Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

STAT > WEEK 1 AND 2 > Flashcards

WEEK 1 AND 2 Flashcards

1
Q

What is the purpose of LITERATURE REVIEW?

A
  • Identifies, evaluates and synthesises the relevant literature within a particular field of research
  • It highlights:
  • what has been done,
  • what is generally accepted,
  • what is the current state of thinking on a topic
  • what is emerging
  • what are the gaps in knowledge
  • A critical analysis of published literature
  • Integration of this information (don’t just summarise & report)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Synthesis vs summary

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Narretive REVIEW vs. SYSTEMATIC REVIEW

Scope of focus
search protocol
study selection
grade
lack of evidenc

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

advantaged and disadvantages of SYSTEMATIC REVIEW

A

ADVANTAGES
▪ Focused, specific research question (PICO)
▪ Comprehensive and Exhaustive
▪ Explicit Methodology (pre-specific criteria)
▪ Accurately assess quality & weight of evidence
▪ Reproducible

DISADVANTAGES
▪ Need to have a very specific question
▪ Need to be very specific about search criteria
▪ Long
▪ Generally requires pre-registration
▪ Requires multiple people
▪ Still subject to biases
▪ Reproducible

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

what is PROSPERO?

A

▪ An international database of prospectively registered systematic reviews in health & social care
▪ Funded by the National Institute for Health Research, UK.
▪ Aims:
1. To provide a comprehensive listing of systematic reviews registered at inception to help avoid duplication
2. To promote transparency in the process and thus minimize the risk of bias in the review
▪ Free to register
▪ Ideally done before screening against eligibility criteria commences.
▪ However, reviews are currently accepted for registration as long as they have not progressed beyond the
completion of data extraction
▪ Scoping reviews and literature reviews are not eligible for inclusion in PROSPERO.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

META-ANALYSES

A

Meta-analyses
* Meta-analysis refers to the use of statistical techniques in a systematic review to
integrate the results of included studies
* Not a requirement & in some cases, it doesn’t make sense to perform one
* Generally NOT recommended to undertake a meta-analyses outside of a systematic
review

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

ORIGINS OF A GOOD RESEARCH QUESTION

A

▪ Guiding judgements of mentors (supervisors) who are experts in the area
▪ A review of the literature is a great first step in identifying “gaps in knowledge”
▪ Be alert to new ideas and techniques
▪ Creativity
▪ Tenacity (returning to a troublesome problem repeatedly)
▪ Conclusion from the research (regardless of the exact findings) would be impactful

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

STEPS IN A RESEARCH PROJECT

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

FORMULATING THE RESEARCH QUESTION

FINER

A

Objective of the study
The uncertainty or gap that the investigator wants to resolve
Begin general but often needs to be narrowed down to be ‘testable’ / ‘feasible’
Needs to meet the FINER formula (Hulley et al., 2001)
✓ Feasible (can be done/tested)
✓ Interesting (of interest/importance)
✓ Novel (not already done)
✓ Ethical (essential)
✓ Relevant (an implication,
meets the “so what test” (should contribute meaningfully to knowledge)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

FINDING THE RIGHT BALANCE - NARROW VS. BROAD QUESTIONS

A

Very broad questions
Difficult to ‘test’
Challenges with methods &
recruitment

Very narrow questions
May have limited generalizability
Could lead to biased
interpretations/conclusions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

PICO

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

DIFFERENT TYPES OF RESEARCH QUESTIONS

6 types,

A
  1. INTERVENTIONS Treatments and therapies to treat a disease/condition
  2. AETIOLOGY, RISK FACTOR What causes the problem? Outcomes of an activity/exposure
  3. FREQUENCY What is the frequency of the problem?
  4. DIAGNOSIS Does this person have the problem?
  5. PROGNOSIS, PREDICTION Who will get the problem?
  6. PHENOMENA Observations of a situation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

example of intervention, frequency, diagnosis and phenomena research question

A

Intervention : In people aged ≥70 years, is 100mg daily Aspirin, versus placebo,
effective in reducing the incidence of acute ischemic stroke
Frequency: In infants born prematurely, what is the subsequent lifetime prevalence
of sensory deafness compared to those born full-term?
Diagnosis : For pregnant women, is ultrasound plus serum biochemistry testing in the 1st
trimester as accurate (with equal or better sensitivity & specificity) as conventional
amniocentesis for diagnosing Down’s Syndrome?
Phenomena: contains P and O
Usually involve a population (P) and an outcome (O) but not an intervention or comparator.
Question is more open ended and informed by the study itself
For mothers of children with a fever, what are the principal concerns?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

HYPOTHESIS TESTING

What is hypothesis ?
when is it required

A

A proposed explanation made on the basis of incomplete evidence, but used as a starting point for further
investigation
▪ A simplified & specific version of the research question that summarises the main elements of the study.
▪ Hypotheses are NOT needed for descriptive studies or observations of a phenomena
▪ BUT essential when any test of statistical significance will be used to compare findings among groups
▪ At least one hypothesis is being ‘tested’ in all experiments & most observational studies
▪ If any of the following terms appear in the research question a hypothesis should be formulated: greater than, less
than, causes, leads to, compared with, more likely than, related to, similar to, correlated with
▪ It MUST be stated in advance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

what is the hypothesis for this
Research Project Title: Social media and medical education
Research Question 1: How can social media best facilitate student learning in medical education?

