The Anatomy of Research
Presented by:
Shahrzad Bazargan-Hejazi, PhD
September 2013
1
Objectives:
 Outline tangible element of a research question
 Identify decisions to be made for selection of research
design
 Describe the purpose of literature review
 Identify decisions to be made in choosing the study
subjects
 Identify decisions to be made in choosing study variables
2
The Research Process
The steps of any population health research project are:
3
What Is Research made of ?
Study Question!
Identifying a study topic is often the most
challenging part of a research project.
Each of the possible study topics has its own set
of virtues and shortcomings.
4
Brainstorming
 Review of previous scholarly work
 Meeting experts
 Attending conferences
 Participating in meetings
 Building relationship with an expert/mentor
5
Brainstorming Questions
6
Key Words: Example
Osteoporosis
Physical therapy
Rehabilitation
Home safety
7
Initial brainstorming about
aging
Falls
Bedsores
Calcium
Prevention
Bone density
Topic Mapping
 Begin by creating a long list of possible study topics:
 Jot down areas of personal interest
 Ask friends / colleagues for ideas
 Skim abstracts, journals, and books for inspiration
 What topics emerge as a repeating theme?
 What might be enjoyable to explore?
8
MeSH (Medical Subject Headings)
9
Exposure, Disease, Population
Most topics in population health research can be expressed in terms
of:
[exposure] and [disease/outcome] in [population]
10
Examples of Types of Exposures
11
Examples of Types of Diseases
12
Examples of Types of Populations
13
“EDP” Study Questions
 Are exercise habits [exposure] related to the risk of bone
fractures [disease] in adults with diabetes [population]?
 Is reproductive history [exposure] related to the risk of
stroke [disease] among women living in rural Ontario
[population]?
 Is household wealth [exposure] related to the risk of
hospitalization for asthma [disease] in Australian children
younger than 5 years old [population]?
14
FINER
Criteria for a Research Question
 F = Feasible
 Adequate # of subjects,
 Adequate technical expertise,
 Affordable in time, money,
 Manageable in scope.
 It helps to know your limits early on to avoid wasting time and
effort over something that is not going to work.
15
FINER Criteria for a Research Question
(cont.)
 I = Interesting
 Answering the question is interesting, not
because, you have to do it.
 Confirm the interest of the question with your
mentor before investing energy and time in
development of research protocol
16
FINER Criteria for a Research Question
(cont.)
 N = Novel
 It contributes to new information by:
 confirming or refuting previous findings,
 extends previous findings,
 provides new findings
 you should be able to answer the “so what”
question
17
FINER Criteria for a Research Question
(cont.)
 E = Ethical
 A good research question should be ethical
 It should not posses unacceptable physical risk to the subjects or
invasion of their privacy
 If so you need to find some other ways to answer your research
question
18
FINER Criteria for a Research Question
(cont.)
 R = Relevant
Is the outcome of the study:
 relevant to scientific knowledge,
 clinical management,
 health policy, and/or
 guides future research direction
19
Problems & Solutions
The RQ is not FINER
1. Not feasible (too broad)
 Not enough subject available
 Method beyond your skills
 Too expensive
2. Not interesting, novel, or relevant
 Consult with your mentor
 Modify RQ
20
Problems & Solutions
3.
•
•
•
Uncertain ethical suitability
Consult with IRB
Modify RP
4. The study plan is vague
• Write the outline of the RP as soon as you can
21
Exercise




22
What is the relationship between depression and
health?
Write in a single sentence a predictor, outcome,
and population.
Discuss whether it meets the FINER criteria
Rewrite the sentence in a form that overcomes any
problems in meeting criteria.
Points to Consider
 Is the question important and relevant?
 Can the question be answered the way it is written?
 To answer the question what needs to be defined?
 Is the answer to the question population sensitive?
23
Possible Answer
 Among college freshmen, does depression assessed by the
CES-D predict health status measured by the Rand General
Health Questionnaire four years later?
24
Study Goals & Specific Objectives
The literature review and consideration of a study approach
should lead to the selection of one very specific study topic
that can be stated in terms of a single overarching study goal or
study question.
