EBM Multiple Choice Questions

For each of the variables listed below, indicate which is discrete and which is continuous: - Number of beds in a hospital - Height in inches - Number of pregnancies a woman has had - Amount of time spent sleeping - Body temperature measured in Fahrenheit degrees - Self-esteem as measured on a 10-question scale - Annual income

- Number of beds in a hospital *discrete* - Height in inches *continuous* - Number of pregnancies a woman has had *discrete* - Amount of time spent sleeping *continuous* - Body temperature measured in Fahrenheit degrees *interval* - Self-esteem as measured on a 10-question scale *discrete* - Annual income *continuous*

A study wishes to assess birth characteristics in a population. Which of the following variables describes the appropriate measurement scale? - birth weight in grams - Birth weight classified as low, medium, high - Type of delivery classified as cesarean, natural, induced - Type of delivery classified as alive and dead

- birth weight in grams *continuous* - Birth weight classified as low, medium, high *ordinary* - Type of delivery classified as cesarean, natural, induced *nominal* - Type of delivery classified as alive and dead *nominal, dichotomous, binary*

A study is conducted to determine the effects of 3 different medications on weight loss (Drug A, Drug B, and Drug C). Patients are randomly assigned to each medication for 12 weeks, and weight is measured at baseline and at the end of 12 weeks. In addition, the study seeks to determine if patient gender has an effect on weight loss as well. What statistical test would be used to determine the effects of both drug therapy and gender on weight loss at the end of 12 weeks?

2 way ANOVA

If the study desired to determine the effect of sodium intake (high/low) and drug therapy simultaneously on blood pressure, what statistical test would be used?

2-way ANOVA

Which of the following statistical tests would provide the most appropriate? a. Multiple regression b. Logistic regression c. Cox proportional hazards d. Kaplan-Meier To estimate the effect of packs of cigarettes smoked during pregnancy on the birth weight while controlling for the mother's age, parity, and pregnancy weight

A

A study comparing two statins finds a mean LDL of 152 mg/dl in the lovastatin group and 165 mg/dl in the newstatin group (p=0.09). How should these results be interpreted? A. Depends on the study power B. Both are equally effective C. Lovastatin and new statin can be used interchangeably D. Setting the p-value at <0.10 would be more useful

A. Depends on the study power

What would be the appropriate inferential statistical test that would be used to compare hemoglobin A1C between two groups, one receiving standard diabetes care and the other cared for by a pharmacist diabetes educator? A. Independent t-test B. Wilcoxon signed rank test C. Paired t-test D. Chi-squared test E. None of the above

A. Independent t-test

A physician at HMC wants to start all of his patients admitted to the ICU on a proton pump inhibitor (PPI). He states a recent study of hospitalized patients found that PPI improve survival when used for stress ulcer prophylaxis. You as a student pharmacist realize this will increase the pharmacy's budget considerably and want to evaluate the evidence. Which one of the following is true? A. The inclusion and exclusion criteria should be evaluated to determine whether the patient population is similar to your patient population before using the study results to make a decision. B. If a significant difference is found between the two groups, then the possibility of a type II error must be considered. C. The methods used to blind the study are not important because the outcome was objective. D. It is not important to review the competency of the authors by reviewing their affiliations and credentials

A. The inclusion and exclusion criteria should be evaluated to determine whether the patient population is similar to your patient population before using the study results to make a decision.

Dr. Supakit conducted a study to compare Drug A and Drug B for treating N/V associated with pregnancy. Drug A has been used for many years and has a large evidence base showing efficacy and safety. Drug B was recently introduced to treat N&V associated with chemotherapy, but has not been studied by pregnant patients with N/V. Patients were randomized to one of these 2 drugs. The primary outcome was patients' ranking of their N/V 3 hours after taking the drug. N/V was graded using the following scale: 0=no nausea, 2=moderate nausea, 3=severe nausea, and 4=vomiting. Which is the best statistical test to test for differences between two drugs? A. Wilcoxon rank sum test B. Wilcoxon signed rank test C. Student's t-test D. Paired student's t-test

A. Wilcoxon rank sum test

You are performing a pharmacokinetic study in which you plan to measure Cmax and AUC of a new drug alone and in combination with a CYP450 enzyme inhibitor in a crossover design. What is the best statistical test for these parameters? Assume data are non-parametric A. Wilcoxon signed rank test B. Chi-squared test C. Student's t-test D. Paired student's t-test E. None of the above

A. Wilcoxon signed rank test

Investigating blood cholesterol levels in patients who follow a diet either low or moderate in fat and who take either a drug to lower cholesterol or a placebo

Analysis of Variance (ANOVA)

Which of the following statistical tests would provide the most appropriate? a. Multiple regression b. Logistic regression c. Cox proportional hazards d. Kaplan-Meier To estimate the risk of death during a hospital stay associated with previous history of heart disease while controlling for age, sex, and socioeconomic status

B

A study tests whether the use of a new drug adherence aid has an effect on CHF exacerbations. Researchers want to evaluate the proportion of patients with an exacerbation of their CHF symptoms in a group of people who used the adherence aid compared with a group of people who did not. To assess whether the changes observed are statistically significant, which one of the following tests is best? A. McNemar B. Chi-square C. Independent t-test D. Mann-whitney U-test E. None of the above

B. Chi-square

What would be the appropriate inferential statistical test that would be used to compare the percentage of patients experiencing a breast cancer associated with Drug A versus Drug B? A. Independent t-test B. Chi-square test C. One-way ANOVA D. Paired t-test E. None of the above

B. Chi-square test

A study is currently designed to detect a 15%difference in response rates between two treatments for breast cancer with a p<0.05. If the investigators wish to be able to detect a 10% difference, which of the following changes do they need to make? A. Decrease the sample size B. Increase the sample size C. Select an level of 0.01 as the cutoff for statistical significance D. Select an level of 0.001 as the cutoff for statistical significance

