-
A standard screening test for a specific genetic condition is based on a combination of maternal age and the level of serum A. Using this test 80% of these cases can be identified at birth, while 5% of normals are detected as positive.
-
What is the sensitivity and specificity of the test?
-
Assume that 1/500 births have this genetic condition. What is the percentage of births who test positive using this test will actually have the condition?
-
-
Disease
No Disease
Positive Test
80%
5%
Negative Test
20%
95%
-
To test the primary hypothesis of a study, infants were categorized by gestational age and were designated as having a severe disease if they had a concentration of a certain hormone which was 2.6 sd below the mean score for the assay from their specimen. Assume also that these assays are done in batches of 240 specimens and that a mean and sd were calculated for each batch based on a sample size of 240. Children in the study were given a standard mental development test at <30 months of age. The results are shown in the table below:
-
Severe Disease
Mean score +/ sd
n
No
106 +/- 21
138
Yes
88 +/- 25
17
-
Perform a test to compare the mean score of the developmental test between children with and without the severe disease (report a p-value).
-
Suppose that we wanted to use data on children of all gestational ages in the study. Suggest a type of analysis that could be used to relate the score on the developmental test to severe disease while controlling for age. (just suggest an analysis don’t actually do it)
-
The following table represents blood pressure recordings on one participant for 10 consecutive days with two readings per day:
-
Reading
Day
1
2
1
98
99
2
102
93
3
100
98
4
99
100
5
96
100
6
95
100
7
90
98
8
102
93
9
91
92
10
90
94
-
Estimate the between-day and within-day components of variance for this participant.
-
Is there a difference in underlying mean blood pressure by day for this participant?
-
Suppose that we had a study that looked at the risk of cavities in 3 different communities according to whether their drinking water had “higher” fluoride concentration as determined by water samples. The table below shows the data (collected over 5 years) on the number of cavities between the higher fluoride vs control for people ages 20-35 and 55-80.
Ages 20-35 |
# with cavities |
Total number |
Ages 55-80 |
# with cavities |
Total number |
Control |
3 |
37 |
Control |
11 |
121 |
Higher fluoride |
1 |
33 |
Higher fluoride |
21 |
148 |
-
What test can be used to compare the cavity rates in these two communities while controlling for age?
-
Implement the test and report a p-value.
-
Estimate the OR relating higher fluoride and cavities while controlling for age.
-
Provide a 95% confidence interval for the OR in part (c).