Data Interpretation

Question 1

What are the different forms of outcome measurement?

Answer 1

Incidence:
-total number of new cases during a specified period within a defined population

Prevalence
-total number of individuals with the disease at particular time

RR
-probability of an outcome in exposed group compared to the unexposed group

OR

• approximates the risk ratio
• used when the outcome is rare

Attributable risk
-excess incidence of outcome attributed to the exposure (risk exposed-risk unexposed)

Question 2

How would you interpret the following results relating to the risk of remission following chemotherapy:
1.35, 95% CI 1.25-1.6

Answer 2

People who received chemotherapy are 35% more likely to enter remission compared to those that did not receive chemo. The likelihood of remission could have been as much as 60% or as little as 25%. The results are statistically significant with due to the 95% CI not including 1. Therefore you can be 95% certain that the effect of chemotherapy on remission will fall within this range.

Question 3

Interpret the following data where we are looking at the effect of asparin preventing CVD:
RD -0.024

Answer 3

There will be 24 fewer cases of CVD per 1000 for people treated with asparin compared to if they received the control

Question 4

Interpret the following where we are looking at the effect of statins on decreasing the risk of developing ARM:
Crude OR = 0.45 (95% CI 0.32-0.64)

Answer 4

People taking statins will be 55% less likely to develop ARM compared to those that did not receive the statins. Taking the statins could reduce the risk of ARM be as much as 68% or as little as 36% compared to the control. The results are statistically significant due to the CI not including 1.

Question 5

Interpret the following results where we are looking at the difference in antibiotic use between the clinical score group and the control:
RR= 0.71 (0.5-0.95; p<0.02)

Answer 5

Those in the clinical score group were 29% less likely to use antibiotics compared to the control. The reduction in antibiotic use between the clinical score group and the comparators could have been as much as 50% or as little as 5%. These results were statistically significant because the p value is less than 0.05 and the confidence interval does not contain 1.

Question 6

What could you interpret 75% sensitivity to mean?

Answer 6

75% of people with the disease are correctly identified.

25% of people with the disease were not correctly identified

If someone receive a negative result they are more likely to be a true negative than a false negatives

Question 7

What you could you interpret 87% specificity to mean?

Answer 7

87% of people who do not have the disease correctly

13% of people who did not have the disease were falsely identified as having the disease

If someone received a positive result they are more likely to be a true positive and have the disease than be a healthy person who has been falsely identified as having the disease I.e. TP>FP