B. Increase the sample size

A new drug, Superstatin, is indicated for treating dyslipidemia; however, the FDA is concerned that this drug may cause liver toxicity. There were a few reports of liver toxicity before the drug came to the market; however, not enough reports existed to conclude an increased risk of liver toxicity. Which one of the following study designs is best suited to identify whether an increased risk of liver toxicity exists with superstatin? A. Randomized controlled clinical trial B. Prospective cohort study C. Case-control study D. Cross-sectional study

B. Prospective cohort study

Censoring in a prospective study whose end point is mortality occurs when A. Subjects die before the planned termination of the study B. The study ends before all subjects have died C. Subjects are excluded from the study because of the presence of a particular factor or complication D. Subjects die and the cause of death is unknown E. Subjects fail to comply with their assigned intervention

B. The study ends before all subjects have died

Which one of the following describes the most important limitation of composite outcomes? A. They allow the counting of many events in the same patient. B. They assume that all outcomes are equally important. C. There is insufficient power to detect treatment effects on the individual components of the composite outcome. D. They are unable to classify all outcomes precisely. E. None of the above

B. They assume that all outcomes are equally important.

The median lung area of the 50 infants was 11.23 cm2 (range 0.82-28.53) and their median FRC was 28 mL (range 3-103). Lung area correlated significantly with lung volume (Spearman's correlation coefficient = 0.60, P<0.001). No other analysis or graph was given. Why did the authors use the Spearman rank correlation coefficient?

Both measured lung area and lung volume have positively skew distributions, as shown by the median being nearer to the lower end of the range than to the upper. Thus neither variable follows a Normal distribution. For a correlation coefficient we need normal data. They could either use a transformation, such as the logarithm, or use a statistical method which is unaffected by skewness. They chose the Spearman rank correlation coefficient

To determine if fasting is associated with dengue fever, data from 60 patients with dengue fever were collected. These patients were matched for age, sex, and race to 40 patients without dengue fever. The hospital charts of these patients were then reviewed to determine whether they also fasted prior to their illness. What is the type of study design? A. Cross-sectional study B. Prospective cohort study C. Case-control study D. Retrospective cohort study E. Randomized clinical trial

C. Case-control study

Researchers evaluated the percentage of patients who achieved less than a target BP (<140/90 mmHg) when initiate on two different doses of amlodipine. In the study of 200 patients, the amlodipine 5-mg group (n=100) versus the amlodipine 10-mg group (n=100) were compared. The researchers used the percentage of patients who achieve the target goal at 6 months as their primary outcome. Which is the best statistical test to answer the researcher question? A. Independent t-test B. Wilcoxon signed rank test C. Chi-square or Fisher's exact D. Wilcoxon rank sum E. None of the above

C. Chi-square or Fisher's exact

You are on the advisory board of a pharmaceutical company that has developed a new drug for stroke prevention in AF. The company decided to conduct a non-inferiority trial to compare the new anticoagulant (Drug A) with warfarin (standard treatment). Which one of the following best describes what you and the researchers set out to prove the alternative hypothesis? A. Drug A is better than warfarin by a factor of delta B. Drug A is worse than warfarin by a factor of delta C. Drug A is not worse than warfarin by a factor of delta D. Drug A is the same as warfarin

C. Drug A is not worse than warfarin by a factor of delta

The median survival time of children with leukemia treated with a combination of drugs and radiation is reported to be 28.2 months. For such patients A. The mean survival time is less than 28.2 months B. The longest possible survival time is 56.4 months C. Half will survive less than 28.2 months D. The modal survival time is greater than 28.2 months E. At least 95% will survive beyond 28.2 months

C. Half will survive less than 28.2 months

You are evaluating an observational study to assess the effect of beta blocker therapy has on time to cardiac event in hemodialysis patient for a 5-year study period. The outcome variable is whether patients suffer a cardiac event, and the IV is whether they initiate beta blocker therapy or not. In addition, the analysis is statistically adjusted for several important covariates including age, gender, race, years with kidney failure, comorbid conditions, and concurrent medications. The authors conducted a multivariate logistic regression analysis. Was this the appropriate statistical test? A. No. Observational studies are never correlational in nature B. No. the authors should have conducted a linear regression analysis C. No. The authors should have conducted a Cox regression analysis D. Yes. Multiple logistic regression analysis was correct E. None of the above

C. No. The authors should have conducted a Cox regression analysis

A study is comparing nicotine patches and nicotine gum. The primary outcome is the rate of smoking cessation (i.e., proportion of subject who can quit). What is the most appropriate study design to determine which treatment is more effective? A. Factorial design B. Crossover design C. Parallel-group design D. Case-control study E. None of the above

C. Parallel-group design

A study is comparing nicotine patches and nicotine gum. The primary outcome is the rate of smoking cessation (i.e., proportion of subject who can quit). What is the most appropriate study design and statistical test to determine which treatment is more effective? A. Parallel-group; t-test B. Crossover; chi-square test C. Parallel-group; chi-square or Fisher's exact D. Crossover; paired t-test

C. Parallel-group; chi-square or Fisher's exact

A study is comparing bupropion and varenicline. The primary outcome is the rate of smoking cessation (i.e., proportion of subject who can quit). What is the most appropriate study design and statistical test to determine which treatment is more effective? A. Parallel-group; t-test B. Crossover; chi-square test C. Parallel-group; chi-square test D. Crossover; paired t-test E. None of the above

C. Parallel-group; chi-square test

A study is conducted to compare the clinical efficacy of 2 antibiotics for the treatment of pneumonia. The patients are randomly assigned to receive either Drug A or Drug B. The main dependent variable is mortality over a 14 day period. The authors provide the following information: Alpha < 0.05; Beta = 0.2; Delta = 4% (for mortality); Std Proportion = 10%; N = 50 patients/group The results indicate that mortality in patients who received Drug A was 12% compared to Drug B which was 14%. The p value for this comparison is p = 0.15. Assume that a reasonable clinical difference between groups in terms of mortality is 5% Which of the following is a TRUE statement: A.The difference in antibiotic mortality between the groups was 2%, and indicates that clinical significance was achieved B.There is sufficient statistical power to detect a reasonable clinical difference but the standard proportion does not support sufficient statistical power C.The statistical power is adequate to detect a reasonable clinical difference between the groups D. None of the above

