blind, People from population
Criterion Validity
 Correlate with other measures
 Compared with established method e.g other instrument or
biomarker
 Construct Validity
 Corresponds to what is understood by a construct or concept
 Socioeconomic and ethnic groups
 Individual self perceived identity
 Minority, racism
 Health status, QOL, Satisfaction (facet-free vs facet-specific)
 Content Validity
 Covers all the relevant issues
 Preliminary fieldwork
 Qualitative enquiry
 Exploratory/Semi structured interview
 Focus groups
 Nominal Group Techniques & Delphi Methods
 Published Work
 External validity
 Generalizability
 Applied to individual beyond the sample
 Probability sampling technique, adequate sample size, good response rate
 Reliability
Reproducible
Internally consistent
Split-half method
Cronbach’s alpha
Test Retest
Factor analysis
No ambiguity of question wording
Less reliable if retrospective
Recall events- Diaries
Too many intervals- less reliable scale
Test-retest- respondents are less consistent
 Survey instrument can be reliable but not valid, it can not
be valid if it is unreliable
 Sample
Power calculation
Smaller the sample size larger the CI.
RaoSoft
PS software
Danielle’s/Daniel formula
Cochrane formula
Power analysis
 Statistical Procedures
Nominal ordinal
Ordinal Data
Continuous Data
Non parametric test
Parametric test
Correlation
Chai Square
Missing data
Regression
OR, RR
etc
 p-value
 While a small p-value can be interpreted as evidence for a real
difference between the groups, a larger ‘non-significant’ p-value must
not be interpreted as indicating that there is no difference.
 It merely indicates that there is insufficient evidence to reject the null
hypothesis, so that there may be no true difference between the
groups.
 Result is called statistically significant if it has been predicted as
unlikely to have occurred by chance alone.
 Null hypothesis refers to a general or default position: that there is no
relationship between two measured phenomena, or that a potential
medical treatment has no effect.
 In the significance testing approach of Ronald Fisher, a null
hypothesis is potentially rejected when the p-value turns out to be
less than a certain significance level, often 0.05
 If your values is 0.005, there is only a 0.5% chance of that result
occurring due to chance.
 One sided and Two sided tests
If it is accepted that the null hypothesis is false,
that means alternate hypothesis is true.
If the claim is about superiority of a new drug, this is a
one-sided alternative.
If the claim is not of superiority or inferiority but only that
they are different, the alternate hypothesis is two sided.
Wrongly rejecting a true null hypothesis is called
type I error and vice versa.
 Qualitative Methodology

 Gained prominence in recent years

 Superiority in answering questions
 “How” & “Why”
 Self Perception
 Receptiveness
 Respondents’ view point
 Leave the researchers’ own view point behind
 Exploratory in nature
 Interactive approach
 Identify all possible explanations of phenomena of interest
 Identification of barriers
 A hypothesis generating method
 Understanding the perceptions, actions and emotions
 Natural context
 Learning from people
 Hidden dimensions
 Controversial relationship
 Population, Settings & Samples
Representativeness
15-50 (8-226)
No new influence emerge (saturation)
Response rate- incentive, pre paid telephone,
honorarium
Respondents’ own home
Independent setting
Face to face
Ensure privacy
Telephone
 Development of Interview Guide
Topic Headings
Content
Open questions
Respondent-led
Additional questions for clarification of views
Neutral probing
Question order; open-specific
Previous work
 Types of Interviews

 Structured
 Also known as a standardized interview or a researcher-administered interview.
 The interview is presented with exactly the same questions in the same order.
 The answers can be reliably aggregated.
 The comparisons can be made with confidence between sample subgroups or
between different survey periods.
 Unstructured
 Also known as non-directive interview.
 Questions are not prearranged.
 These non-directive interviews are considered to be the opposite of a structured
interview.
 Semi-structured
 Loose structure
 Open ended questions
 Flexible agenda
 Serves as check list
 Liberty to pursue questions in depth
 Probing approach
 Opportunity to clarify ambiguous replies
 Sampling procedures

