200 likes | 331 Views
How to assess an abstract. Objectives. Understand the principle differences between qualitative and quantitative research Understand the basic statistics employed in research Be able to assess a piece a research with confidence!. Qualitative research. Which type of questions does it answer?
E N D
Objectives • Understand the principle differences between qualitative and quantitative research • Understand the basic statistics employed in research • Be able to assess a piece a research with confidence!
Qualitative research • Which type of questions does it answer? • What methodologies are employed? • Improving their validity
Assessing a qualitative paper • Is the qualitative approach appropriate? • Methodology • Data analysis • Results and conclusion
Quantitative • Types of quantitative research • RCT – design features, advantages & disadvantages • Cohort Studies • Case control studies • Cross section surveys
BIAS • Selection bias • Observer bias • Participant bias • Withdrawal or drop out bias • Recall bias • Measurement bias • Publication bias
Commonly used statistics • P values • Relative Risk Reduction • Absolute Risk Reduction • Numbers Need to Treat • Sensitivity • Specificity • Positive Predictive Value • Negative Predictive Value
P values & CI • p value = the probability of the outcome being due to chance • p = 1 in 20 (0.05). • > 1 in 20 (0.051) = not significant • < 1 in 20 (0.049) = statistically significant CONFIDENCE INTERVALS This defines the range of values between which we could be 95% certain that this result would lie if this intervention was applied to the general population
RR, AR, ARR & RRR • What are they? • How do you calculate them?
Warfarin & AF study • The annual rate of stroke was 4.5% for the control group • Absolute Risk (Control group) = 0.045 • 1.4% for the warfarin group • Absolute Risk (experimental group) = 0.014 • Absolute Risk Reduction = 0.045 – 0.014 = 0.031 • NNT = 32 • Relative Risk = 0.014/0.045 = 0.31 = 31% • Relative Risk Reduction = 0.045 – 0.014/0.045 = 0.68 = 68%
NNT How many people you need to treat with the study intervention to stop the study event from happening once. 1/ARR = Number Needed to Treat.
Sensitivity • The test’s ability to correctly identify those people with disease. • If Sensitivity is <100% Disease is missed. • So = True Positives • True Positives + False negatives i.e. all those who truly Have the disease
Specificity • The test’s ability to correctly exclude those people without disease • If Specificity <100% then healthy people are told they may have disease = True Negatives True Negatives + False Positives i.e. all those who truly don’t have the disease
Positive predictive value • If the test is positive, what is the chance of the person having the disease = positive predictive value. True Positives True positives + False Positives
Negative Predictive Value • If the test is negative, what chance is there that the person doesn’t have the disease = negative predictive value. • True negative True negative + False negative
Accuracy • True positive + True negative True negative +true positive+ false negative + false positive
Urine dipstick to screen for Diabetes • Example- urine dip test vs GTT (the gold standard) Diabetes +ve Diabetes –ve • Result of urine test (n=27) (n=973) • Glucose present (13) True +ve 6 False +ve 7 • Glucose absent (987) False –ve 21 True -ve 966