C. The statistical power is adequate to detect a reasonable clinical difference between the groups

A clinical trial is designed to compare INR measurements during treatment with brand-name and generic warfarin. Mean INR during brand-name treatment was 2.45, and mean INR during generic treatment was 2.51 (p=0.0001; paired t-test) How should the results be interpreted? A. There is a clinically significant difference between the two formulations B. There is a 0.01% chance that a Type II error has occurred C. There is a statistically significant difference between the two treatments D. No conclusion can be drawn because the wrong statistical test was used

C. There is a statistically significant difference between the two treatments

READ CASE B for the next 3 questions

CASE B. The following 3 questions are based on CASE B A study is developed to compare the efficacy in lowering diastolic blood pressure (DBP) between 2 different antihypertensives. The authors provide the following information: Alpha < 0.05 Beta = 0.2 Power = 80% Delta = 5 mm Hg (also known as the expected effects size) Sigma = 10 mm Hg N = 120 patients/group The study results indicate that on average, Drug A can reduce BP by 20 mm Hg, and Drug B can reduce BP by 8 mm Hg. The p value for this comparison is p = 0.01. NOTE: Assume that a reasonable clinical difference in blood pressure reduction is 5 mm Hg between groups and that the population variance is expected to be 10 mm Hg.

READ CASE C for the next 3 questions

CASE C. The following 3 questions are based on CASE C A study is conducted to compare the clinical efficacy of 2 antibiotics for the treatment of urinary tract infections. The authors provide the following information: Alpha < 0.05 Beta = 0.2 Delta = 25% (difference in antibiotic cure rates) N = 20 patients/group The results indicate that Drug A was efficacious in 92% of the patients, and Drug B was efficacious in 86% of the patients. The p value for this comparison is p = 0.33. NOTE: Assume that a reasonable clinical difference between groups in terms of antibiotic efficacy is 10%

READ CASE D for the next 9 questions

CASE D. The following 3 questions are based on CASE D A study comparing the efficacy of two H2 blockers in the treatment of patients with duodenal ulcers was conducted. The study determined that Drug A had a 90% cure rate, and that Drug B had an 84% cure rate. The following information is supplied: (Curediff) = 6% (Curediff = Drug A ‐ Drug B = 90% ‐ 84%) (95%CIdiff) = +2%, +10% (95%CIdiff = 95% confidence interval for the difference in ulcer cure rates)

READ CASE E for the next 11 questions

CASE E. The following 4 questions are based on CASE E An analgesic medication was compared to an identical appearing placebo by random assignment to separate treatment groups. Pain was measured on a scale where 0 = none, 1 = mild, 2 = moderate and 3 = severe. Pain was measured at baseline to determine if the two groups were similar. Before the data were collected, the investigators stated that the level of pain 8 hours after a single dose of analgesic was the primary end point. Nevertheless, they measured pain every hour after the dose for 8 hours.

Which of the following statistical tests would provide the most appropriate? a. Multiple regression b. Logistic regression c. Cox proportional hazards d. Kaplan-Meier To determine the effect of age on the time to recover from fracture of the hip

D

A randomized, double=blinded, parallel trial compare three hypnotic drugs for their effect on sleep latency (measured as the mean time to sleep onset) and finds that drug A is better than drug B (p<0.05) and that drug C is better than drug A (p<0.01). The investigators used a student t-test to test the hypotheses that each drug was equal to the other. Which one of the following is TRUE? A. Investigators used the appropriate statistical test to analyze their data B. Drug C is the most effective of these drugs C. The statistical methods used bythe investigators likely resulted in a type II error D. ANOVA would have been a better hypothesis test for their data E. None of the above

D. ANOVA would have been a better hypothesis test for their data

Researchers designed a study looking at cardiovascular deaths comparing a new drug to placebo. They determined that they would need 200 patients in each group to detect a 15% difference in cardiovascular end points given 90% power and a significance level of 0.01. Which of the following changes would require the researchers to increase their sample size? A. Aim to detect a difference of 20% B. Specify a power of 80% C. Use a significance level of 0.05 D. Aim to detect a difference of 10% E. None of the above

D. Aim to detect a difference of 10%

Dr. Supakit conducted a cohort study to determine the effect of fat intake (i.e., High vs.Low) on the development of breast cancer (i.e., Yes vs.No). He knew that types of fat intakes (i.e., monounsaturated vs. polyunsaturated) might influence on the development of breast cancer and any woman in both group was exposed to types of fats. Which statistical test would be the most appropriate to determine whether fat intake is a risk factor for the development of breast cancer, while taking types of fat into account as a confounding factor? A. Independent t-test B. Chi-square test C. McNemar's test D. Cochran Mantel-Haenszel test E. None of the above

D. Cochran Mantel-Haenszel test

You are performing a pharmacokinetic study in which you plan to measure Cmax and AUC of a new drug alone and in combination with a CYP450 enzyme inhibitor in a crossover design. What is the best statistical test for these parameters? A. Wilcoxon signed rank test B. Chi-squared test C. Student's t-test D. Paired student's t-test

D. Paired student's t-test

An analysis was conducted to determine the impact of drug therapy on blood pressure (measured in mm Hg) using a crossover design. Drug A was given to all patients in the first period and measurements were taken, a washout occurred, and then Drug B was given to the same patients and measurements were again taken. What would be the appropriate statistical test to determine whether the mean blood pressures after taking Drug A were different from the mean blood pressures after taking Drug B (assume the data are parametric and homoscedastic)? A. Wilcoxon signed rank test B. Chi-squared test C. Independent t-test D. Paired t-test E. None of the above

