Study Design, Biostatistics And Evaluating

3. A newly available NSAID was designed to reduce the incidence of adverse GI events compared with traditional NSAIDs. A large retrospective cohort study compares the incidence of ulceration and bleeding associated with the use of this new NSAID with that of ibuprofen and naproxen. The results indicate that the new agent is associated with no statistically or clinically significant reduction in ulceration or bleeding with long-term use compared with ibuprofen and naproxen. The investigators of the study argue that the lack of difference in safety is because the drug is being promoted as safer; therefore, most patients receiving it are at a much higher baseline risk of NSAID-induced ulceration and bleeding. If this phenomenon did indeed affect the study results, which potential source of bias would most likely be present?

• A. Recall bias.
• B. Misclassification bias.
• C. Interviewer bias.
• D. Channeling bias.

7-Researchers conduct a study to determine the impact of potential risk factors on the occurrence of breast cancer. Of the techniques available to researchers useful for controlling for confounding variables, which method could ultimately reduce the external validity of their study?

• A.Matching
• B.Multivariate analysis
• C.Restriction
• D.Stratification

Answer: C. Restriction

There are three or four key ways for researchers to control for, or reduce the impact of, confounding variables.Of those,restricting the study population to individuals who do not have the confounding variable will reduce the external validity (i.e., generalizability) of the study findings.Block randomization, matching, stratification, and multivariate statistical analysis are other ways to control for confounding and do not limit the inclusion or enrollment of study subjects,thereby not reducing external validity (generalizability).certified

11-Researchers desire to conduct a randomized, active-comparator clinical trial evaluating two currently available anti-hypertensive medications. Which of the following study-design terms most appropriately describes the situation in which study subjects and investigators know the identity of the treatment being utilized for each individual subject?

• A. Open-labeled
• B. Single-blinded
• C. Double-blinded
• D. Triple-blinded

Correct Answer:"A".

Open-label studies are those in which all study subjects, investigators, and evaluators know which treatment allocation group that each study subject has been allocated (albeit randomly)

Your oncology colleagues would like to conduct a research project to answer a question, and they approach you to join them in conducting the project and would appreciate your input regarding the type of study design that would be most appropriate to answer the research question. The question is the following: Is there a difference in exposure to cell phone- and cell tower-radiation in patients with neuroblastoma? Your colleagues anticipate obtaining patients with neuroblastoma from their institutions’ neuro-oncology clinic. Your colleagues wish to determine the differences in exposure between those with neuroblastoma and those without this disorder. They also wish to complete the research in time to meet the deadline for submission of the abstract to their favorite oncology organization’s annual meeting (deadline in 6 months). Which study design would be most appropriate for you to recommend?

• A. Case-control
• B. Randomized clinical
• C. Cross-sectional
• D. Cohort

A. Case-control

The colleagues clearly indicate that they wish to start with the individuals with neuroblastoma, patients with the disease. They would also like to assess the difference in the exposure status of these patients and of the patients without the disease. In addition, because of the limited time for completion, the only appropriate choice would be a case-control study. Because patients with neuroblastoma are readily available, and you wanted to assess exposure, this is the classical design and function of a case-control study. Because case-control studies are retrospective, they are also able to be completed in a shorter time than either a prospective cohort or a randomized clinical trial.

A study reports an increased incidence of falls associated with the use of a certain type of antidepressant. The study selected groups of patients based upon whether or not they had received medical care for a fall in the previous 3 years. Patients in both groups (i.e., those with a fall or those without a fall) were matched based on their age, preexisting psychiatric and medical conditions, other medications, and sex. Which one of the following represents the type of trial design used in the above investigation?

• (A) prospective cohort
• (B) retrospective cohort
• (C) case-control
• (D) randomized, controlled trial

Answer: C.

In this trial, participants are divided into two groups based on whether or not they experienced a particular clinical event (i.e., a fall). Both groups, those experiencing a fall during the study period and those not experiencing a fall during the study period, are assessed for their use of a particular type of antidepressant. This description is consistent with a case-control design (answer c). In a cohort trial, patients are grouped based on their exposure (e.g., to a certain type of drug) and then followed to see if they experience a particular clinical event (e.g., a fall). In a prospective cohort trial (answer a) the exposure happens in the past, but the event may occur in the future as patients are followed prospectively through time. In a retrospective cohort design (answer b), both the exposure and outcome have already occurred. A randomized, controlled trial (answer d) assigns people to groups randomly and no part of the study is based upon retrospective observations.