 Purposive samples
 Relevant to study
 Most informative
 Ensure variability

 Representative samples
 Quota from all centers

 Convenience samples
 Accessible
 Unrepresentative

 Theoretical samples
 Data Collection
Interviews
Focus Groups
Observations
Written descriptions of observations such as field notes,
and reflections, ideas and conjectures
Transcripts
Narratives
Audiotaped
Transcribed verbatim
Contextual data
Paraphrasing of the response
 Data processing & Analysis
Organize into themes
Analytical procedures
Computers
Reporting of results
Validity
 Communicative validation
 Researcher returns to the field to collect additional data to verify or further develop their findings.
 Argumentative validation
 An attempt is made to use the dataset to argue a contradictory viewpoint.
 Cumulative validation
 The researcher uses the literature to demonstrate how their findings are consistent with existing knowledge of the
subject.

Reliability
 Maintaining meticulous records of the interviews and observations
through documenting the process of analysis in detail.
Generalizability
 Theories
Phenomenology
Cyrandeal theory
Discourse analysis
 Mix Methodology
 The integration of both Qualitative and Quantitative methodologies
broadens the perspective of a research and unravel the complexity of many
different factors that influence health.

There are two designs used for mix methodology

 Concurrent design
 Compare and validate data collected from both methods.
 Usually the same individuals provide both the qualitative and quantitative data.
 Sequential design
 The first one informs the analysis of the second.
 A preliminary qualitative study may precede a large quantitative study.
 Qualitative method provides complementary assistance to the quantitative
research.

For Example: Findings from in-depth interviews and focus group discussion could be used to
develop the content of questionnaire (to include important topics and
improves the effectiveness of the survey.
Epidemiological studies
Non-interventional
AKA observational studies
Researchers do not manipulate situations/objects
Only describes or analyses situations/objects

Interventional
Participants are assigned to receive one or more
interventions (or no intervention) so that researchers can
evaluate the effects of the interventions on biomedical or
health-related outcomes.
Non-interventional studies
Descriptive studies
Case series
Community diagnosis or need assessment
Epidemiological description of disease occurrence
Descriptive cross-sectional studies
Ecological descriptive studies
Analytical studies
Case-control studies
Cohort studies
 Prospective cohort studies
 Retrospective cohort studies
Analytical cross-sectional studies
Descriptive studies
When an epidemiological study is not structured
formally as an analytical or experimental study i.e.
when it is not aimed specifically to test a hypothesis, it
is called a descriptive study

The wealth of material obtained in most descriptive

studies allows the generation of hypotheses, which can
then be tested by analytical or experimental designs
Case series
based on reports of a series of cases of a specific
condition, or a series of treated cases, with no
specifically allocated control group

These represent the numerator of disease occurrence, and

should not be used to estimate risks
Community diagnosis or needs assessment

This kind of study entails collection of data on

 existing health problems
 Programs
 Achievements
 Constraints
 social stratification
 leadership patterns
 focal points of resistance or high prevalence, or groups at highest
risk.
Its purpose is to identify existing needs and to provide
baseline data for the design of further studies or action
 Epidemiological description of the disease occurrence
 This common use of the descriptive approach involves the collection
of data on the occurrence and distribution of disease in populations
according to specific characteristics of individuals
 Age
 Sex
 Education
 Smoking habits
 Religion
 Occupation
 social class
 Marital status
 Health status
 Personality
 Place (rural/urban, local, national, international)
 Time (epidemic, seasonal, cyclic, secular)
 Descriptive cross-sectional studies
 KAP population/community surveys

 Entail the collection of data on, as the term implies, a cross-section of

the population, which may comprise the whole population or a
proportion (sample) of it

 Many cross-sectional studies do not aim at testing a hypothesis about

an association

 Provide a prevalence rate at a particular point in time → point

prevalence

 Or over a period of time → period prevalence

Ecological descriptive studies
When the unit of observation is an aggregate (e.g.
family, clan or school)