A

Hypothesis: Not needed because descriptive study (no use of following words within research question: greater than,
less than, causes, leads to, compared with, more likely than, related to, similar to, correlated with)

17
Q

WHY IS STUDY DESIGN IMPORTANT?

A
  • Study design determines the way in which the study will be conducted
  • Includes sampling (who, what, when) & types of information (data) gathered
  • It will also govern:
    ✓ methods used to collect the data
    ✓ how the data are analysed
  • It WILL directly affect your results & the conclusions drawn
  • It will dictate more precisely the research question that your study will address
18
Q

WHY IS STUDY DESIGN IMPORTANT?

A
  • Consider the differences between
    Your research question - What you want to answer ?
    Study plan & design - What the study is designed to answer?
    Given the errors/limitations of implementation etc. - What the study actually addressed?
  • In many ways it is considered more important than the analysis
    A badly designed study→ can never be retrieved
    Poorly/incorrectly analysed data→ can usually be reanalysed
19
Q

HOW DO I DETERMINE THE MOST APPROPRIATE STUDY DESIGN?

A

It will depend on the type of research question
Aim is to optimise efficiency to answer the research question

20
Q

BROAD CATEGORIES OF RESEARCH QUESTIONS

A
21
Q

WHERE IS YOUR DATA COMING FROM?

A

Primary data
▪ Originated by the researcher for the specific purpose of addressing the research question
Secondary data
▪ Data have already been collected for purposes other than the specific research question

22
Q

SECONDARY DATA ANALYSIS

A

▪ Previous research studies may provide a rich source of secondary data
▪ Many studies collect more data than the investigators analyse
▪ Access to such data is controlled by principal investigator
▪ Other sources of data include national data sets and registries
▪ Ancillary studies where an investigator adds one or several measurements to
an existing study to answer a different research question

23
Q

ADVANTAGES / DISADVANTGES OF primary and secondary data

A
24
Q

YOUR RESEARCH PROJECT
You need to consider

A

Consider:
1. Primary or Secondary data analysis
2. Descriptive, analytical or causal study design?
3. Any further specification of the exact study design to be used?
4. How you will actually going about implementing the study?
5. What other aspects might you need to consider?

25
Q

HOW DO I DETERMINE THE MOST APPROPRIATE STUDY DESIGN?

A

The language of the research question can be helpful in deciding what
research design and methodology to use.

26
Q

DESCRIPTIVE – OBSERVATIONAL

A
  • Quantify the information in terms of numerical data
    “How many mothers of children with a fever have concerns?”
    “How many mothers of children with a fever are concerned about the correct medication to use?”
    “How common is night shift work in health care?”
27
Q

DESCRIPTIVE – PHENOMENON

A
  • Determine whether a phenomenon exists & describe it’s characteristics
  • Through observation, analysis & description

“Do mothers of children with a fever have concerns?”
“What are the principal concerns for mothers of children with a fever?”
“What are the experiences of people working night shifts in health care?”

28
Q

RELATIONAL – ANALYTICAL
Study design

A
  • Once the phenomena have been described
  • Specific questions about how factors relate to one another
    “Are new mothers more likely to worry when their child has a fever?”
    “Is there a correlation between the duration of a fever and the extent of concern a mother has?”
    “Is night shift work among health care workers, associated with a poorer
    quantity and quality of sleep?”
29
Q

EXPERIMENTAL - CAUSALITY
Research question

when?
PICO

A
  • Once the phenomena have been described & relationship been factors established
  • Interested in whether an intervention or an activity affects something else
    “Does intervention X versus standard practice, reduce the ‘degree’ of a mother’s concern,
    when their child has a fever?”
    “Is drug X versus paracetamol, better at reducing fever in young children?”
    Remember PICO: Population, Intervention, Control, Outcome
30
Q

TWO MAIN APPROACHES TO A RESEARCH PROBLEM

A

Quantitative Research (late stage, narrow, focused, conclusive)
* Tries to find answers to concrete questions by generating numbers & facts
* Only measurable data are being gathered and analysed
* Research question often starts with “How many/often”, “Why” or includes terms like
“Effect/Influence/Correlation/Impact/Cause“
Qualitative Research ( early stage, broad, whole picture, exploratory)
* Exploratory or interrogative research: describe, discover, understand, tries to get “under the surface”
* Gathering of mainly verbal data rather than measurements
* Analysed in an interpretative manner, subjective, impressionistic or even diagnostic
* Research question often starts with “How” or “What”

31
Q

QUANTITATIVE STUDY DESIGN
Map

A
32
Q

HIERARCHY OF EVIDENCE PYRAMID

A
33
Q

DIFFERENT STUDY DESIGNS
Ecological
Case study
Cross sectional
Case control
Nested case control
Nested case cohort
Cohort
Clinical trials

A
34
Q
A

STAT (7 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions