wk 4- error

Question 1

systematic error is?

Answer 1

there is a systematic difference between the groups and it is not due to chance.

Question 2

random error is?

Answer 2

due to chance

type 1/type 2 with a % chance that it is random

Question 3

selection bias/error is

Answer 3

to do with the participants in the study
when the subjects studied are not representative of the target population about which conclusions are to be drawn

Question 4

measurement bias

Answer 4

error that occurs in the collection of data in a study.

Question 5

precision deals with and accuracy deals with

Answer 5

systematic error, random error

Question 6

risk ratio is also called?

Answer 6

relative risk

Question 7

types of bias without a control group

Answer 7

-volunteer bias: more motivated
-maturation: not isolating intervention
-natural progression: conditions will naturally improve
-regression to the mean: outliers move closer to mean after repeated tests
-placebo effect: effect due to subject believing something is happening
-Hawthorne effect: effect due to attention received from study
-rosenthal effect: participants perform better because of expectation

Question 8

factors to consider when understanding results of RCTs (2)

Answer 8

• the size or magnitude of the intervention effect (this may be provided as continuous data or dichotomous data); and
• the precision of the intervention effect (which can best be determined by inspecting the confidence interval for the intervention effect).

Question 9

appraising a paper includes what? 3

Answer 9

1. internal validity
2. impact (size and effect found)
3. external validity / generalizability

Question 10

types of randomization processes 5

Answer 10

simple- no factors just random division of groups
cluster- randomise an entire group (town, clinic, school, etc)
stratified- group individuals with confounding factors evenly (heart attack in both groups)
block- equal numbers in each arm of study by organizing randomization in blocks
quasi-random allocation- not as true randomization. grouped by a common factor (DOB, alphabetical, etc)

Question 11

what is a consort

Answer 11

a flow diagram tracking the participants throughout a trial

Question 12

concealment v blinding

Answer 12

concealment is from the health professionals
blinding is the participants

Question 13

difference between external validity and clinical significance

Answer 13

external validity is the generalizability of the study results to be applied to people other than those in the study

clinical significance is if the study results are worthwhile

Question 14

confounding and examples

Answer 14

muddling of the result when the relationship between exposure and outcome is confused by the effects of something else - the confounder. A confounder must be associated with both the exposure and the outcome, but independent of both.
Some examples include: Age, socioeconomic status, gender, education level, ethnicity

Question 15

difference in event rates is known as

Answer 15

absolute risk reduction

Question 16

relative risk reduction has what no effect value

Answer 16

0

Question 17

absolute risk reduction has what no effect value

Answer 17

0

Question 18

odds ratio

Answer 18

odds of an event in the intervention group divided by the odd of an event in the control

Question 19

relative risk /risk ratio

Answer 19

probability of the event occuring in the intervention group verses the control group

Question 20

only apply if what

Answer 20

the findings are statistically and clinically significant

Question 21

questions for applying

Answer 21

do the results apply to our situation?
do the benefits outweight the harm, costs, inconviences?
what factors need to be considered when applying?