Or an ecological unit (a village, town or country) the

study becomes an ecological descriptive study
Analytical strategies of epidemiological studies

Observational studies, where establishing a relationship

between a ‘risk factor’ (etiological agent) and an
outcome (disease) is the primary goal, are termed
analytical

 hypothesis testing is the primary tool of inference

Case-control studies
Simplest and most commonly used analytical strategy in
epidemiology

Designed primarily to establish the causes of diseases by

investigating associations between exposure to a risk
factor and the occurrence of disease

Design is relatively simple, except that it is backward-

looking (retrospective) based on the exposure histories of
cases and controls
one investigates an association by contrasting the
exposure of a series of cases of the specified disease with
the exposure pattern of carefully selected control groups
free from that particular disease
Advantages Disadvantages
Relatively quick and inexpensive great risk of bias in the selection of cases
and
Controls
feasible for rare disease Prone to recall bias (it may be very difficult
or impossible to obtain information on
exposure if the recall period is long)
Requires a smaller sample size than cohort Cannot measure incidence, only relative
study risk is measured
Little problem with attrition, as when
follow-up requires periodic investigations
and some subjects refuse to continue to
cooperate
Cohort studies
 Prospective cohort study
 The common strategy of cohort studies is to start with a reference
population (or a representative sample thereof)

 Some of whom have certain characteristics or attributes relevant

to the study (exposed group), with others who do not have those
characteristics (unexposed group)

 Both groups should, at the outset of the study, be free from the
condition or conditions under consideration

 Both groups are then observed over a specified period to find out
the risk each group has of developing the condition(s) of interest
Design of a cohort (prospective) study
Example of Prospective cohort study design
Design features
 Selection of cohort
 a community cohort of specific age and sex
 an exposure cohort, e.g. radiologists, smokers, users of oral
contraceptives
 a birth cohort, e.g. school entrants
 an occupational cohort, e.g. miners, military personnel
 a diagnosed or treated cohort, e.g. cases treated with radiotherapy,
surgery, hormonal treatment

The usual procedure is to locate or identify the cohort, which

may be a total population in an area or sample thereof
Data to be collected
 data on the exposure of interest to the study hypotheses
 data on the outcome of interest to the study hypotheses
 characteristics of the cohort that might confound the association
under study
Methods of data collection
 Several methods are used to obtain the above data, which should
be on a longitudinal basis.
 These methods include
 Interview surveys with follow-up procedures
 Medical records monitored over time
 Medical examinations and laboratory testing
Measures of frequency
Cumulative incidence

Incidence density (Person-year approach)

  In September 2009, 3000 Birmingham university freshman enrolled in a study
of health behaviors that aimed to track them over 4 years of college. Data from
September 2012 are shown below
 What is the cumulative incidence of obesity over 4 years?

No. of new No of
cases students
“at risk” at
the
begining
Admitted to the emergency room for an alcohol-related injury 400 3500
Unintended pregnancy? 252 2000
Diagnosed with depression? 310 3000
Obese? 543 2100
Person-time approach/incidence density
This approach is an improvement over the conventional
measure of incidence, because it takes into consideration
both the number observed and the duration of
observation for each individual

If 30 individuals were observed as follows: 10 for two

years, 5 for three years, and 15 for four years, they would
contribute (10x2)+(5x3)+(15x4) = 95 person-years of
observation, which would become the denominator. The
numerator is the number of new cases observed in these
groups over the specified period of time………
Advantages Disadvantages
Allow calculation of incidence rates, Expensive
relative risk , attributable risk etc.
If a probability sample is taken from the long-term and sometimes are not always
reference population, it is possible to feasible
generalize from the sample to the reference
population with a known degree of
Precision
capable of identifying other diseases that Sample sizes required for cohort studies are
may be related to the same risk factor extremely large
The most serious problem is that of attrition
Historical/retrospective cohort study
Analytical cross-sectional study
Investigator measures exposure and disease
simultaneously in a representative sample of the
population
By taking a representative sample, it is possible to
generalize the results obtained in the sample for the
population as a whole
Cross-sectional studies measure the association between
the exposure variable and existing disease (prevalence)
Unlike cohort studies, which measure the rate of
developing disease (incidence)
Design of a cross-sectional study
Experimental study designs
An experiment is the best epidemiological study
design to prove causation

