24th Annual ACPM Preventive Medicine Board Review Course
Maintenance of Certification
Self-Assessment Questions
-Please mark all of your answers in the CME/MOC Request form.
-This process is for self-assessment; your answers will not be graded.
Reminder: You can only earn CME/MOC for the sessions you attended. Please complete questions only for the specialty (AM, OM, GPM/PH).
Biostatistics
1. Each of the following are examples of what kind of data:
1.1) race
1.2) age
1.3) temperature
1.4) stage of disease
a) nominal
b) ordinal
c) interval
d) ratio
2. If we are doing a study to determine the effect of mother’s age and BMI on infant birth weight, correctly identify the independent and dependent variables.
a) Mother’s age and BMI are the dependent variables and birth weight is the independent variable.
b) BMI is the dependent variable and mother’s age and birth weight are the independent variables.
c) Birth weight is the dependent variable and mother’s age and BMI are the independent variables.
3. What are the mean, median, and standard deviation of the standard normal distribution?
a) 1, 1, 1
b) 1, 1, 0
c) 0, 0, 1
d) 0, 0, 0
e) Cannot be determined without a table
4. Which of the following is not a requirement for the binomial distribution?
a) n independent trials
b) probability of success and failure constant from trial to trial
c) 2 possible outcomes (“success”, “failure”) for each trial
d) probabilities of success and failure equal in each trial
5. Which of the following is not a measurement of central tendency?
a) mean
b) standard deviation
c) mode
d) median
6. Which of the following is not a measurement of dispersion?
a) range
b) standard deviation
c) geometric mean
d) coefficient of variation
7. A random sample of 100 patients attending a diet clinic was found to have lost an average of 30 pounds, with a sample standard deviation of 10. What is the 95% confidence interval for the true mean weight loss for all patients attending the clinic?
a) 27.00, 33.00
b) 10.40, 49.60
c) 29.22, 30.78
d) 28.04, 31.96
8. Students within an elementary school are divided into grade levels and then a random sample is taken from each grade level. This is an example of which kind of sampling?
a) Simple random sampling
b) Cluster sampling
c) Stratified random sampling
d) Systematic sampling
9. Assume that the birth weights of infants are normally distributed with mean of 3405 grams and standard deviation of 225 grams. What is the probability that a randomly selected infant from this population will have a birth weight greater than 3846 grams?
a) 0.05
b) 1.96
c) 0.95
d) 0.025
e) Can’t be determined with the information given
10. We are doing a study to determine if drug A is more effective than a placebo in treating cancer patients. Each patient is only given drug or placebo. The outcome is the proportion of patients who experience a remission. Which are the appropriate null and alternative hypotheses for this study assuming that we are using 2-sided alternative hypotheses?
a) H0: pA=pP; Ha: pA¹pP
b) H0: pA=pP; Ha: pA>pP
c) H0: pA¹pP ; Ha: pA=pP
11. We are doing a study to determine if males and females have different baseline cholesterol levels. What are the null and alternative hypotheses to look for any difference in these means assuming 2-sided hypotheses?
a) H0: mM > mF HA: mM ¹ mF
b) H0: mM = mF HA: mM > mF
c) H0: mM ¹ mF HA: mM > mF
d) H0: mM = mF HA: mM ¹ mF
12. Assume you gather data, compute a test statistic, and reject the null hypothesis of no difference.
12.1) If, in fact, the null hypothesis is true, you have made a ______.
12.2) If, in fact, the null hypothesis is false, you have made a ______.
a) Type I error
b) Type II error
c) Type I and Type II error
d) Correct decision
13. Assume you gather data, compute a test statistic, and fail to reject (accept) the null hypothesis.
13.1) If, in fact, the null hypothesis is true, you have made a ______.
13.2) If, in fact, the null hypothesis is false, you have made a ______.
a) Type I error
b) Type II error
c) Type I and Type II error
d) Correct decision
14. Consider the following two distribution curves. Which numerical summary measure would allow you to discriminate between the two distributions?
a) Median
b) Mean
c) Mode
d) Standard Deviation
15. We are doing a study comparing pre and post test scores for 2000 students after they have completed a physical training program. We conducted a paired t-test. Suppose we find that the test statistic is 3.50. What conclusions should you draw for a=0.05?
a) Reject the null hypothesis. There is a significant difference between pre and post test
scores.
b) Don’t reject the null hypothesis. We cannot state that there is a significant difference
between pre and post test scores.
c) Reject the null hypothesis. We cannot state that there is a significant difference between
pre and post test scores.
d) Don’t reject the null hypothesis. There is a significant difference between pre and post
test scores.
16. We compared the BMI of mothers who exercised during pregnancy to the BMI of mothers who did not exercise during pregnancy. We found that the 95% CI for the difference in BMI between the exercising and non-exercising mothers was (-0.5, 0.5). What conclusions can we draw?
a) Reject the null hypothesis. There is a significant difference between exercising and non- exercising mothers.
b) Don’t reject the null hypothesis. We cannot state that there is a significant difference between exercising and non-exercising mothers.
c) Reject the null hypothesis. We cannot state that there is a significant difference between exercising and non-exercising mothers.
d) Don’t reject the null hypothesis. There is a significant difference between exercising and non-exercising mothers.
e) We need more information to answer the question.
17. In a sample of low back pain patients, we asked each patient to rate his pain on a scale from 1 to 50. There are 10 patients with sciatica and 15 patients without sciatica. Pain score, sciatica, age, and type of provider are recorded in the following table for 5 patients from the study group. We are interested in whether or not pain score differs by whether or not the patient has sciatica.
