Scribd
Upload
8 views6 pages

Checking Questions

Uploaded by

Hafsa Omer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
8 views6 pages

Checking Questions

Uploaded by

Hafsa Omer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Checking Questions

Elfatih M. Malik (MBBS, MD, FPH-UK)


Sudan Medical Specialization Board
Community Medicine Council
Epidemiology, biostatistics and research methodology module
Hand-washing # Patients # non- aOR (95%CI)
Influenza and hand score patients
washing 0 -3 (poorest) 42 (42%) 07 (07%) Ref

4 16 (16%) 11 (11%) 0.26 (0.075 – 0.89)

• Researchers (Liu M et al, 2016/ doi:


10.1097/MD.0000000000003046) 5 16 (16%) 19 (19%) 0.12 (0.032 – 0.44)
conducted a study to test whether the risk
of influenza transmission associated with 6 13 (13%) 24 (24%) 0.073 (0.019 – 0.28)
self-reported hand-washing and unhealthy
hygienic habits among residents in 7 06 (06%) 15 (15%) 0.044 (0.009 – 0.20)
southeastern China. Laboratory confirmed
seasonal influenza cases were
8 -9 (best) 07 (07%) 24 (24%) 0.029 (0.0068 -0.13)
consecutively included in the study
patients (n = 100). Negatively tested
patients of similar age and city of
residence were also included in the study.
1. What is the study design?
Both were recruited at hospital. Telephone
interview was used to collect information 2. What is/are the common types of bias in such a study?
on hand-washing and hygienic habits in 3. How the researchers guard against the possible confounders?
the preceding 2 weeks and the following 4. What is the conclusion of this study?
were obtained.
Results of 5 trials
1. Which trial you trust more?
The 95% confidence interval (95%CI) for a. Trial 1
the difference in mean cholesterol b. Trial 2
(mg/dl) in a five trials of drugs used to c. Trail 3
lower serum cholesterol level is as d. Trial 4
follows
Trial Drug Sample Difference in 95% CI
no. used size mean 2. Why is the difference in confidence interval
cholesterol
1 A 30 -40 (-118.4) – (38.4)
in drug A in trials 1 and 2?
2 A 3000 -40 (-47.8) – (-32.2) a. Bias
3 B 4000 -02 (-8.5) – (4.5) b. Chance
4 C 5000 -05 (-8.9) – (-1.1)
c. Sample size
d. Wrong calculation
Mean plasma volume
1. What is your interpretation of the results?
a. There is a significance association between
A researcher compared the mean body weight and plasma volume
plasma volume between male and b. There is no association between body
female using t-test, she further weight and plasma volume
conducted linear regression. The c. Cannot tell
table below shows the computer d. Too small sample size to interpret
output for the linear regression of
2. What is the advantage of linear regression on
plasma volume on gender in 8 t-test?
healthy males. e. No added value
f. The t-test shows the association and the its
direction
Plasvol Coefficient Std err t P>[t] 95% CI g. The linear regression shows the association
Gender 0.0436 0.0153 2.857 0.029 0.0063 to 0.0810 and its direction
(Female) h. The advantage cannot be thought from this
Constant 0.0857 1.024 0.084 0.936 -2.420 to 2.591 example
Fasting blood glucose and
HbA1c
A female doctor wanted to predict the
laboratory HbA1c result from the fasting blood
glucoses. On 12 consecutive diabetic patients
she noted the fasting glucose and
simultaneously drew blood for HbA1c. She
compared the pairs of measurements and drew
the graph on the right. Linear regression was
used to estimate this relationship and the
coefficient of determination, R2 is 77%.

1. The correlation coefficient, r ; can be


estimated from the figure to be: 2. How do you interpret the R2 ?
a. The correlation coefficient is – 0.88 a. 77% of the variability can be explained by
b. The correlation coefficient is 0.88 the model
c. The correlation coefficient is 0.12 b. 77% of the HbA1c results can be projected
d. The correlation coefficient is – 0.12 from the fasting blood glucose
Sensitivity and specificity of a new test
1. The test sensitivity is:
We have 1000 patients, a. 10%
100 with a disease and b. 45%
900 without. They are c. 90%
all subjected to a new d. 100%
test and 180 return a 2. The test specificity is:
positive result (of them a. 10%
90 were having the b. 45%
disease) and 820 return c. 90%
a negative result. Using d. 100%
this data answer the 3. The test positive predictive value is:
following questions a. 10%
(question 22 -24)
b. 50%
c. 90%
d. 99%
4. Would you recommend the test?

You might also like