D. Paired t-test

In a meta-analysis of RCT examining the effects of antihypertensive drugs, investigators found that the OR for treatment with low-dose diuretics compared with beta-blockers for CV disease events was 0.86 (95%CI, 0.77-0.97). Which one of the following statements is the most appropriate interpretation of these findings? A. Treatment of hypertension with low-dose diuretics was 14% more effective in preventing CV disease events than treatment with beta-blockers. B. Treatment of hypertension with beta-blockers was 14% more effective in preventing CV disease events than treatment with low-dose diuretics. C. The difference observed between treatment with beta-blockers and low-doses of diuretics was not statistically significant. D. The odds of developing CV events when treating hypertension with low-dose diuretics was lower than treatment with beta blockers

D. The odds of developing CV events when treating hypertension with low-dose diuretics was lower than treatment with beta blockers

Many cardiovascular trials now use composite outcomes such as hospitalization for heart failure, stroke, MI, or CV death. Which one of the following best describes why composite outcomes are useful? A. They can provide more data on adverse effects. B. They add together outcome events for a single patient to form a single end point for a study. C. They reduce the complexity of describing treatment effects. D. They can reduce sample size requirement. E. None of the above

D. They can reduce sample size requirement.

The p-value for nicotine clearance was reported as follows: Clearance: p < 0.01 (CO, A > CS) What kind of the test was used to determine CO, A > CS.? A. Chi-square test B. Wilcoxon's Rank Sum C. Paired t-test D. Tukey's test E. None of the above

D. Tukey's test

What statistical test would be used to provide this p value?

Data parametric - Paired t test Data nonparametric or ordinal data - Wilcoxon Signed Rank

Two new investigational drugs A and B were tested in 20 patients with skin disease. Each patient received only one of the two drugs selected by random. Ten patients received B and 10 received A. At the end of the trial, dermatologists examined the patients and rank ordered them from best to worst. What is the most appropriate statistical test to use to compare the differences between two drugs? A. Independent t-test B. Chi-square test C. Kruskal-Wallis H test D. Paired t-test E. Mann-Whitney U test

E. Mann-Whitney U test

What is the log-rank test for and what feature of the data makes the authors use this technique?

It is a method of comparing survival in two groups of patients. It allows for the possibility that some subjects have not yet died and that subjects have been followed for varying lengths of time

A study is conducted to determine the effects of professor office hours on students' satisfaction with their education. The following data is collected: Student satisfaction (yes/no) prior to office hours and student satisfaction after office hours. What would be the appropriate statistical test to determine differences in satisfaction before and after office hours?

McNemar's Test

Investigating physical functioning in patients with diabetes measure as a continuous variable on the basis of demographic characteristics and level of diabetic control

Multiple Regression

Is the result statistically significant? Explain why or why not

No, the results are not statistically significant as zero is included in the interval, indicating that there is a possibility of accepting the null hypothesis

Investigating a possible association between exposure to an environmental pollutant and miscarriage

Odds Ratio

If the study wanted to determine the differences in blood pressures at all visits for both drug groups, what statistical test would be used?

Repeated Measures of ANOVA

Comment on the statistical and clinical significance of the resul

Statistically insignificant and clinical significance should not be determined

Define the dependent variable(s) for this study

Student Boredom

Define the independent variable(s) for this study

Student age, gender, previous education, personality profile, socioeconomic background, ethnicity

If the data were homoscedastic and parametric, what test would be used to calculate the p value?

Student's T test

If the study wanted to determine differences in blood pressure at ONLY Week 3, what would be the appropriate statistical test to use?

Student's T test

What statistical test would be used to determine differences in blood pressure at baseline?

Student's t-test

The self-assessment was recorded into two categories, "cleared or pronounced improvement" or "little or no improvement". What statistical tests could be used to carry out the significance test for the recoded data?

The data are paired and dichotomous. We could use McNemar's test to compare the proportions who had improved for the two treatments.

What can we conclude?

The response was better using calcipotriol ointment than using betamethasone, though we are not told by how much. As this was a randomized double blind trial, the evidence suggest that the ointment itself was the cause

Determine if the study has sufficient power to detect a reasonable clinical difference between the groups

The study has sufficient power to detect a reasonable clinical difference, but only if the sigma for the population is small (5 mm). Since the est population difference is 10mm, the study may not be adequately powered to detect a 5mm difference between groups if the std dev is 10mm or larger. The around delta of 5 mm is higher than the SD the authors used (SD = 5 mmHg) make the power goes up, but power also goes down for the expected SD in the population of 10 mmHg. Therefore, we don't know where the power ends in this case.

What question does the test try to answer and what conclusion would you draw?

The test tried to determine the evidence for a difference in survival between patients with and without P53 antibodies in breast cancer patients as a whole. The P value is very small and so we have strong evidence that patients without P53 antibodies survive longer than patients wit P53 antibodies in the population from which the sample was taken.

Rejecting the null commits what type of error?

Type 1 error

Accepting the null commits what type or error?

Type 2 error

Interpret the 95% CIOR for the data and determine if it is statistically significant

We are 95% confident that the true odds ratio for the population lies within the interval 1.5 and 4.0 (male are more likely to be bored than females, and this is statistically significant as the interval does not include 1.0)

What is the correct interpretation of the 95% confidence interval for this study?

We are 95% confident that the true population difference in cure rates between patients receiving Drug A vs those receiving Drug B lies between the interval of -1%, +13%

What test would be used to provide this p value?