21-You are a member of the hospital Pharmacy and Therapeutics (P&T) Committee. The hospital is considering formulary addition of a new drug to treat intensive care unit (ICU) delirium. A prospective head-to-head study comparing this novel drug with haloperidol showed similar efficacy between the two drugs, with a favorable trend in favor of the new drug because of minimal/no effect on the QT interval. A comparative retrospective analysis showed that patients treated with the new ICU delirium drug had a shorter length of hospital stay and decreased hospital costs. Using an evidence-based literature evaluation, what is your recommendation?

• A. The new drug should be added to the formulary because of potential cost savings in the retrospective analysis.
• B. The new drug should be added to the formulary because of its improved safety profile with minimal/no QT-interval prolongation.
• C. The new drug should not be added to the formulary because retrospective analyses are subject to information bias.
• D. The new drug should not be added to the formulary because the results of retrospective analyses are subject to misclassification bias.

C. The new drug should not be added to the formulary because retrospective analyses are subject to information bias.

The Pharmacy and Therapeutics Committee is responsible for evaluating the potential benefits of therapies up for formulary consideration compared with current formulary therapies. In addition, a comparison of the possible risks associated with the new therapy versus current formulary therapies is critical. Retrospective chart reviews are among the lowest quality of data available and are subject to recall bias and possible inaccurate or invalid data/information bias. Typically, agents are considered for formulary status on the basis of several clinical trials, particularly randomized, double-blind, controlled clinical trials, not just one retrospective chart review, even if cost-savings are involved. Potential cost-savings must be considered in the context of the frequency of anticipated use of the newer agent compared with current formulary agents. Just because a retrospective review showed a slight safety advantage does not require it to be added to the formulary because many other factors, including efficacy, must also be weighed in the decision.

24-There have been many small reports showing that a drug used to treat cancer also caused hypertension.

Which of the following studies would provide rapid information regarding this fact?

• A. Case control study
• B. Metanalysis
• C. Double blind trial
• D. A review article

Answer: b) metanalysis.

Meta-analysis can be performed when there are several small studies; the study would take into account all the data from the studies and provide good estimate of whether an effect was present. This would also be quicker and more cost-effective than performing a double blind randomised trial. Overall, this question has been attempted 225 times (56.89% correct) in an average time of 28 s.

33-As the incidence for tuberculosis increases, the most likely effect on prevalence is that it

• a.Increases.
• b.Decreases.
• c.Stay the same.
• d.Cannot be determined from the given information.

42-As a member of a clinical trials group at a large pharmaceutical company, you are asked to design a study to ascertain whether a link exists between new-onset bronchospasm and a newly marketed drug for the treatment of ALS (acute lateral sclerosis). The U.S. Food and Drug Administration (FDA) has expressed concern about data that suggest a trend toward increased risk of bronchospasm, observed in early clinical trials; therefore, the FDA has asked for some phase IV data after marketing. Which one of the following should be suggested as the most appropriate study design?

• A. A case series.
• B. A randomized, controlled trial.
• C. A prospective cohort study.
• D. A retrospective cohort study.

Answer D:

A retrospective cohort study. Bronchospasm is a common disorder; however, ALS is uncommon, and thus, exposure to this agent would be uncommon, too. This study lends itself to a cohort trial. A prospective cohort would potentially take longer to conduct; thus, a retrospective cohort would be chosen.

44-You are one of the pharmacists on the Pharmacy and Therapeutics Committee. A formulary drug is being scrutinized because of rare reports of serious drug-induced rhabdomyolysis. The FDA is reevaluating the safety of the drug, but that reevaluation is likely to take a considerable amount of time. The committee is considering whether to take a drug off the formulary.

Which one of the following study designs would you consider best suited for this problem?

• A. Randomized, controlled crossover clinical trial.
• B. Case-control study.
• C. Prospective cohort study.
• D. Randomized, controlled parallel-group clinical trial.

44-Answer B: Case-control study.

A case-control is the most appropriate study design in this situation since this study would be investigating a relatively common exposure (drug is on the market and being routinely used clinically) and a relatively rare/uncommon outcome. Neither A nor D would be ethically possible, since we know there is a published rare, but serious toxicity. Additionally, A, D, and C are incorrect in that these designs would also take a considerable amount of time to conduct.

A study that measures the number of persons with influenza in a calendar year

• A. Cohort study
• B. Case control
• C. Cross-sectional
• D. Case report

A study was designed to assess the association between glucocorticoid use and the development of acute pancreatitis. A total of 6,161 individuals with acute pancreatitis and 61,637 individuals without pancreatitis were enrolled and then the investigators looked retrospectively for differences in glucocorticoid use between the study groups.

