Download How to Develop a Research Protocol ?

How to Develop a
Research
Protocol
?
By
Dr.Shaik Shaffi Ahamed
Asst. Professor
Dept. of Family & Community
Medicine
What is Research?
A systematic investigation, involving the
collection of information (data), to solve a
problem or contribute to knowledge about a
theory or practice
Relies on methods and principles that will
produce credible and verifiable results
Research helps provide scientific
understanding and solves practical problems
MODEL OF THE RELATIONSHIP BETWEEN DIFFERENT TYPES OF RESEARCH
Epidemiological
Biomedical
Statistical
Behavioural
Social &
economic
Health
systems
research
Policy making,
planning,
Management
evaluation
BCG vaccination is not effective. Why ?
(a) BCG vaccination coverage is good but not effective
(i)
Immuno-microbiological factors
Poor nutrition (low protein intake)
Poor immune reaction (race-specific)
Tubercle bacillus strains
Atypical mycobacterial infection
---- BIOMEDICAL PROBLEMS
(ii) Technical factors
Quality of BCG vaccine
---- BIOMEDICAL PROBLEMS
(iii) Operational factors
Storage and transport of vaccine
Handling of vaccine after reconstitution
Technique of vaccination
Logistic support (supply of vaccine, vaccinating equipment)
----- HEALTH SYSTEMS RESEARCH
(b) BCG vaccination coverage is poor
(i) Operational factors
Coverage and efficiency of local
health services
---- HEALTH SYSTEMS RESEARCH
(ii) Human factors
Indifference of population toward
immunization
Fear of reaction and low level of
confidence in BCG
Decline in concern about tuberculosis
----- BEHAVIOURAL PROBLEMS
Clinical issues and questions in the
practice of medicine
Issue
Question
Normality/abnormality
Is a person sick or well? What abnormalities are
associated with having a disease ?
Diagnosis
How accurate are diagnostic tests or strategies used to
find a disease ?
Frequency How often does a disease occur ?
Risk What factors are associated with an increased likelihood
of
disease ?
Prognosis What are the consequences of having a
disease ?
Treatment How does treatment change the
future course of a disease ?
Prevention Does intervention on people
without disease keep disease from
arising?
Does early detection and treatment
improve the course of disease ?
Cause What conditions result in disease ?
What are the pathogenetic mechanisms of
disease ?
Introduction
Research is critical in medicine because it
leads to new discoveries and can change
peoples’ lives by improving health and well being
All research starts with an idea or question
based on personal experiences
Research shapes the world we live in by
continually questioning and testing human
knowledge and understanding
Introduction
There are many ways in which humans
acquire knowledge and gather information
in order to solve problems
Many questions are answered and
problems solved based on inherited
customs, traditions and experiences
More complex questions may be
answered through a process of logical
reasoning
What is Reasoning?
Inductive - the process of
developing generalization from
specific observations
Deductive - the process of
developing specific predictions
from general principles
Scientific Approach
The most sophisticated method of acquiring
knowledge that has been developed
Combines important features of induction
and deduction together with other methods
to create a system of acquiring knowledge
Generally more reliable than other methods
Approach used most often to perform
medical research
Scientific research
is an integrated approach of deduction and induction
Deduction
Hypothesis generation
Testing the hypothesis
Induction
(Inference on hypothesis)
generalization of the results
known truths
In sample
population
Universe
OBJECTIVE OF MEDICAL
RESEARCH
Increase understanding of casual
association , both etiologic agents and risk
factors to disease.
 To improve methods of diagnosis
 To optimize therapy and management of
the sick.