PATIENT ID / AGE / PROVIDER TYPE / SCIATICA / Pain Score1 / 57 / Primary Care / No / 2
2 / 24 / Chiropractor / No / 3
3 / 69 / Primary Care / Yes / 13
4 / 42 / Orthopedist / Yes / 18
5 / 53 / Orthopedist / No / 7
17.1 The variance of the pain score is 49.0. What is the standard deviation?
a) 49
b) 9.8
c) 7
d) 1.4
17.2 What is the standard error?
a) 9.8
b) 7
c) 1.4
d) 0.28
17.3 How could you visually display the distribution of the pain scores?
a) Bar graph
b) Scatterplot
c) Histogram
d) 2x2 table
17.4 Which of the following would not be a way of expressing the number of primary care provider patients in this sample?
a) Median
b) Frequency
c) Percent
d) Proportion
17.5 There is evidence (from previous studies) that pain increases with age. What would be the best way visually to assess this is this sample?
a) Scatterplot
b) Histogram of age by categories of pain scores
c) Histogram of pain scores by categories of age
d) Pie chart
18. We are interested to see if the risk of pre-term birth differs for women with pre-eclampsia as compared to women without pre-clampsia. We determine that the odds ratio for pre-term birth comparing women with pre-eclampsia to those without pre-eclampsia is equal to 2.55 (95% CI: 1.35, 3.85) and the p-value from a chi-square test is equal to 0.025. What conclusion can we draw?
a) Women with pre-eclampsia have a significantly higher risk for preterm birth because the p-value is <0.05 and the 95% CI does not include 1.
b) Women with pre-eclampsia have a significantly higher risk for preterm birth because the p-value is <0.05, but the 95% CI does not agree with this result.
c) There is no significant difference between these two groups with regard to risk for pre-term birth.
19. We did a study among children and determined that the correlation (r) between BMI and
hours of television watched was 0.24 with a p-value=0.02. What conclusions can we draw?
a) Children who watch a lot of television are significantly more likely to have a higher
BMI.
b) Although statistically significant, television watching accounts for only about 6% of the
variation in BMI among children in this study.
c) There is a weak correlation between television watching and BMI.
d) Both B and C.
e) None of the above are appropriate conclusions.
20. For each of 100 patients with emphysema, a clinical researcher tabulated the number of years the patient smoked and the attending physician’s subjective evaluation of the extent of lung damage (measured on a scale of 0 to 100). She determined that the regression equation describing the relationship between the number of years the patient smoked (X) and the extent of lung damage (Y) was: Y = 11.24 + 1.31 X
20.1 Which of the following statements is NOT true?
a) The slope of the regression line equals 1.31.
b) The correlation coefficient, if computed for the above data, would be positive.
c) Patients who smoked less on average than others had lower lung damage scores.
d) The regression line crosses the X-axis at 11.24.
20.2 What is the predicted extent of lung damage for a patient who has smoked for 10 years?
a) 22.24
b) 24.34
c) 10.0
d) This cannot be determined from the data
21. Grade point averages (GPA) at the end of the first two years of medical school are used to predict scores on Part I of the National Board Examination. Data was collected on medical students across the United States during the past academic year.
21.1 The correlation coefficient between the exam score and GPA is r=0.75. How could this be interpreted?
a) There is a negative linear relationship between exam score and GPA,
i.e., lower exam scores are associated with lower GPAs
b) There is a negative linear relationship between exam score and GPA,
i.e., lower exam scores are associated with higher GPAs
c) There is a positive linear relationship between exam score and GPA,
i.e., higher exam scores are associated with lower GPAs
d) There is a positive linear relationship between exam score and GPA,
i.e., higher exam scores are associated with higher GPAs
21.2 What would not be the null hypothesis for the regression equation to test whether there is a statistically significant linear relationship between exam score and GPA?
a) bGPA = 0
b) GPA predicts exam score
c) Slope of the equation for the line is zero
d) There is no linear relationship between GPA and exam scores
21.3 Suppose that the slope is 132.2. What is interpretation of this slope?
a) As GPA increases by 1 point, test scores increase by 132.2 points.
b) As test scores increase by 132.2 points, GPA increases by 1 point.
c) There is no relationship between test score and GPA.
d) As GPA increases by 1 point, test scores decrease by 132.2 points.
22. Match the appropriate statistical test with the scenario.
22.1 In a sample of 100 subjects we are comparing time to walk one mile before and after drinking a cup of coffee.
22.2 We are comparing customer satisfaction scores (on a scale of 1 to 5) before and after a new billing system is added.
22.3 We would like to compare two treatments for migraine, Drug A and Drug B. Patients are randomized to receive different treatments. We want to compare the proportion of patients reporting migraines between the two drugs.
22.4 A study is designed to compare cholesterol levels in patients who took a multivitamin as compared to patients who did not take a multivitamin.
22.5 We would like to examine the relationship between gestational age and birth weight.
22.6 We would like to determine if there is a difference in time to remission between patients treated with a new type of chemotherapy and patients given the standard treatment.
22.7 We would like to determine if the odds of having a c-section delivery differ between by maternal age and race.
22.8 We want to compare the average weight loss among women who were randomly assigned to four different diets.
22.9 We would like to examine the relationship between blood pressure and cholesterol level while controlling for age.
a) Multiple linear regressions
b) Analysis of variance
c) Survival analysis
d) Simple linear regression or correlation
e) Student’s t-test
f) Chi-square analysis
g) Wilcoxon signed rank test
h) Paired t-test
i) Logistic regression
Epidemiology
Prostate cancer is the most common cancer in American men and has the third highest cancer mortality rate. There has been no improvement in the age-adjusted death rate from this disease since 1949. Potential areas for improvement include primary prevention (identification and control of risk factors), secondary prevention (early diagnosis through screening), and tertiary prevention (improved treatment).