Which option best describes this study design?

• A Case Series
• B Case-control
• C Cohort
• D Cross-sectional

Answer B.

This is an example of a case-control study. Case-control studies are most appropriate when investigators want to study differences in exposure (and the presence of the outcome is known). In this item, investigators know if the patient has paceratitis or not, and wish to evaluate the differences in glucocorticoid use (exposure). This is different from cohort studies which study differences in the outcome (and the presence of the exposure is known). A case series is when more than one patient with a similar experience is reported in a descriptive review. A cohort study would classify a patient population by whether patients received a corticosteroid and then determine whether they developed acute pancreatitis. A cross-sectional study is an evaluation of the prevalence or characteristics of a group of individuals at a particular time.

The Sudden Cardiac Death in Heart Failure trial evaluated the efficacy of amiodarone or an ICD versus placebo in preventing all-cause mortality in ischemic and nonischemic NYHA class II and III patients with HF. There was a 7.2% absolute risk reduction and a 23% relative risk reduction in all-cause mortality at 60 months with an ICD versus placebo. Which best shows the number of patients needed to treat with an ICD to prevent one death versus a placebo?

• A. 1.
• B. 4.
• C. 14.
• D. 43.

Answer: C

The number needed to treat can be calculated by 1/absolute risk reduction. Because the absolute risk reduction in mortality at 60 months was 7.2% with ICD versus placebo, 1/0.072 would be used to calculate the number of patients needed to treat to prevent one death during this time. About 14 patients would need to be treated with ICD to prevent one death in 60 months versus placebo. Other calculations in this fashion, including relative risk reduction and 100% minus the absolute or relative risk reduction, do not provide useful information for interpreting the trial results and yield an incorrect number of patients.

10-A randomized, double-blind, placebo-controlled clinical trial of patients with disseminated intravascular coagulation (DIC) showed a mortality reduction in patients treated with antithrombin III (44%) versus placebo/supportive care (56%).

What is the number of DIC patients needed to treat with antithrombin III in order to prevent 1 death?

• A. 5
• B. 8
• C. 22
• D. 40

Answer B. 8

To calculate the number needed to treat to prevent an event, the percentage in the treatment group must first be subtracted from the percentage in the comparator group. 

56% − 44% = 12% / 100 = 0.12, 1/0.12=8.33. Eight patients would require treatment to prevent one death.

16-A study reports an odds ratio of 1.18 for ventricular arrhythmia associated with the use of a recently marked diuretic when compared to hydrochlorothiazide. The 95% confidence interval reported for this odds ratio is 1.04 to 1.33.

Which one of the following is the correct interpretation of these results?

• (A) The risk of ventricular arrhythmia may be decreased between 4% and 33% with the use of this new diuretic when compared to hydrochlorothiazide
• (B) The risk of ventricular arrhythmia may be 4 to 33 times less with the use of this new diuretic when compared to hydrochlorothiazide
• (C) The risk of ventricular arrhythmia may be elevated between 4% and 33% with the use of this new diuretic when compared to hydrochlorothiazide
• (D) The risk of ventricular arrhythmia may be 4 to 33 times greater with the use of this new diuretic when compared to hydrochlorothiazide

Answer: C.

Odds ratios are compared to baseline risks of one for their interpretation. Numbers greater than one indicate an increased risk, while numbers less than one indicate a decreased risk. In this question, an odds ratio (OR) of 1.18 indicates a potentially increased risk. As such, answers a and b are incorrect. Answer d (4 to 33 times greater risk) is consistent with a confidence interval for the OR of 4.0 to 33.0. A confidence interval of 1.04 to 1.33 means that the risk of ventricular arrhythmia may be increased anywhere from 4% to 33% when compared to the baseline risk of using hydrochlorothiazide (answer c).

In a cohort study designed to determine an association between measles, mumps, and rubella vaccination and autism, investigators report the relative risk of autistic disorder in the vaccinated group compared to the unvaccinated group as 0.92 (95% confidence interval [CI] = 0.65–1.07).

Which one of the following p-values is consistent with these reported findings?

• A. A p-value of less than 0.05.
• B. A p-value of less than 0.01.
• C. A p-value of greater than 0.05.
• D. A p-value of greater than 0.10.

The mean difference in diastolic blood pressure between patients taking either a new or traditional antihypertensive is 10 mm Hg (95% confidence interval = -15 mm Hg to 51 mm Hg). The difference in diastolic blood pressure was not statistically significant at the 0.05 level.