Viewed as a final or definitive step in the research

process

A mechanism for confirming or rejecting the validity

of ideas, assumptions, postulates and hypotheses about
the behavior of objects, or effects upon them which
result from interventions under defined sets of
conditions
Investigator has control of the subjects, the
intervention, outcome measurements, and sets the
conditions under which the experiment is conducted

In particular, the experimenter determines who will be

exposed to the intervention and who will not

This selection is done in such a way that the

comparison of outcome measure between the exposed
and unexposed groups is as free of bias as possible
General experimental study design
Randomized Controlled Clinical Trial
(RCT)
 A prospective, analytical, experimental study using primary data

generated in the clinical environment.

 Individuals similar at the beginning are randomly allocated to two or

more treatment groups and the outcomes the groups are compared after
sufficient follow-up time.
 Properly executed, the RCT is the strongest evidence of the clinical

efficacy of preventive and therapeutic procedures in the clinical setting.

 The most commonly encountered experiment in health science research, and the

research strategy by which evidence of effectiveness is measured, is the

randomized, controlled, double-blind clinical trial, commonly known as an RCT
Clinical Trials
Prospective or behavioral research studies on human

subjects with the aim to answer specific questions

regarding biomedical or behavioral interventions
(vaccines, drugs, treatments, novel modalities of known
interventions)
Research studies to examine how well medical

approaches work in people

Clinical trials may be done for various purposes
Some of the common types of clinical trial (according
to purpose) are:

prophylactic trials, e.g. immunization, contraception

therapeutic trials, e.g. drug treatment, surgical procedure
safety trials, e.g. side-effects of oral contraceptives and
injectable
Risk-factor trials, e.g. withdrawing the agent (e.g.
smoking) through cessation
Therapeutic trials may be conducted to test efficacy
does a therapeutic agent work in an ideal, controlled
situation?

or to test effectiveness (e.g. after having established

efficacy, if the therapy is introduced to the population
at large, will it be effective when having to deal with
other co-interventions, confounding, contamination,
etc.?
 The intervention in a clinical trial may include
drugs for prevention, treatment or palliation
clinical devices, such as intrauterine devices
surgical procedures, rehabilitation procedures
medical counselling
diet, exercise, change of other lifestyle habits
risk factors
communication approaches, e.g. face-to-face communication
versus pamphlets
different categories of health personnel, e.g. doctors versus
nurses
treatment regimens, e.g. once-a-day dispensation versus three
times a day
Design of a Clinical Trial
 A good clinical trial design controls or minimizes known or suspected

sources of bias and other errors so that clinical treatment/device

effectiveness may be assessed clearly and objectively.

 Error is the result of our inability to accurately measure a variable.

 Type I – detecting an effect when it is not present (false positives)

 Type II – failing to detect the effect when it is present (false negatives)

Bias
 Bias affects results when any characteristic of the investigator,

study population, or study conduct interferes in a systematic way

with the ability to measure a variable accurately
 Selection bias – diff in base line characteristics of groups

 Performance bias – diff in care that is provided

 Detection bias – how outcomes are determined

 Attrition bias – withdrawals

 Reporting bias – reported and un-reported findings

Minimising bias and confounding
Variation - measurements are nearly always subject to random
variation. Minimise error by ensuring adequate sample size and
using statistical analysis

Bias - caused by systematic variation/error in selecting patients,

measuring outcomes, analysing data – take extra care

Confounding - factors that affect the interpretation of outcomes

e.g people who carry matches are more likely to develop lung
cancer, but smoking is the confounding factor – so measure likely
confounders too
Ethical issues – the wider aspects

what information to give before seeking consent?

deviation from normal clinical practice?
what full burden will be imposed on participants?
what risks will participants/others be exposed to?
what benefit might participants or others receive?
how might society/future patients benefit in time?
might publication reveal patients’ identities?
 Exactly what are you planning to do?

