Elderly men with hip fractures were followed to determine the time needed after surgery for them to recover their pre-fracture walking ability. The recovery times had a mode of 9 months, a median of 11.5 months, and a mean of 15.2 months.
The 25th percentile of the recovery time distribution was
9 months.
11.5 months.
less than 11.5 months.
more than 11.5 months.
3.
The standard deviation for the recovery time was 8.3 months. The interval given by the mean plus or minus two standard deviations cannot be used to establish normal limits for recovery time because
the normal curve is not a good model for this distribution.
the standard deviation is too large.
the standard deviation is too small.
the standard error is needed instead of the standard deviation.
Sixty percent of patients with pulmonary dysfunction are smokers and 80% of individuals with normal pulmonary function are non-smokers. The prevalence of pulmonary dysfunction in the population is 30%.
4.
What is the probability of pulmonary dysfunction for smokers?
0.12
0.32
0.56
0.60
0.80
5.
What is the proportion of the total population who are non-smokers with pulmonary disfunction?
0.12
0.32
0.56
0.60
0.80
6.
What is the prevalence of smoking in the population?
0.12
0.32
0.56
0.60
0.80
The performance measures of a screening test can always be interpreted as probabilities.
7.
The predictive value of a positive test is
the probability of testing positive given that the disease is present.
the probability of testing negative given that the disease is absent.
the probability of the disease being present given that the test is positive.
the probability of the disease being present given that the test is positive.
the probability of testing positive given that the disease is present.
the probability of testing negative given that the disease is absent.
the probability of the disease being present given that the test is positive.
the probability of the disease being absent given that the test is negative.
9.
The predictive value of a negative test is
the probability of testing positive given that the disease is present.
the probability of testing negative given that the disease is absent.
the probability of the disease being present given that the test is positive.
the probability of the disease being absent given that the test is negative.
10.
The specificity of the test is
the probability of testing positive given that the disease is present.
the probability of testing negative given that the disease is absent.
the probability of the disease being present given that the test is positive.
the probability of the disease being absent given that the test is negative.
Four screening test are being considered for early detection of Disease X. The sensitivity and specificity of each test is shown below:
Test
Sensitivity
Specificity
Blood Test
80%
70%
Physical exam
60%
90%
Ultrasound
90%
60%
Urinalysis
70%
90%
11.
Of these four tests, which will fail to detect the most cases of disease X in a screened population?
Blood test
Physical exam
Ultrasound
Urinalysis
One thousand people (500 who have colon cancer and 500 who do not have colon cancer) receive a fecal occult blood test (FOBT). The results are shown in the table below:
Based on these data, we can conclude that the specificity of FOBT as a screening test for the detection of colon cancer is closest to
20%
40%
60%
80%
The mean cholesterol level for a sample of cases of women with hypertension during pregnancy was found to be 212 mg%. The lower 95% confidence limit on the population mean was determined as 192 mg%.
13.
The standard error was
5 mg%
10 mg%
15 mg%
20 mg%
14.
The upper 95% confidence limit was
212 mg%
222 mg%
232 mg%
242 mg%
15.
If the mean cholesterol for normotensive pregnant women was 200 mg%, it would be safe to conclude that women with prenatal hypertension have cholesterol levels that are
lower than expected.
higher than expected.
exactly as expected.
perhaps higher or lower than expected.
Volunteers who smoked ten or more cigarettes per day were randomly assigned to use either a nicotine or a placebo inhaler. The subjects were blind to the identity of their treatment. Of the 145 subjects assigned to the nicotine group, 22 eventually quit smoking. There were 7 who quit smoking among the 141 subjects assigned to the placebo group.
16.
The relative risk for quitting, associated with the use of the nicotine inhaler, is
0.32
3.06
3.42
30.6
34.2
17.
Based on a test of the null hypothesis that the quit rates are identical in the two groups, P<0.001, the best interpretation is that
a statistically significant difference in quit rates was attained.
the difference in quit rates could easily be explained by chance alone.
less than one in a thousand trials of this kind will result in a difference as large as that found.
18.
In a subsequent trial, similar subjects were randomized to either a nicotine inhaler or a nicotine patch. A quit rate of 18.1% was attained in the former group and one of 21.6% in the latter, P>0.10. Therefore,
the P-value shows that the difference in quit rates is not clinically significant .
a statistically significant difference in quit rates was attained.
the difference in quit rates could easily be explained by chance alone.
more than one in a ten trials of this kind will result in a difference as large as that found.
The postnatal weight gain (Y) in pounds over a specified period of time was related to varied amounts of a formula supplement (X), in unit doses, taken during the same period for a sample of 70 infants. The following results were obtained:
Regression equation: Y = 1.0 + 0.8 X
Correlation coefficient: r = .64, P<0.05
19.
All of the following statements are reasonable conclusions except
Infants not taking the supplement are expected to gain one pound during the period.
The extra weight gain, attributed to the supplement, is not likely to be the result of random variation in the data.
There is a strong correlation between weight gain and the amount of supplement taken.
An infant taking 0.5 units of the supplement during the time period is expected to gain 1.4 pounds.
An infant taking 0.25 units more of the supplement than another infant during the time period is expected to gain 0.2 punds more
20.
If the weight gain was expressed in grams rather than pounds,
the slope would increase and the correlation coefficient would remain the same.
the slope would decrease and the correlation coefficient would remain the same.
the slope would remain the same and the correlation coefficient would increase.
the slope would remain the same and the correlation coefficient would decrease.
the slope would remain the same and the correlation coefficient would remain the same.
Plasma serum levels of betacarotene (Y) in mg/dl were related to body mass index (X1) and cigarette smoking (X2) for a sample of 200 volunteers with a multiple regression equation. The independent variable X2 took a value of 1 if the volunteer was a current smoker and 0 otherwise. The following results were reported:
Regression equation: Y = 3.1 - 0.10 X1 - 1.0 X2
Coefficient of determination: R2 = .30
21.
Tests of the null hypothesis of zero slope gave P-values less than .05 for both independent variables. Thus a smoker with a body mass index of 20 as compared to a non-smoker with a body mass index of 30 is expected to have a beta carotene level which is
2 mg/dl lower.
1 mg/dl lower.
the same.
1 mg/dl higher.
2 mg/dl higher.
22.
In using the regression analysis given in the previous problem to estimate the effect of cigarette smoking on levels of betacarotene (i.e. mean levels) we must assume
100% of the variance in betacarotene is accounted for by body mass index and smoking.
the effect of smoking is constant over all levels of body mass index.
all individuals in the population will have the same body mass index.
about half of the population will be current cigarette smokers.
all of the above.
23.
The reduction in betacarotene attributed to smoking is likely to be the result of
a low coefficient of determination.
random variation in the data.
the fact that smokers tend to have high body mass.