What number is included in the range of values that allows you to make this decision?

• A. 0.
• B. 1.
• C. 2.
• D. 20.

A study that compares the cholesterol-lowering effect of a new medication to traditional therapy found no difference in total cholesterol between the two groups after 1 year of treatment. What type of error would you be making if you wrongly concluded that there is no difference between the two different treatment arms in total cholesterol?

• A. Type I error.
• B. Type II error.
• C. Power.
• D. α Error.

When calculating the sample size to determine the difference between two means, which one of the following are elements to consider in the equation?

• A. Effect size.
• B.Sensitivity
• C.The statistical test that will be used
• D.Measurement process

A prospective study was done over 3 years to determine if a new antihypertensive (Drug A) would offer any benefit over a placebo for the treatment of hypertension. 500 patients were randomized to one of 2 groups: Group A (250 patients total), Drug A 50 mg by mouth once daily or Group B (250 patients total), placebo by mouth once daily. The primary endpoint was the development of a stroke. At the end of the trial, 88 patients in Group A had a stroke, and 95 patients in Group B had a stroke.

What would be an appropriate statistical test for analyzing the primary endpoint of this study?

• Fisher’s Exact
• Chi-Square
• Student’s-Hest
• 2-way ANOVA

Answer is B

A group of investigators has designed a study to determine if simvastatin (Zocor) is more effective than pravastatin (Pravachol) for the treatment of high cholesterol. They designed the study so that 1000 patients would be randomly assigned to one of 2 groups. Zocor 40 mg once daily or Pravachol 40 mg once daily. At baseline, patients had a fasting lipid profile done and were assessed for myopathy (muscle aches/pain). These patients were then prospectively followed for 3 months. At 3 months, the patients had a follow-up visit where fasting lipid profiles and patient assessments on myopathy (none, mild, moderate, or severe pain) were completed. The primary endpoint was to determine the change in LDL-c (mg/dL) from baseline. The secondary endpoints included: 1-The proportion of patients achieving their LDL-c goals per guidelines; 2-The patients rating of myopathy. Which of the following statistical tests should you use when analyzing secondary endpoint number 1 (the proportion of patients achieving their LDL-c goal)?

• McVennar Test
• Chi Square
• Wilcoxon Rank Sum
• Sign Test

The average weight of a group of 100 patients is 60 kg, standard deviation of 5 kg. Weights are normally distributed. The standard error of the weights is which one of the following?

• A-12 kg
• B-0.6 kg
• C-5 kg
• D-Larger than the standard deviation
• E-Smaller than the standard deviation

A research group is analyzing the relationship between various independent patient demographics (e.g., age, height, weight, Alb, creatinine clearance [CrCl]) and phenytoin pharmacokinetics. Which is the best statistical test to use in assessing the relationship?

• A. One-way analysis of variance.
• B. Analysis of covariance.
• C. Multiple regression.
• The shortest way to be board-certified
• D. Spearman's rank correlation.

Answer: C

The correct statistical test is multiple regression. Multiple regression is used to describe the relationship between a dependent variable and two or more independent variables when both the dependent and independent variables are numeric. Analysis of variance is used to describe the relationship between a dependent variable and two or more independent variables when the dependent variable is numeric and the independent variables are nominal. Likewise, analysis of covariance is used to describe the relationship between a dependent variable and two or more independent variables when the dependent variable is numeric and the independent variables are nominal with confounding factors. Spearman's rank correlation is a nonparametric test used to describe the relationship between one dependent and one independent variable when the data are ordinal or numeric and not normally distributed.

2-A placebo-controlled clinical trial is conducted comparing an investigational antiplatelet agent for preventing stroke. The hemoglobin levels are monitored for adverse effects of the investigational agents. The mean hemoglobin value for the investigational agent-treated group is 13.7 g/dL with a standard deviation of 1.9 g/dL. Based on these values, and assuming a normal distribution of the groups' hemoglobin values, which of the minimum (below the mean) and maximum values (above the mean) would most completely incorporate over 99% of all the values of the placebo-treated study sample?

• A. 12.8-14.7 g/dL
• B. 11.8-15.6 g/dL
• C. 9.9-17.5 g/dL
• D. 8.0-19.4 g/dL

D. 8.0–19.4 g/dL

4-In reviewing a research proposal for a randomized, double-blind, placebo-controlled trial of a new antihypertensive versus placebo, you note that the primary outcome is listed as differences in mean blood pressure between the two groups after 8 weeks of treatment. The primary outcome can be classified as what type of data?

