HOW TO PRESENT LIMITATIONS AND ALTERNATIVES Flashcards

1
Q

what is the key principle in presenting limitations?

A

transparency, we want to identify and present them

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

what are the four steps in presenting limitations?

A
  1. describe the limitation
  2. describe the impact of this limitation on our study findings
  3. discuss alternatives and why they were not selected
  4. describe the methods proposed to minimise the impact of the limitation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

describing the limitation

A

specific the type e.g. error, dangerous limitation such as selection/info bias/confounding
consider limitations of prior literature, however don’t use jargon

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

describing the impact of limitation on study findings

A

project the likelihood, magnitude, direction of limitation f study findings
bias towards null = less dangerous than those away from null

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

discussing alternatives

A

discuss the pros and cons of these alternatives and say why you chose the approach you did
be transparent into why you chose a type of design/analysis as shows reviewer your aware of alt as well as impact of decision on interp of findings

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

examples of design techniques to minimise study limitations

A

examples
- choosing prospective design of case-control to avoid recall/selection bias
- binding interviewers reduces interviewer bias
- incorporate repeated questionnaires to minimise nondifferential misclassification of exposure to changes in beh over time
- use of life event calendars to boost accuracy of recall

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

examples of analysis techniques to minimise study limitations

A
  • compare baseline characteristics of experimental and standard care group showing randomisation successful
    -sub group analysis with/without missing data on key variables of interest
How well did you know this?
1
Not at all
2
3
4
5
Perfectly