Research Quiz 1

Front 1

EBM = integration of:

Back 1

1. Individual clinical experience
a. Vital for determining whether the evidence applies to pt AT ALL and if so, how?
2. Best available external clinical evidence
a. From clinical research among intact pts.
b. Short doubling time (10 yrs)
c. Replace current tests and treatments with new ones that are more accurate, powerful and safer
3. Patient’s values and expectations
a. Getting better at quantifying

Front 2

Ideas Behind New Medical Practice

Back 2

• Many aspects of clinical practice cannot, or will not, ever be adequately tested
• Study of basic diseases is necessary but not sufficient
• Understand certain rules of evidence is necessary to correctly interpret literature

Front 3

How can clinical performance be kept up to date??

Back 3

1. Learning how to practice EBM
2. Seeking and applying EBM
3. Applying EBM for changing our clinical behavior

Front 4

Desirable characteristics of research:

Back 4

1. The study addresses specific clinical question of the PT
2. Study participants are similar to your pt.
3. The study was published in a **peer-reviewed format**
4. The context of the study is consistent with **contemporary** healthcare

Front 5

What is Bias

Back 5

Systematic deviation from the truth that occurs as a result of uncontrolled influences during the study

Front 6

Research Design Types (1-4)

Back 6

1. Experimental
2. Quasi-experimental
3. Observational
4. Case Reports

Front 7

Research Design Types
1. Experimental

Back 7

Classic design approach (RCT); most control you can have in a study!

Front 8

Research Design Types
2. Quasi-experimental

Back 8

Not as controlled, but still has some elements of experimental design
(Ex: all pt get same tx but objectively look at results)

Front 9

Research Design Types
3. Observational

Back 9

Non-experimental; just observing over time, but in a systematic way (same outcome measures)

Front 10

Research Design Types
4. Case Reports

Back 10

Descriptive studies; documenting what happens with pt; may be about diagnostic process, intervention, etc.

Front 11

Three Factors that make Research Difficult

Back 11

- Research with small or "rare" populations
- Topics that are hard to define (fatigue, endurance, etc)
- If there's not enough funding

Front 12

PICO

Back 12

Population
Intervention
Comparison
Outcome

Front 13

Levels of Evidence (best to worst)

Back 13

1. Randomized Control Trial (RCT)
2. Lesser quality RCT
3. Case Control or Retrospective Study
4. Case Series
5. Expert Opinion

Front 14

Levels of Evidence (best to worst)
1. Randomized Control Trial (RCT)

Back 14

a. Systematic review of RCT
b. Individual RCT (high quality)

Front 15

Levels of Evidence (best to worst)
2. Lesser quality RCT

Back 15

a. Systematic review of cohort studies
i. Cohort = studying one specific group of people over time
b. Individual cohort study
i. Including lesser quality RCT
c. Outcomes Research

Front 16

Levels of Evidence (best to worst)
3. Case Control or Retrospective Study

Back 16

a. Systematic review of case-control studies
b. Individual case-control studies
c. Case = people with disease of interest
d. Control = people without disease of interest

Front 17

Levels of Evidence (best to worst)
4. Case Series

Back 17

a. More than one pt.
b. Stronger evidence than case report

Front 18

Levels of Evidence (best to worst)
5. Expert Opinion

Back 18

a. Includes textbooks

Front 19

Primary Ethical Issues (that should be considered when designing a research study)

Back 19

• Safety
• Informed consent
• Confidentiality
• How adverse events will be handled

Front 20

Belmont Report – set 3 ethical principles

Back 20

1. Autonomy = obligation of investigator to respect each subject as a person capable of making informed decisions regarding participation in study
a. Subjects receive full disclosure
2. Beneficence = obligation of investigator to attempt to maximize benefits for subject which minimizing risk of harm
a. Honest risk/benefit analysis should be performed
3. Justice = equitable selection of subjects (avoiding coercion of prisoners, poor people, institutionalized people, etc)