PICO

P - who are the patients or what’s the problem?

I - what is the intervention or exposure?
C – what is the comparison group?
O - what is the outcome or endpoint?
More on PICO
Patients
 disease or condition
 stage, severity
 demographic characteristics (age, gender, etc.)
Intervention
 type of intervention or exposure
 dose, duration, timing, route, etc.
Comparison
 risk or treatment
 placebo or other active treatment
Outcome
 frequency, risk, benefit, harm
 dichotomous or continuous
 type: mortality, morbidity, quality of life, etc.
 Selection of Variables
 Clinical study involve two types of variables:

 outcome variables and

 influencing variables.

 Outcome variables define and answer the research question and should have

direct impact on the claims for the medication.

 These variables, also known as response, endpoint, or dependent variables,

should be directly observable, objectively determined measures with minimal

bias and error.

 They should be directly related to biological effects of the clinical condition

 Selection of Variables
 Influencing variables, also known as baseline variables, prognostic

factors, confounding factors, or independent variables, are any aspect

of the study that can affect the outcome variables (increase or

decrease), or can affect the relationship between treatment and

outcome.

 Imbalances in comparison or treatment groups in influencing

variables at baseline can lead to false conclusions by improperly

attributing an effect observed in the outcome variable to an

intervention when it was merely due to the imbalance.

Study Inclusion and Exclusion Criteria
 Inclusion criteria are characteristics that the

prospective subjects must have if they are to be

included in the study. Exclusion criteria are those

characteristics that disqualify prospective subjects

from inclusion in the study.

Inclusion/Exclusion Criteria
 Inclusion/exclusion criteria should include an

assessment of prognostic factors for the outcome

variable(s), since one or more of these variables may

influence the effectiveness of a treatment.

 For example, gender may be a prognostic factor for a

particular disease process.

Inclusion/Exclusion Criteria
 Inclusion and exclusion criteria may include factors such as

age, gender, race, ethnicity, type and stage of disease, the

subject’s previous treatment history, and the presence or
absence (as in the case of the “healthy” or “control”
subject) of other medical, psychosocial, or emotional
conditions. Healthy, or control, subjects may be defined as
those individuals who are free of certain specified attributes
of non-health
Inclusion/Exclusion Criteria
 It seems reasonable to assess what role, if any, that gender

might play in treatment evaluation and then determine

inclusion/exclusion criteria, other design, and analytical
considerations accordingly.
 Consideration should also be given to: patient age;

concomitant disease, therapy or condition (at both baseline

and subsequent follow-up times); severity of disease; and
others.
 Poorly Justified Reasons for Exclusion:

 Any criterion has a direct bearing on condition/intervention/results.

 Strongly Justified Reasons for Exclusion:

 Unable to provide informed consent

 Harmful effects of Placebo or intervention

 Lack of equipoise (intervention harmful)

 Effect of intervention difficult to interpret

 Potentially Justified Reasons for Exclusion

 Individual may not adhere

 Individual may not complete follow up

 Example

 An example of inclusion criteria for a study of chemotherapy of breast

cancer subjects might be postmenopausal women between the ages of 45

and 75 who have been diagnosed with Stage II breast cancer.

 An exclusion criterion for this study might be abnormal renal function

tests, if the combination of study drugs includes one or more that is

nephrotoxic. In this case it would be required to specify which tests of

renal function are to be performed to evaluate renal function and the

threshold values that would disqualify the prospective subject (e.g.,

serum creatinine above 1.9 mg/dl).

 Control population
 Every clinical trial intended to evaluate an intervention is

comparative, and a control exists either implicitly or explicitly.