Welch's Test

Eighty‐seven patients with bronchioalveolar cancer, a rare condition were compared to 286 non cancer patients and 297 patients with other types of cancer, matched by age, sex and race. They were interviewed about their history of cigarette smoking. A strong relationship was observed between smoking and alveolar cancer (Morabia and Wynder 1992). This is an example of a a. Case control study b. Cohort study c. Cross‐sectional study

a. Case control study

A study is conducted prospectively to determine the effects of Drug A versus Drug B in preventing myocardial infarctions (MIs). One hundred patients in each group were treated with either Drug A or Drug B, and the following results were obtained over a 10 year period: Drug A = 7% of patients had an MI Drug B = 10% of patients had an MI What would be the appropriate inferential statistical test that would be used to compare the percentage of patients experiencing a myocardial infarction associated with Drug A versus Drug B? a. Chi‐square analysis b. One‐way ANOVA c. Independent student t‐test d. None of the above

a. Chi‐square analysis

In a clinical trial, 60 people received a new drug and 50 people received a placebo. Of the people on the new drug, 40 of the 60 showed a positive response and 25 of the 50 people on placebo showed a positive response. What is the most appropriate statistical test to use in determining if the new drug shows more of an effect than the placebo? a. Chi‐square analysis b. Mann Whitney U test c. Independent student t‐test d. None of the above

a. Chi‐square analysis

The use of inferential statistics permits the researcher to: a. Generalize to a population based on information gathered from a sample b. Interpret descriptive statistics c. Describe information obtained from empirical observation d. Reject the null hypothesis e. Select a level of significance

a. Generalize to a population based on information gathered from a sample

If you decided to reject the null hypothesis when it was false you a. Made a correct decision b. Made both a type I and type II error c. Made a type I error d. Made a type II error

a. Made a correct decision

Which of the following research designs would be the MOST optimal in terms of controlling the study and producing valid results for a clinical trial comparing two drugs? a. Prospective, randomized, double‐blind parallel study b. Prospective experimental with historical control, open‐label, comparative study c. Observational, double‐blind parallel study d. Retrospective, parallel, longitudinal, open‐label study e. A and C

a. Prospective, randomized, double‐blind parallel study

A study was conducted to evaluate an antibiotic in the treatment of possible bacteremia. Five hundred children with fever but no focal infection were randomly assigned to the antibiotic or to a placebo. All patients were reevaluated after 48 hours. The design used in this study is best described as a a. Randomized clinical trial b. Placebo‐controlled trial c. Controlled clinical trial d. Cohort study

a. Randomized clinical trial

A group of pharmacists have successfully implemented a new, evidenced-base protocol around medical intervention for ACS patients. How should the pharmacists follow up this success? a. Tell others about the results of their project b. Elicit feedback from patients and other stakeholders c. Evaluate the costs of the change in practice d. Search the literature for research that relates to the practice change

a. Tell others about the results of their project

When p‐values are statistically significant, the following conclusions can be reached a. The amount of Type II error is not evaluated because the null hypothesis is rejected and a difference between groups exists b. The amount of Type I error is excessive. c. The test might not have had enough power to produce a statistically significant result d. None of the above

a. The amount of type II error is not evaluated because the null hypothesis is rejected and a difference between groups exists

The investigator presented a statistical test that was appropriate for the level of measurement to compare the pain in the two treatment groups at baseline. The pain measurements within each group were not identical, but the p‐value for the baseline comparison was 0.22. Which of the following could you conclude from this information? a. The baseline differences in pain between the two groups could be due to sampling variation b. The differences at baseline would not cause the groups to be difference at 8 hours c. The study had inadequate power to compare pain after 8 hours d. The likelihood that the observed differences would occur due to sampling variation was less than 5%.

a. The baseline differences in pain between the two groups could be due to sampling variation

Which of the following is a TRUE statement (based on data from CASE B)? a. The difference in DBP is statistically significant because the amount of Type I error is less than alpha (the level of significance) b. The amount of Type II error is excessive because the sample size is too small c. Statistical power could only detect differences of 12 mm Hg or larger between groups (with 80% power) d. B and C e. None of the above

a. The difference in DBP is statistically significant because the amount of Type I error is less than alpha (the level of significance)

Which of the following is a TRUE statement (based on data from CASE C)? a. The result indicates a statistically insignificant finding, and the null hypothesis is accepted b. The result is statistically insignificant and clinically significant c. The p value indicates that there is insufficient power to evaluate the data d. A and B e. All of the above

a. The result indicates a statistically insignificant finding, and the null hypothesis is accepted

When we test H0 : mu1 - mu2 = 0 vs. H1 : mu1 - mu2 doesNOT 0, we get a P‐value of 0.02. What would the decision be at a significance level of 0.05 a. We would decide to reject the null hypothesis. We would have sufficient evidence that there was a difference between the two population means b. We would decide to reject the null hypothesis. We would have sufficient evidence that there was no difference between the two population means c. We would decide to not to reject the null hypothesis. We would have sufficient evidence that there was no difference between the two population means d. We would decide not to reject the null hypothesis. We would have sufficient evidence that there a difference between the two population means

a. We would decide to reject the null hypothesis. We would have sufficient evidence that there was a difference between the two population means

A study is internally valid if: a. A repeated‐measures design was used. b. A causal relationship is confirmed c. An experimental design was used d. All alternative explanations of the results can be ruled out e. It can be generalized to the population

b. A causal relationship is confirmed d. All alternative explanations of the results can be ruled out

The researcher wants to investigate blood cholesterol levels in patients who follow a diet either low or moderate in fat and who take either a drug to lower cholesterol or a placebo. What is the most appropriate statistical test to use in analyzing the data? a. Independent t‐test b. ANOVA c. Chi‐square test d. Mann‐Whitney U test e. Wilcoxon signed rank test

b. ANOVA

When the researcher simultaneously manipulates two independent variables, the design is a: a. Randomized control trial b. Factorial design c. Cross‐over design d. Repeated measures design e. Cohort study

b. Factorial design

You are evaluating a study that used multiple linear regression assessing the effect of age on prostate-specific antigen (PSA) doubling time (weeks) after controlling for race and comorbit conditions. The slope (i.e., regression coefficient) for age is 1.50. This slope indicates that: a. The average PSA for all men in the sample is 1.50 b. Holding race and comorbid conditions constant, for every 1 unit increase in age, PSA doubling time increase 1.50 units c. Age is statistically significant predictor of PSA doubling time d. The analysis is inappropriate based on the scale of measurement for the DV

b. Holding race and comorbid conditions constant, for every 1 unit increase in age, PSA doubling time increase 1.50 units

