3. Research Planning
Identify knowledge gap to select and to justify the
research problem.
Transform the problem into clear researchable aims
and research question (formulate the research
question).
Search for existing information.
Focus the research question
Design the study.
4. The Research Process
You begin with an observation
Generate explanations or “Theories”
Make Predictions “Hypothesis”
Data processing
Identify Variables
Collect data
Data Analysis
Data Presentation
5. Sources of Research Questions
• Health development and promotion depends on
researchers asking the „right‟ questions and identifying
solvable problems.
• Sources:
• Expertise of researchers (professional background):
• Theoretical knowledge
• Practical experience
• Culmination of intensive preliminary observations
• Reading in the library
• Discussion
6. Where do questions come from?
1. From patient-centered questions in routine clinical
practice:
Diagnosis
Etiology
Prognosis
Treatment or prevention
2. From new treatment or diagnostic tests.
3. From physician and patient experiences.
7. Defining a good question
• Importance
• Interest
• Motivation
• Innovation
• Ethical considerations
• Answerability
• Type III error: Asking the wrong question
• Type IV error: Asking a question not worth answering
8. Formulate an answerable question
1. What is the question?
Variance questions: Focuses on difference and correlation
(Quantitative, Clinical)
Process questions: Focuses on how and why things happen
(Qualitative)
2. What is the problem, intervention, comparator, and
outcome?
3. What is the best feasible study type?
9. What is the question?
TASK
Think of some of the clinical questions you have asked
recently.
Write one of these questions down in your own
words…
10. Example
“What is the role of antibiotics in people who get recurrent
skin infections”
11. Example
“What is the role of antibiotics in people who get
recurrent skin infections”
What is wrong with the way we expressed our
question?
Too board, not specific.
Not clear what information is needed.
Unanswerable.
12. • To formulate an answerable question we first need to
think . . . . . . .
• What do we really want to know?
• What type of question are we asking?
• Does our question concern background or foreground information?
13. Background questions
• Questions concerning basic biological processes.
• e.g. What is . . . . .? How does . . . . .?
• Best information source: regularly updated electronic
textbooks
14. Foreground questions
• Generated in the clinical setting
• Specific and relevant to clinical decision making
• Observational? Frequency? Diagnosis? Aetiology?
Prognosis? Or Intervention?
15. “What is the role of antibiotics in people who get recurrent
skin infections”
“In people with recurrent skin infections do prophylactic
antibiotics reduce recurrence rates”
• Foreground question
• Intervention
16. What is the problem, intervention, comparator,
and outcome?
• Problem/Population:
• Who are the relevant patients/population group and what is the
problem.
• Intervention or exposures and comparator:
• What are the treatment / exposure being considered?
• What is the comparator?
• Outcome:
• What are the person-relevant consequences of the exposure
that we are interested in.
17. What is the problem, intervention, comparator,
and outcome?
Population
In patients with recurrent skin infection
Indicator (intervention, test, etc)
Do prophylactic antibiotics
Comparator
Compared with no treatment
Outcome
Reduce recurrence rates
18. The Finer criteria for a good research question
1. Feasible
Adequate number of patients, adequate expertise and resources.
Affordable in time and money.
Manageable in scope.
2. Interesting to the investigator
3. Novel
Provides new findings
Extends previous findings
Confirms previous findings
4. Ethical
5. Relevant
To scientific knowledge
To clinical and health policy
To future research directions
20. Types of Study
1. Observational studies
Researcher has an observational role
Researcher does not intervene, leaves nature takes its course
Researcher role is to record what happens or what happened in
the past.
“Bread-and-Butter”
2. Experimental studies (Interventional Studies)
Researcher actively attempts to change something to alter the
disease course.
21. Observational Studies
1. Descriptive studies:
Describes the occurrence of disease and exposure.
Most commonly used
Look for patterns of disease, to measure the occurrence of
disease, to identify risk factors for disease
Concerned with the „person, place, and time‟
Questions “Who? What? Where? and When? (Not Why?)
Includes:
○ Case Reports
○ Case Series
22. Observational Studies
1. Analytical studies:
Incorporate analysis of association between exposure and disease.
Involve planned comparisons between people with and without
disease, between people with or without exposures thought to cause
disease.
Try to answer question “Why”
Includes:
○ Cohort Studies (follow-up Studies).
○ Case-Control Studies.
○ Cross-sectional Studies.
○ Ecological Studies.
23. Cross-sectional Studies
• Sample of the subjects in a population are investigated for
outcome and/or exposure.
• Used as a first step in more complex design
• Simple description of disease prevalence (blood pressure,
height, DM)
• Known as: Prevalence Studies
24. Cross-sectional Studies
Advantages
1. May study several
outcomes and exposure
2. Short Duration
3. Good first step
4. Yield prevalence and
relative association
5. Inexpensive
Disadvantages
1. Does not establish
sequence of events.
2. Survivor bias
3. Not feasible for rare
conditions
4. Does not yield incidence.
25. Cohort Studies
• Study in which people, who are free of the disease of interest
(outcome) but differ on a certain exposure (study factor), are
followed and the incidence of disease measured.
• Follow – up Studies: follow people over time to see what
happen to them
• Included participants must be free of the outcome of interest.
• Mostly Prospective study
26. Cohort Studies
Advantages
1. The exposure has
definitely preceded the
outcome (causal
association).
2. Establish Incidence
3. Multiple outcomes
4. Other factors can be
measured
Disadvantages
1. Not efficient for rare
diseases.
2. Expensive
3. Need long time
27. Case-Control Studies
• The subjects are defined or selected by disease status
(outcome) not by exposure status.
• Choose individuals with the disease or outcome of interest
and a comparison group without the disease (controls,
reference group), and the measure their past exposure to
certain risk factors.
• Retrospective Study
28. Case-Control Studies
Advantages
1. Ideal for rare cases
2. Short duration
3. Inexpensive
4. Small subjects are
required.
Disadvantages
1. Limited to one outcome variable.
2. Selection bias (cases or controls)
3. Does not establish a sequence of
event
4. Measurement bias
1. recall bias
2. Interviewer bias
5. Survival bias
29. Intervention studies or experiments
• A study in which the investigator intentionally alters one or
more factors under controlled conditions in order to study the
effect of doing so.
• Includes:
• Randomized Controlled (clinical) trials (RCTs)
• Preventive Trials
• Community Trials