Guidance to Understand Research AMSRA – AINSHAMS MEDICAL STUDENTS RESEARCH ASSOCIATION
Apr 06, 2016
Guidance to Understand Research AMSRA – AINSHAMS MEDICAL STUDENTS RESEARCH ASSOCIATION
Today’s ILOs
Definition of Research
Scientific Validity Definition
Writing research Question Using PICOT Formula
FINER Criteria to choose your research question
Differentiate between Qualitative and Quantitative Research Design
Study Hypothesis Definition and types of Hypothesis
What Does Random Error Mean ( Type 1 and Type 2 Error) Mean?
What Does Variable Mean
What Does Effect Size Mean
What Does Outcome mean
An Introduction to Sample Size definition
Clinical Reseaech Vs Clinical PracticeClinical research
Systematic Approach designates activity designed to test hypothesis, permit conclusions, and thereby
develop or contribute to generalizable knowledge (expressed in theories, principles, and statements of
relationships)
Clinical practice
interventions designed solely to enhance well-being of individual patient or client and have reasonable
expectation of success, purpose to provide diagnosis, preventive treatment or therapy to particular individuals
Or
To Provide The Best Available Care for The patient ( Socially , Economically , Scientifically )
Thinking as A Researcher
Step 1 : An Idea of Making Research
Step 2 : Writing Research Question
Step 3 : Choosing a Research Design
Step 4 : Developing Hypothesis
Step 5 : Identifying my Variables
Step 6 : Identifying my outcomes
Step 7 : Writing a Protocol and Getting IRB approval
Step 8 : Statistical analysis Plan
Step 9 : Data collection and Management
Step 10 : Data entry and analysis
Step 11 : Writing Down a paper
Thinking as A Researcher Measurement Terminologies : Subject : living To be studied
Predictor Variable : data who changes during study and need to be recorded
Outcome Variable : the resultant of the study
Clinical Endpoint : the target outcomes of the trial or Cause of Withdrawal of Trial
Surrogate endpoint : a biomarker intended to substitute for a clinical endpoint
Effect Size : The size of association between The predictor and Outcome Variable
Precision VS accuracy
Types Of Subjects :
Animal Subjects
Human Subjects
- ordinary population : Have the capacity to consent
- Vulnerable Groups e.g : Children, Pregnants , Institutionalized .
Types Of Variables :
Categorical
Qualitative Data
Nominal : Mutual Exclusive
e.g : Sex – Blood Groups
Ordinal : ordered Mutual Exclusive
e.g Disease Severity – Pain Severity
Numerical Quantitative Data
Discrete: have Considerable
Value e.g Days Sick in last
Month
Continuous: Can Take Any Value e.g Weight – Height
Categorical Values can be Binary or Dichotomous : have Two values e.gdead/alive – Male/female
Event A Event B Event A Event B
Mutual Exclusive Not Mutual Exclusive
Define Type of Variable
Blood Pressure Measures
Size of Ulcer
Glagow Coma Scal
Hb Level
Hand Deformities in Rhuematoid Arthritis
Osteoarthritis Grading in Xray
Age groups
Pregnancy status
Outcomes
Event outcomes
MorbidityMortality
Surrogate/biomarker outcomes
Resvascularization (e.g., CABG, PTCA)
Patient-specific outcomes
Pain/symptoms
Quality of Life/PROs
Composite outcomes
Precision Vs Accuracy
Precision also Means : Reproducibility – Consistency – Reliability
Accuracy Nearly Equal to Validity
Confounding Factor Variables that Affect the Predictor and Outcome and didn’t
Measured
Confounding Factor
OutcomePredictor
Thinking as A Researcher
Hypothesis : a statement about what investigator believes to
be true about nature and relationships of two or more
variable to each other
Null Hypothesis H0 : No Difference between the predictor
and outcome
Alternative Hypothesis H1 : There’s A difference between the
predictor and the outcome
One Sided hypothesis : Takes one direction .Used When one Direction of
the effect is Clinically important
Two sided hypothesis : state that association exist but doesn’t Specify
Direction
The Random Errors مامسوك فى قضية حشيش و مخدرات ... سنة 20... محمد أحمد عبد السالم
محمد فى فرصة القضاء اتهم محمد برأ محمد القضاء
الحكم عليه
محمد مذنبشامخالقضاء!!هرب
محمد برئ!!ظلم القضاء شامخ
The Random Errors مامسوك فى قضية حشيش و مخدرات ... سنة 20... محمد أحمد عبد السالم
No Association
between Predictor and outcome
ASSOCIATION
BETWEEN
PREDICTOR AND OUTCOME
Type 1
FP
CorrectReject Null
H0
CorrectType 2
FN
Fail to
Reject H0
The Random Errors
Type 1 Error (FP) : Rejecting Null which is Actually
true in Nature
Type 2 Error ( FN ) : Accepting Null Which is Actually Not True in Population
Thinking as A Researcher Literature : every source of Info to be used in searching process
Errors in Research :
a) Systematic Error : errors due to Bias => Affect Accuracy and Validity
of Research
b) Random Error: errors due to Chance => Affect Precision
Bias : The Distortion of Results from its correct answer
Translational Research : Translating findings from
A) Basic Science to clinical Research ( T1) B) From Clinical Research to Population (T2)
Thinking as A Researcher
Sampling Terminologies : Population ( of interest ) : Set of people with specific Character with Certain
characteristics ( Target Population )
Sample : subset of population of interest that represent them in nature
Thinking as A Researcher
Source Population
Study Population
SampleSample Size Calculation
Statistical Inferences
Thinking as A Researcher Consider an epi study carried out in Cairo to assess whether obesity is associated with
hypertension in young adults. The investigators decided that it was not feasible to
consider taking a sample from among all young adults in the city. It was decided that
fitness centers would provide a large source of young Egyptian adults. A sample of
subjects is taken from several randomly selected fitness centers throughout the city and
their blood pressure is measured to determine hypertension status.