Execution of Research:
Conceptualizing the problem:
Need
Background & Ratinale
Formulating the Objectives
Generating hypotheses
Testing hypotheses
Designing the Approach
Research design
Methods and Materials
Target population
Study population
Methods of collection
Analysis and Interpretation of results
Choosing a topic
 Should be interesting – to investigator, funding agency,
journal editors, consumers (colleagues, public, medical
community), etc.
 Relevance- add new information to the scientific world
 Simple and manageable in scope (feasibility in terms of
money, time, manpower)
 Expected results likely to alter clinical or health policy
decisions in future
 New interventions chosen for trial should have some
supportive evidence to its superiority over the conventional
treatment in one way or other
Steps in conduct of research
designing, planning and execution
The first and foremost is
formulating a research question,
the most challenging part
Identify the Question
Good or poor research is defined by the
question being asked
The question should be well understood, and
the problem well defined
Selecting a question should not be rushed
If the question is hurried, proceeding in an
orderly fashion may be difficult and may
produce unreliable results
Identify the Question
Sources for identifying the question or
problem:
Personal experiences
Literature review
Theories
Ideas from others
Identify the Question
Criteria for evaluating the question:
Significance
Practicality
Feasibility
Interest to researcher
RESEARCH QUESTION
IT SHOULD BE A SINGLE SENTENCE IN
THE FORM OF A QUESTION.
IT SHOULD BE CLEAR UNAMBIGUOUS
AND SPECIFIC
RESEARCH QUESTION

IS DRUG “A” BETTER THAN DRUG “B” IN
THE MANAGEMENT OF HEPATIC FAILURE
IN PATIENTS WITH CIRROSIS?

IS ALCOHOLISM RELATED TO THE
DEVELOPMENT OF CIRROSIS LIVER?
Why Research question?
to communicate &
convince
the need and nature of
the study
in a simple but single
sentence





Scientific community
Health professionals
Funding agency
Journal editors
Administrators, health
policy makers
 Lay public
 Ethical committee
Fully refined RQ
 The fully refined research question should
indicate the objective of the study,
 specify the major outcome and predictive
variables
 the setting and the intended study subjects.
 The implied biological rationale
and study design should be explicit in the
research question.
Refining Research Question
Fully refined Research Question
“Do beta-carotenoids protect against human cancer?”
Is the risk of developing lung cancer low among cohorts
with high beta-carotenoid dietary intake, compared to
cohorts with low beta-carotenoid dietary intake among
male smokers residing in Riyadh?
Refining Research Question
Fully refined Research Question
“Do beta-carotenoids protect against human cancer?”
Is the risk of developing lung cancer low among cohorts
with high beta-carotenoid dietary intake, compared to
cohorts with low beta-carotenoid dietary intake among
male smokers residing in Riyadh?
Cohort Study
Refining Research Question
Fully refined Research Questions
• Is there an association between serum retinoic acid
level and development of lung cancer among male
smokers residing in Riyadh?
- A case-control study
• Does administration of beta-carotenoid (specify dose,
route and duration) reduce the risk of developing lung
cancer among male smokers residing in Riyadh?
– A randomized placebo controlled trial.
REVIEW OF LITERATURE

SHOULD GO BACK ATLEAST FIVE YEARS

SHOULD BE STRUCTURED BY CONTENT
OR GEOGRAPHIC REGION

SHOULD BE A CIRITICAL REVIEW THAT
ASSESSES THE STRENGTHS AND
WEAKNESS OF THE DESIGN USED IN THE
STUDIES REVIEWED.
Literature Review
Determine what published data
suggests about the question or
problem
Clarify the value of the question
Clarify what is already known
Provide sources for reference
Literature Review
Primary Sources:
Journal articles, books, abstracts
Written by the person(s) who conducted
the research
Secondary Sources:
Review articles that summarize research
Written by someone other than primary
investigator
OBJECTIVE

SHOULD BE GENERAL AND SPECIFIC
gen: prevalence of hiv
 specific: prevalence of hiv in unmarried adults

SHOULD CONTAIN WHAT YOU EXPECT TO
DO
 SHOULD BE ONLY ONE PRIMARY
SPECIFIC OBJECTIVE.
 IF YOU HAVE SECONDARY OBJECTIVE - IT
WILL BE DIFFICULT TO DESIGN A STUDY
TO ANSWER MORE THAN ONE
OBJECTIVES AT ONCE.