A study is conducted to determine the effects of 2 drugs on blood pressure reduction (measured in mm Hg). When plotted out on a frequency distribution, the data are NOT normally distributed. Which of the following measures of central tendency would be most appropriate to use for these data? a. Mean b. Median c. Mode d. A and B e. None of the above

b. Median

A study is conducted to determine the efficacy of 2 drugs in preventing seasickness. Measurements are taken to assess the following: no seasickness, mild seasickness, or severe seasickness. Which of the following data types best describes these measurements? a. Nominal b. Ordinal c. Interval d. Ratio e. Integer

b. Ordinal

A study is conducted to determine the differences in ulcer relapse rates between 2 anti‐Helicobacter pylori drug regimens. The study is conducted on an intention‐to‐ treat basis (ITT). Which of the following is FALSE with regard to ITT studies? a. The researchers are forced to make assumptions about the missing data, which is a potential source of error b. Per‐protocol analyses utilize data from all patients entered into the study, regardless of whether they completed the study or not. c. Last‐observation‐carried forward (LOCF) is a common technique used in ITT analyses d. A and B e. All of the above

b. Per‐protocol analyses utilize data from all patients entered into the study, regardless of whether they completed the study or not.

A study is conducted to compare the efficacy of two medications (Drug A and Drug B) for the treatment of hypertension. One hundred patients are enrolled in the study (50 receiving Drug A and 50 receiving Drug B). Which of the following terms BEST describes patients receiving the study medications? a. Population b. Sample c. Experimental group d. Control group e. None of the abov

b. Sample

The proportion of people with disease, who have positive test results is referred to: a. Specificity b. Sensitivity c. Responsiveness d. None of the above

b. Sensitivity

The standard deviation of a sampling distribution is called a : a. Sampling error b. Standard error c. Variance d. Mean square e. Coefficient of variation

b. Standard error

Non‐probability sampling includes all of the following except: a. Convenience sampling b. Stratified sampling c. Purposive sampling d. Quota sampling e. Snowball sampling

b. Stratified sampling

Based on the 95% confidence interval for the difference, which of the following is TRUE? a. The null hypothesis is accepted b. The alternative hypothesis is accepted c. The data is insufficient to determine whether we can accept or reject the null hypothesis d. Drug A is statistically superior compared to Drug B, but clinically we are 95% confident that there is a possibility that they have similar clinical efficacy e. None of the above

b. The alternative hypothesis is accepted

When p values are statistically significant, the following conclusion(s) can be reached: a. The amount of Type I error is excessive and null hypothesis is accepted b. The amount of Type II error is not evaluated, because the null hypothesis is rejected and differences between groups exist c. There is a clinically significant difference between the groups d. B and C e. None of the above

b. The amount of Type II error is not evaluated, because the null hypothesis is rejected and differences between groups exist

Based on data from the 95% confidence interval for the difference, which of the following is TRUE? a. Drug A is always superior to Drug B in the population of patients with ulcers b. The confidence interval states we are 95% confident that the ulcer cure rates for Drug A is superior to Drug B in the population c. We are 95% confident that Drug A and Drug B may have equal cure rates in the population d. A and B e. All of the above

b. The confidence interval states we are 95% confident that the ulcer cure rates for Drug A is superior to Drug B in the population

Determine which samples are dependent. a. The effectiveness of Prilosec for treating heartburn is tested by measuring gastric acid secretion in a group of patients treated with Prilosec and another group of patients given a placebo b. The effectiveness of Prilosec for treating heartburn is tested by measuring gastric acid secretions in patients before and after drug treatment. The data consist of before/after measurements for each patient c. The flu vaccine is tested by treating one group of subjects with the vaccine while another group of subjects is give placebos d. None of the samples are dependent

b. The effectiveness of Prilosec for treating heartburn is tested by measuring gastric acid secretions in patients before and after drug treatment. The data consist of before/after measurements for each patient

Which of the following is a TRUE statement (based on data from CASE C)? a. The difference in antibiotic efficacy between the groups was 6%, and indicates that clinical significance was achieved b. The statistical power is sufficient to detect a difference of 25% between the groups, but this difference exceeds what a reasonable clinical difference should be c. The study has sufficient power to detect a efficacy difference of 10% between the groups d. All of the above e. None of the above

b. The statistical power is sufficient to detect a difference of 25% between the groups, but this difference exceeds what a reasonable clinical difference should be

From question 46, how should the results be interpreted? 1. Individuals who drink coffee were more likely to develop gallstone 0.6 times when compared with those who do not. 2. Individuals who drink coffee were more likely to develop gallstone 40% lower than those who do not. 3. Individuals who drink coffee were more likely to develop gallstone 60% greater than those who do not. a. 1 b. 3 c. 1, 2 d. 2, 3 e. 1, 2, 3

c. 1, 2

You are evaluating a randomized, double‐blind, parallel‐group, controlled trial that compares four antihypertensive drugs for their effect on BP. The authors conclude that hydrochlorothiazide is better than atenolol (p<0.05) and that enalapril is better than hydrochlorothiazide (p<0.01), but no difference is observed between any other drugs. The authors used independent t‐test to test the hypothesis that each drug was equal to the other. Which of the following statements is TRUE? a. The authors used the appropriate statistical test to analyze their data b. Enalapril is the most effective of these drugs c. ANOVA would have been a more appropriate test d. A paired t‐test is a more appropriate test e. None of the above

c. ANOVA would have been a more appropriate test

A pharmacist wants to implement EBP about the Medication Therapy Management on Hypertension and Diabetes. How should the he/she begin the EBP process? a. By reviewing the literature around the practice b. By consulting with expert nurses c. By fostering an attitude of curiosity d. By evaluating the outcomes of current practices