1. What is the source population for this study?
2. What is the study population in this study?
3. Does the sample represent the study population?
4. Does the study population represent the source population?
Types of Sample
Types of Sample
Non-probablitySampling
Convenient Sampling
Snowball Sample
Quota Sample
Judgmental Sample
Probability
Simple Random Sample
Systematic Sample
Stratified Sampling
Cluster Sampling
Types of Sample
Probability Sample :
a sampling technique in which the probability of getting any particular
sample may be calculated
Simple Random Sample : individual is chosen randomly and entirely by
chance, such that each individual has the same probability of being
chosen at any stage during the sampling
Systematic Sample : selection of subjects from an ordered sampling
frame.
Stratified Sample : dividing population into stratum then choosing simple random or systematic sample from each
Cluster Sample: used when "natural" but relatively homogeneous groupings
are evident. the total population is divided into these groups (or clusters)
and a simple random sample of the groups is selected.
Simple Random Sample
Cluster Sample
Systematic Sample
Stratified Sample
Types of Sample
Non - Probability Sample :
Convenience, Haphazard or Accidental sampling - members of the population are chosen based on their relative ease of access.
Snowball sampling - The first respondent refers a friend. The friend also
refers a friend, and so on. Used with population who fear of Social
Stigma
Judgmental sampling or Purposive sampling - The researcher chooses
the sample based on who they think would be appropriate for the study. This is used primarily when there is a limited number of people that have
expertise in the area being researched. Such samples are biased
because prominent experts may differ from other, equally expert, less prominent persons.
Thinking as A Researcher Longitudinal Studies : Studies that assess finding over Time ( Prospective
or Retrospective )
Prospective ( concurrent ) Study : assess Findings in advance
Retrospective( Non- Concurrent ) Study : assess Findings backwards
Cross- Sectional : assess Findings in point of Time
Longitudinal Study is Like A Film while Cross-sectional Study is Like A Photo
Step One: An Idea of Making Research
Origin of Good Research Question Mastering the Literature
Being Alert to New ideas and Technology - Conferences
- Published Work in this Area
Keep the imagination rooming
My Clinical practice:
“What do I have” (diagnosis)
“How bad is it” (prognosis)
“Can I give it to my family?” (natural history)
“How did I get this?” (etiology)
“Will this stuff help me?” (treatment or prevention)
Choosing Mentor - Person who has experience in that Field with published work in
Criteria for Good Research Idea : FINER
Feasible
Interesting
Achievable
Novel
Ethical
Relevant
Feasibility and Scientific Validity
Feasiblity Includes :
a) Number of Subjects
B) Techinical Expertise
C) Cost in time and Money
D) Scope
Scientific Validity : ability of study to correctly answer research question
posed
Feasible but Not Valid : Not ETHICAL
Study Results
Truth in the Study
Truth in population
Actual Subjects
IntendedSample
PopulationIn
fere
nc
e
Imp
lem
en
tatio
n
Internal Validity ( Control )
External Validity (Generalizability)D
esi
gn
ing
Study Results
Truth in the Study
Truth in population
Actual Subjects
IntendedSample
Population
Infe
ren
ce
Imp
lem
en
tatio
n
Subjects Enrollment
Inclusion Criteria
De
sig
nin
g
Step 2 : Designing a Good Research Question
Criteria of Good Research Question : PICOT
Population
Intervention
Comparison
Outcome
Time
Phases of Clinical Trials
Preclinical Phases : Testing of drug in non-human subjects, to gather
efficacy, toxicity and pharmacokinetic information
Phase 0 : First in Human Trial
Phase I : Testing of drug on healthy volunteers for dose-ranging using
Subtheraputic doses
Aim : To Know Dose –Limiting Toxcity (DLT )
Phase 2 : Testing of drug on patients to assess efficacy and safety
Phase 3 : Testing Drug on Patients in comparison to the standard
treatment
Phase 4 : Post Marketing Surveys
Phase 5 : looping – Translational Research
Step 3 : Study Designs
Study designs
Observational
Descriptive
Case Report
Case Series
Ecological Study
Analytical
Cross-sectional
Study
Case – control Study
Cohort Study
Experimental
Clinical Trial
Sequential Trial
Cross-over Trial
Historical Withdrawal Trial
Superiority Trial
Non- Inf. Trial
Factorial Trial
References
Principles and practice of Clinical Research , 3rd edition
Designing Clinical Research , 4th Edition
Medical Statistics at a glance , 3rd Edition
A Pocket Guide to Epidemiology , 2009
Kaplan Books , Epidemiology
Wikipedia , Probability and Non- Probability Sampling