 The safety and effectiveness of a treatment is evaluated through

the comparison of differences in the outcomes (or diagnosis)

between the treated patients (the group on whom the

device/treatment was used) and the control patients (the group

on whom another intervention, including no intervention, was

used).
A scientifically valid control population should be
comparable to the study population in important
patient characteristics and prognostic factors, i.e., it
should be as alike as possible except for the treatment
Randomization
 Assurance that subject populations are similar in test and control

groups is best attained by randomly dividing a single sample

population into groups that receive the test or control treatments.
 Randomization avoids systematic differences between groups

with respect to known or unknown baseline variables that could

affect outcome.
 Randomization also provides a sound basis for statistical inference
Blinding
 The groups should not only be similar at baseline, but should be

treated and observed similarly during the trial, except for

receiving the test and control drug.

 Clinical trials are often “double-blind” (or “double-masked),

meaning that both subjects and investigators, as well as sponsor

or investigator staff involved in the treatment or clinical

evaluation of subjects, are unaware of each subject's assigned

treatment.
Blinding is intended to minimize the potential biases

resulting from differences in management, treatment,

or assessment of patients, or interpretation of results

that could arise as a result of subject or investigator

knowledge of the assigned treatment

Phases
Usually, clinical trials of new therapies or devices pass
through four phases

Phase I
Phase II
Phase III
Phase IV
Phase I clinical trial
This first phase in humans is preceded by
considerable research

Including pharmacological and toxicological studies in

experimental animals

To establish that the new agent is effective and may be

suitable for human use,

And to estimate roughly the dose to be used in man

Phase I trials include studies of volunteers who
receive, initially, a fraction of what the anticipated
dose is likely to be
And are monitored for effects on body functions, such
as
Hepatic
Cardiovascular
Renal
Gastroinstestinal
Endocrinal functions
The metabolism of the drug may also be investigated
at this stage

These studies are normally done on volunteers, who

are usually institutionalized, and occupy what are
called ‘research beds’

Require close supervision

short duration
Phase II clinical trial
Also carried out on volunteers selected according to
strict criteria

Purpose of Phase II is
to assess the effectiveness of the drug or device
to determine the appropriate dosage
to investigate its safety

Further information on the pharmacology, especially

the dose-response relationship of the drug, is collected
Phase III clinical trials
This is the classical phase
The one usually referred to as a ‘clinical trial’ and
reported in health research journals
Performed on patients, who should consent to being in
a clinical trial
Strict criteria for inclusion in and exclusion
purpose of this phase is
to assess the effectiveness
to assess safety in continue use of the drug or device in a
larger and more heterogeneous population than in Phase
II
It includes more detailed studies and monitoring than
those given in a normal service situation

Usually carried out on hospital inpatients, but may be

performed on outpatients with intense monitoring
follow-up

It requires superior clinical and epidemiological skills,

in addition to the required laboratory technology
Results from these trials are used by regulatory
agencies to evaluate whether a new product or device
should be licensed for general public use
Phase IV
To identify and evaluate the long-term effects of new
drugs and treatments over a lengthy period for a
greater number of patients.

Phase IV research takes place after the FDA approves

the marketing of a new drug

Through Phase IV clinical studies, new drugs can be

tested continuously to uncover more information about
efficacy, safety and side effects after being approved
for marketing
Factors that affect the design of and analysis of trial (phase III)
 The agent, treatment or experimental factor
 Conditions to be treated
 The target population
 Ethical issues
 Outcomes to be measured
 Side-effects
 Study instruments
 Blinding
 Stopping rule
 Plans for the analysis
 Selective attrition
 Methods for ensuring the integrity of the data
 The choice of design
 Time required
Community intervention trials
CITs are usually carried out in hospitals or clinics, and
are usually directed at a patient group with specific
health conditions

a test of whether the introduction of iron-fortified salt

in the community would reduce the incidence of
anemia in the community

The major difference from the RCT is that the

randomization is done on communities rather than
individuals
Conclusion

“Research is for the people,

institutions, and activities whose
primary purpose is to generate high
quality knowledge that can be used
to promote, restore, and or maintain
the health status of populations“