c. By fostering an attitude of curiosity

The independent variable is marital status (single/married/separated or divorced). The dependent variable is whether or not the person had been hospitalized in the preceding 12 months. The most appropriate statistical test is a(n): a. t‐test b. ANOVA c. Chi‐square test d. Mann‐Whitney U test e. Paired t‐test

c. Chi‐square test

A study is conducted to determine the effect of Vitamin C tablets on mortality. After the study is completed, the data indicate that patients taking Vitamin C tablets had a decrease in mortality. However, the study design did not control for the effects of dietary Vitamin C. Which of the following terms best describes the role of DIETARY Vitamin C in this study? a. Independent variable b. Dependent variable c. Confounding variable d. Control variable e. None of the above

c. Confounding variable

Suppose a researcher is interested in assessing the adequacy of an instrument to measure the theoretical conceptualization of "Quality of Pharmaceutical Care". The type of validation procedure would most probably be: a. Content b. Concurrent c. Construct d. Face e. Predictive

c. Construct

Subjects serve as their own controls in which of the following research design? a. Factorial design b. Parallel design c. Cross‐over design d. Randomized control trial e. Case‐control study

c. Cross‐over design

In a randomized trial of patients who received a cadaver renal transplant, 100 were treated with cyclosporine and 50 were treated with conventional immunosuppression therapy. The difference in treatment was not statistically significant at the 5% level. Therefore, a. This study has proven that cyclosporine is not effective b. Cyclosporin could be significant at the 1% level c. Cyclosporin could be significant at the 10% level d. The groups have been shown to be the same e. The treatment should not be compared because of the differences in the sample sizes.

c. Cyclosporin could be significant at the 10% level

If you decided to reject the null hypothesis but it was in fact true, you a. Made a correct decision b. Made both a type I and type II error c. Made a type I error d. Made a type II error

c. Made a type I error

The next step in the evaluation of the statistical analysis is to select a test that is appropriate for the levels of measurement and a test that tests the primary hypothesis. You must also decide if a test for repeated measures is warranted. Which of the following statistical tests would be most appropriate to test the primary hypothesis stated in Case D? a. Fisher's exact test b. Wilcoxon signed‐rank test c. Mann‐whitney U test d. Student's t‐test for independent groups

c. Mann‐whitney U test

A study was conducted to evaluate an antibiotic in the treatment of possible bacteremia. Five hundred children with fever but no focal infection were randomly assigned to the antibiotic or to a placebo. All patients were reevaluated after 48 hours. The authors reported the proportion of children with major infectious morbidity among those with bacteremia was 13% in the placebo group and 10% in the antibiotic group. The 95% confidence interval for the difference in proportions was ‐2.6% to +8.6%. Thus, the most important conclusion is that a. The difference in major infection morbidity between placebo and antibiotic is statistically significant b. The proportion of children with major infection morbidity is the same with placebo and antibiotic c. No statistically significance exists in the proportions of children that received placebo and antibiotic. d. The study has low power to detect a difference owing to the small sample size, and no conclusion should be drawn until a larger study is done e. Using a chi‐square test to determine significance is preferable to determining a confidence interval for the difference

c. No statistically significance exists in the proportions of children that received placebo and antibiotic.

The most effective method of controlling extraneous variables is by: a. Analysis of covariance b. Matching c. Randomization d. Use control group e. None of the above

c. Randomization

In some non‐randomized studies, patients are assigned to certain treatment groups based on some characteristic that they possess, which could affect the outcome of the study. This type of bias is known as: a. Volunteer bias b. Contamination bias c. Selection bias d. Correlation bias e. None of the above

c. Selection bias

In a placebo‐controlled trial of the use of oral aspirin‐dipyridamole to prevent arterial restenosis after coronary angioplasty, 38% of patients receiving the drug had restenosis, and 39% of patients receiving placebo had restenosis. In reporting this finding, the authors stated that p>0.05, which means that a. Chances are greater than 1 in 20 that a difference would again be found if the study were repeated b. The probability is less than 1 in 20 that a difference this large could occur by chance alone c. The probability is greater than 1 in 20 that a difference this large could occur by chance alone d. Treated patients were 5% less likely to have restenosis e. The chance is 95% that the study is correct

c. The probability is greater than 1 in 20 that a difference this large could occur by chance alone

Investigating gender of the head of household in families of patients whose medical costs are covered by insurance, Medicaid, or self

chi square test

Which of the following is TRUE with respect to the term "random chance?" a. Outcomes of clinical trials may be affected by random chance b. Statistical tests help to determine the effect of random chance on the study outcomes c. Random chance is also known as "systematic variation" d. A and B e. All of the above

d. A and B

Which of the following would INCREASE the amount of Type II error in a study? a. Reduction in sample size b. Reducing the size of delta (assuming sample size remained constant) c. Lack of randomization d. A and B e. All of the above

d. A and B

Which of the answer sets below best matches the definitions in the left column with the stages of testing new therapies listed in the right column? __ Studies involving animals or cell cultures __ Conducted to determine the safety of a treatment in humans. Patients go through intense monitoring __ Large studies (may or may not be a randomized trial) conducted after the therapy has been approved by the FDA to assess the rate of serious side effects and explore futher therapeutic uses __ Relatively large randomized blinded trials used to evaluate the efficacy of an intervention __ Investigator explores test tolerability, safe dosage, side effects, and how the body copes with the drug A. Preclinical B. Phase I C. Phase II D. Phase III E. Phase IV a. A, C, E, D, B b. C, A, B, D, E c. E, A, B, D, C d. A, B, E, D, C

d. A, B, E, D, C

A trial is likely to be underpowered, biased, or misleading if which of the following occurred? a. A large number of participants are lost to follow‐up. b. A substantial number of participants do not adhere to the study intervention. c. Baseline variables are not balanced between intervention groups. d. All of the above

d. All of the above

When p‐values are not statistically significant, the following conclusions can be reached. a. The amount of Type II error should be evaluated b. The p‐value is greater than the significance level c. The test might not have had enough power to produce a statistically significant result d. All of the above

d. All of the above

Patients may alter their behavior when they know they are being observed under study conditions. Which of the following terms best describes this phenomenon? a. Contamination bias b. Selection bias c. Volunteer bias d. Attention bias (Hawthorne effect) e. Admission rate bias (Berkson's paradox)

d. Attention bias (Hawthorne effect)