A study goal often includes the specific exposure, disease, and
population that will be the focus of the study
Examples of Study Goals
Study Goals & Specific Objectives
After finalizing the overarching study goal, the researcher
should identify three or more specific objectives (also called
specific aims or specific hypotheses) that stem from the main study
goal.
 Each of these specific objectives should take the form of a
measurable question or a “to” statement.
 Each should represent a logical step toward answering the
main study question.
Example
 Study goal: “to assess the impact of lead poisoning on school
performance in kindergarten students in southeast
Michigan.”
 Specific objective #1:
1. To measure the prevalence of high blood lead levels in a random
sample of kindergarten students in southeast Michigan.
Example
 Study goal: “to assess the impact of lead poisoning on school
performance in kindergarten students in southeast
Michigan.”
 Specific objective #2:
2. To determine whether children in that sample with high blood
lead levels have lower scores on academic tests than children
with lower blood lead levels.
Example
 Study goal: “to assess the impact of lead poisoning on school
performance in kindergarten students in southeast
Michigan.”
 Specific objective #3:
3. To estimate the total impact of high blood lead levels on
kindergarten performance in southeast Michigan by applying the
rates in the sample population to the total population of the
region.
Example
 Study goal: “to assess the impact of lead poisoning on school
performance in kindergarten students in southeast
Michigan.”
 Note that all three of these specific objectives relate to
the overall study goal and provide a clear pathway for
achieving the main goal.
ICMJE
ICMJE = International Committee of Medical Journal Editors
Most journals in the population health sciences require
manuscripts to conform to ICMJE’s Uniform Requirements for
Manuscripts Submitted to Biomedical Journals
ICMJE Authorship Criteria
Each coauthor must have met all three of the following
conditions:
1. Substantial contributions to conception and design and/or
acquisition of data and/or analysis and interpretation of
data
2. Drafting the article and/or revising it critically for
important intellectual content
3. Final approval of the version to be published
ICMJE Authorship Criteria
 “Acquisition of funding, collection of data, or general
supervision of the research group alone does not constitute
authorship.”
Authorship Example #1
 A person who conducts interviews for the project but does
not contribute further would not be eligible for authorship.
 An interviewer who also writes a paragraph for the
discussion section would meet authorship criteria.
Authorship Example #2
 A hospital laboratory technician who analyzes blood samples
of patients included in a clinical study but makes no further
contributions would not be eligible for authorship.
 A lab tech who analyzes the samples and writes part of the
methods section describing laboratory techniques would be
a coauthor.
Authorship Example #3
 A data entry assistant who makes no additional contributions
to the project would not be considered an author.
 A data manager who runs statistical tests and creates a table
for the manuscript would meet authorship criteria.
Authorship Example #4
 A technical editor who cleans up the grammar and spelling
in a manuscript does not earn authorship.
 An editor who raises important questions about the
interpretation of the results and the meaning of the work
may be eligible for authorship.
Authorship Criteria
 “All persons designated as authors should qualify for
authorship, and all those who qualify should be listed.”
 No gift authorships
 No ghost authorships
Authorship Order
A typical justification for authorship order:
 The first author is the person who was the most involved in
writing the manuscript.
 The remaining authors are listed in order of contribution,
defined in terms of time dedicated to the project as well as
intellectual contribution.
 List authors with equal contributions in alphabetical order
 The senior (supervising) author is listed last, unless s/he
prefers to be listed in order of contribution.
Decisions about Authorship
 Publications are an important metric of success in the
sciences and academia, and authorship is often the only
reward for the time put into a project. As a result,
authorship decisions can be very stressful.
 It is helpful to decide before a contributor does any work on the
project whether that person will be a coauthor and what role
s/he will play.
 At the end of the project, there should be no surprises about
who is being included or excluded as an author.
Decisions about Authorship
 If someone making a minor contribution will not be a
coauthor, make sure that s/he is not asked to write any part
of the paper or to provide critical feedback.
 If someone will be a coauthor, make sure that s/he has the
opportunity to make an important intellectual contribution
to the writing of the paper.
 Any disputes over authorship criteria or the order of authors
are usually best referred to the senior author on the paper.