Frame a Hypothesis
The hypothesis is a statement that describes the
results the researcher expects
It examines relationships or differences
The null hypothesis is a statement that
expects no relationships or differences to
exist
A study should be designed to test the
hypothesis or null hypothesis
Frame a Hypothesis
The nature of the hypothesis will
determine:
Sample group for study
Measuring instruments
Study design
Procedures
Statistical techniques
Refining Research Question
1. What is the Conceptual hypothesis (CH)
After deciding on the topic
- gather knowledge from all possible sources
- arrive at a meaningful conceptual
CH:hypothesis
“Beta-carotenoids have protective role
against development of human cancer”
Can you think of a Research Question (RQ) ?
RQ: “Do beta-carotenoids protect against
human cancer?”
Refining Research Question
2. Pick up an operational hypothesis
CH usually contains many theoretical principles
“Do beta-carotenoids protect against human cancer?”
OH1: Dietary deficiency of beta-carotenoid is associated
with increased risk of lung cancer.
OH2: Exposure to or treatment with beta-carotenoid can
decrease the risk of lung cancer
OH3: Low levels of serum retinoic acid are associated with
increased risk of lung cancer.
Refining Research Question
3. Identify the study variables
“Do beta-carotenoids protect against human cancer?”
Exposure variables
 Diet habits (beta-carotenoids)
 Serum retinoic acid level
 Receptors of retinoic acid
 Markers of retinoic acid
Outcome variables
 All cancers (cancer registry)
 Organ Specific cancers e.g. lung cancer
 Cell atypia e.g. sputum cytology
Refining Research Question
4. Specify the nature of comparisons
“Do beta-carotenoids protect against human cancer?”
E.g. Strength of Association
Comparison of risk between 2 groups
Case-control study:
Odds ratio
Cohort study, RCT:
Relative risk
Refining Research Question
5. What is the Study design?
“Do beta-carotenoids protect against human cancer?”
OH1: Dietary deficiency of beta-carotenoid is associated
with increased risk of lung cancer.
Cohort or Case-control or RCT
OH2: Exposure to or treatment with beta-carotenoid can
decrease the risk of lung cancer
Cohort or RCT
OH3: Low levels of serum retinoic acid are associated with
increased risk of lung cancer.
Cohort or Case-control
Develop the Study Design
A study design is the researcher’s overall plan
to obtain the answer(s) to the question being
asked and the hypothesis being tested
It spells out strategies to develop information
that is accurate, objective and meaningful
It explains methods that will be used to collect
and analyze data
It includes time frame to conduct study
Research Designs
Purpose
Study Design
To determine frequency &
burden of a disease
* Cross sectional survey
(Prevalence)
* Cohort study (Incidence)
To identify the risk factors
* Cohort study
* Case-Control study
To determine prognosis of a * Cohort study
disease
To determine efficacy/
effectiveness of new
treatment
* Clinical trials
To evaluate community
programs
* Evaluation
*Community intervention
Methodology
--Study subjects
--Selection of study subjects
--Sampling method
--Criteria for inclusion/exclusion
--Sample size
--Study & Outcome variables
--Measurement of study & outcome
variables (data form, Questionnaire)
--Place of study (Community, OP, IP, Case
records, College., )
Feasibility
 Availability of resources (funding)
 Infrastructure
 Technical expertise (subject expert, methodological expert
& statistical expert)
 No extra stress to the patients or existing system
(in terms of money, manpower or other resources)
 No Ethical violation
Feasibility- Study subjects

What is the estimated sample size?

Who is the study subject (case
definition)?

Selection criteria (inclusion & exclusion)

How they are sampled ? (sampling)

Time span for meeting the sample size
Can we meet the sample size?

Estimated sample size (based on the research
hypothesis, outcome variable)

Estimated subjects likely to be available for
recruitment

Estimated subjects likely to refuse

Estimated subjects likely to be lost to follow up
• Extend of the problem in target population
• Knowledge of biological behavior of disease &
study subjects
• Pilot study required ?
Analyzing the Data
Upon completion of the study, data should
be analyzed
List out the appropriate statistical tests
based on the type of data
Conclusions
Developing, conducting and
communicating a research protocol is a
sophisticated and time-consuming
process
It is important to understand the steps in
developing a research protocol in order to
perform an appropriate study and obtain
reliable results
Thank You