Which of the following best describes the function of the inclusion criteria for a study? a. Assures patient safety b. Describes the characteristics of the sample c. Is important in evaluating the external validity of the study d. B and C e. None of the above

d. B and C

The disadvantages of the study are a. There may be recall bias concerning smoking history b. Choice of control group may affect results c. Since the disease is rare it will be difficult to get enough cases that will get the disease d. Both A and B

d. Both A and B

A study is conducted retrospectively and collects data on patients who developed a myocardial infarction (MI) and gathers data on another set of patients who haven't had a MI. Differences in exposures to Drug A are compared between the groups, and an odds ratio is calculated. This type of research design is known as: a. Cross‐sectional b. Case report c. Cohort study d. Case/control study e. None of the above

d. Case/control study

Which of the following study designs is best suited for assessing an exposure‐disease relationship when the outcome is rare and/or when the latency period involved is relatively long? a. Cross‐sectional b. Prospective cohort c. Retrospective cohort d. Case‐control

d. Case‐control

Which of the following online evidence sources is most likely to provide preappraised evidence? a. CINAHL b. MEDLINE c. PubMed d. Cochrane Library

d. Cochrane Library

Now that you know the level of measurement, which descriptive statistics would be most appropriate to summarize the pain in each group at each time point? a. 95% confidence interval for the median b. Mean and standard error of the mean c. Mean and standard deviation d. Median and interquartile range

d. Median and interquartile range

Patients are recruited into a clinical trial. The main dependent variable is to determine the average change in cholesterol level between 2 different medications in a randomized, parallel, prospective, double‐blinded trial for the average person living in America. An advertisement is placed on the wall of a local fitness center, and patients from the fitness center respond to the advertisement. Based on this information, which is the most likely bias to affect this study? a. Publication bias b. Therapeutic personality bias c. Selection bias d. Membership bias

d. Membership bias

Results of randomized, double‐blind, controlled clinical trial revealed that the difference in rehospitalization rates between the intervention group and the control group is 6% (p=0.01). The authors conclude that there is a statistically significant difference between the groups. Which of the following statements is TRUE? a. The chance of making type I error is 5 in 100 b. The trial does not have enough power c. There is a high likelihood of having made a type II error d. The chance of making an alpha error is 1 in 100 e. None of the above

d. The chance of making an alpha error is 1 in 100

Which of the following is TRUE with regards to the clinical significance or clinical insignificance of the result (based on data from CASE B)? a. Clinical significance was achieved, because the null hypothesis was accepted, and the difference of 5 mm Hg between groups is clinically significant b. Clinical significance was not achieved, because a difference of 5 mm Hg is not clinically significant c. Clinical significance should not be evaluated, since the null hypothesis was accepted d. The result is clinically significant, because the null hypothesis was rejected, and the difference of 12 mm Hg is clinically significant e. None of the above

d. The result is clinically significant, because the null hypothesis was rejected, and the difference of 12 mm Hg is clinically significant

An early‐phase clinical trial of 50 subjects was designed to evaluate a new drug known to increase HDL‐C levels. The purpose of the trial was to compare the new drug's ability to increase HDL‐C with that of lifestyle modifications (active control group). at the beginning of the study, the mean baseline HDL‐C was 37 mg/dL in the active control group and 38 mg/dL in the new drug group. At the end of the 3‐ month trial, the mean HDL‐C was 44 mg/dL for the control group and 49 mg/dL for the new drug group. The p‐value for the comparison at 3 months was 0.08. Which of the following statements is the BEST interpretation of these results? a. An a priori alpha of less than 0.10 would have made the study more clinically useful b. The new drug and active control appear to be equally efficacious in increasing HDL‐C c. The new drug is better than lifestyle modifications because it raises HDL‐C to a greater extent d. This study is potentially underpowered e. A paired t‐test is appropriate test to compare between two group at the end of the 3‐month trial

d. This study is potentially underpowered

One of the main limitations of a cross‐over design in drug studies in humans is: a. The study is conducted during a specified time "window" b. Requires a washout period to remove any residual effects from previous drug treatments c. If a drug induces a permanent effect (i.e. antibiotic cure), cross‐over designs cannot be used d. B and C e. All of the above

e. All of the above

Which of the following is a TRUE statement? a. Statistical significance does not assure that the results are clinically significant b. If the p values are small (i.e., less than alpha), then clinical significance should always be interpreted c. If the p values are large (i.e., greater than alpha), then clinical significance should NOT be evaluated d. A and C e. All of the above

e. All of the above

Leitzmann et al. examined the link between drinking coffee and gallstone disease. Individuals who drank between 2 and 4 cups of coffee per day had a relative risk of developing symptomatic gallstones of 0.6. What type of study was this? a. Case report b. Cross‐sectional study c. Case control study d. Clinical trial e. Cohort study

e. Cohort study

In clinical trials, which of the following statements is TRUE? a. The acceptable amount of Type I error is usually > 5% b. The acceptable amount of Type II error is usually > 20% c. Statistical power will increase as the size of delta decreases (assuming the sample size remains constant) d. A and B e. None of the above

e. None of the above

What is the correct interpretation of the 95% confidence interval for the difference in ulcer cure rates? a. We are 95% confident that the difference in ulcer cure rates for the sample lies within the interval +2, +10 b. We are 95% confident that the difference in ulcer cure rates for the sample lies within the interval between 84% and 90% c. We are 95% confident that the difference in ulcer cure rates for the population lies within the interval between 84% and 90% d. B and C e. None of the above

e. None of the above

Can a study with poor internal validity have a high level of external validity?

no

Investigating average body weight before and after a supervised exercise program

paired t test

In what type of study is blinding most useful?

subjective data