Front 21

5 Components of Informed Consent

Back 21

1. Disclosure
2. Understanding
3. Volunteerism
4. Competence
5. Consent

Front 22

P of PICO

Back 22

Population-details about pt, i.e. gender, dx, severity

Front 23

I of PICO

Back 23

Intervention-could be a dx test, risk factor or outcome
- "Indicator" in diagnostic or prognostic studies (talking about risk factor)

Front 24

C of PICO

Back 24

Comparison-if appropriate

Ex: pt has shoulder impingement, trying to find dx test. Learned about Neer and Hawkin. Comparison may be "Is te Neer test better than Hawkin for identifying impingement?"
Ex2: Diagnositic "Is TxA more effective for shoulder impingement than TxB", and then Outcome can change and be whatever you'd like or what the pt is intereested in

Front 25

O of PICO

Back 25

Outcome-consequence of interest for Rx, Dx or
outcome
- This is what will change in your search

Front 26

What are the 3 most important factors in a Diagnostic Question

Back 26

tests or measures that give ...
1) the best information
2) in the least amount of time 3) with the least amount of cost and risk

Front 27

Questions about Prognosis

Back 27

• Social factors affect prognosis
• Predicted optimal level of importance through intervention and amount of time needed to reach that level

Front 28

Questions about Interventions

Back 28

• Is intervention in line with pt. goals?
• Cost vs. benefit
• Practical intervention for pt.?

Front 29

Questions about Outcomes

Back 29

• Validity for population (does it measure what you want it to measure?)
• Reliable
• Standardized
• Sensitive to change (if a pt. gets better, will the outcome measure show that?)

Front 30

Concept vs. Construct:

Back 30

Concept = mental image of an observable phenomenon expressed in words (i.e. fatigue, pain, flexibility, age)

Construct = non-observable abstraction created for research purpose and defined by observable measures (i.e. motivation, patient satisfaction, health related QOL)

Front 31

Hypothesis:
Null Hypothesis (Ho)

Back 31

1. Null hypothesis (Ho)
a. Aka “statistical hypothesis”
b. No difference or no relationship  everything related to statistical testing falls into these 2 categories
c. Tests are designed to challenge the null and attempt to reject the null

Front 32

Hypothesis:
Alternative hypothesis (Ha or H1)

Back 32

a. Aka “research hypothesis”
b. There is a difference or there is a relationship
c. Best if directional language is used

Front 33

Theory

Back 33

Organized set of relationships among concepts and constructs that is proposed to systematically explain a phenomenon or interest (Darwin’s Theory of Evolution, Nagi’s Disablement Model)

Front 34

Research Design Features & Components:
• Philosophical Perspective

Back 34

o Quantitative - Objective reality, measured in
numbers

o Qualitative - Subjective reality, measured in
words

Front 35

Research Design Features & Components:

Back 35

o Experimental
o Quasi-experimental
o Non-experimental

Front 36

Research Design Features & Components:

Back 36

o Cross-sectional (Data collected from one point
in time)
o Longitudinal

Front 37

Research Design Features & Components:

Back 37

o Retrospective
o Prospective

Front 38

Research Design Features & Components:

Back 38

o Within groups
o Between groups (this is always the best to use!)

Front 39

Prognosis – one of the following 3 designs is usually used:

Back 39

1. Prospective cohort
2. Retrospective cohort
3. Case-control study

Front 40

In a prognostic study, Causality can only be inferred when:

Back 40

• Exposure clearly precedes the outcome
• A dose-response relationship between exposure and outcome was demonstrated
• Findings are consistent with previous, well-designed studies (contemporary)
• Results are biologically plausible

Front 41

RCT: Efficacy vs Effectiveness

Back 41

- Efficacy = performed under ideal conditions (in vitro)

- Effectiveness = performed in real world circumstances (in vivo)