• A. Nominal
• B. Ordinal
• C. Interval
• D. Ratio

D. Ratio: Changes in blood pressure are measured as continuous variables. Because the zero point for blood pressure is absolute, it is considered ratio data.

6-Researchers are evaluating the potential beneficial effects of an investigational therapy for the treatment of Crohn's disease. Investigators are asking study participants to record their bowel movements into a personal diary, listing the number and consistency of bowel movements and any associated pain throughout the day. Researchers will categorize each patient’s bowel movements into one of the following categories: 1-5 bowel movements per day, 6-10 bowel movements per day, or 11 bowel movements per day. Which data type most appropriately represents the categories of bowel movements that the researchers will utilize?

• A. Composite
• B. Interval
• C. Nominal
• D. Ordinal

D. Ordinal

The data type for the number of bowel movements is interval because this variable has magnitude and a fixed interval within the counts. However, the stem indicates that the number of bowel movements would be forced into one of the fixed categories. Because of this data change, the data become categorical and therefore ordinal. Ordinal scales have magnitude, but not even, with a fixed interval within the categories.

7-You are conducting an investigation to compare the hospitalization rates of asthma exacerbations for patients managed by a clinical pharmacist-led pharmacotherapy clinic with their usual care (e.g., primary care physician [PCP]-office management). You intend to compare differences in the percentage of hospitalization rates for each group at the end of 1 year as your primary outcome. What would be the most appropriate statistical test to assess any apparent differences in this outcome?

• A. Chi-square
• B. Student t-test
• C. Analysis of variance (ANOVA)
• D. Wilcoxon signed rank

A. Chi-square: Comparing proportions/percentages between independent groups with a dichotomous outcome.

8-A study of patients with moderately severe persistent asthma reports the mean number of allotterol puffs that were used over a 2-week period. In a sample of 100 patients, the mean number of puffs used over the observation period was 26. The authors report a standard error of the mean (SEM) of 1. You would like to know about the variability of allotterol usage in the sample. Which one of the following represents the value of the standard deviation (SD) that gives you such information?

(A) 1

(B) 10

(C) 100

8-Answer: B

The standard deviation and the standard error of the mean are related mathematically. Usually, the SEM is calculated from the SD and sample size, but, like any algebraic relationship, you can solve the equation for any of the variables when given the other two. The SEM can be calculated by dividing the SD by the square root of the sample size. As such, the SD can be calculated by multiplying the SEM by the square root of the sample size. In this case, the SEM (i.e., 1) times the square root of 100 (i.e., 10) equals 1 puff (answer a).

10-In addition to these results, the authors also describe a “rule” to predict which patients in the study were finally admitted to the hospital based on their FEV1 at baseline, their change in FEV1 following the first dose of drug, their oxygen saturation at baseline, and their history of use of inhaled corticosteroids. Which one of the following is the appropriate statistical technique to derive such a proposed rule?

• (A) a correlation analysis
• (B) a survival analysis
• (C) a regression analysis
• (D) an analysis of variance

10-Answer: C.

A regression analysis (answer c) is the type of analysis that is used to construct predictive models. Regression allows for different variables to be added or subtracted from a model to discern the best predictive equation. Correlation analysis (answer a) only explains how a variable changes in association with another variable and does not allow the prediction of one variable based on the value of the other. Survival analysis (answer b) helps us to understand the probability of experiencing an outcome through a period of time. Analysis of variance (answer d) is a type of inferential statistical test and is not used to predict values of variables.

11-A randomized trial investigates the difference in the incidence of symptomatic deep vein thrombosis (DVT) when using two different injectable anticoagulants in a medically ill inpatient population. After studying 32 patients, the p-value comparing the percentage of DVT in each group is calculated as 0.12. Which one of the following is an important consideration in order to apply the results of this trial correctly?

• (A) Type I error
• (B) type II error
• (C) confounding by indication
• (D) interviewerbias

11-Answer: B.

Because this is a randomized trial, the risks of influencing the outcome due to confounding by indication (answer c) or interviewer bias (answer d) are minimized. Even if these were potential concerns (which they generally are not for prospective trials), their risks should be distributed equally in each group due to the randomization. Type I error (answer a), by definition, is only a consideration if a statistically significant difference is found. Type II error (answer b) is important to consider whenever a non-significant result is reported in a clinical trial. The small sample size in this trial highlights this as a consideration even further.

A reader cannot assume that a "small sample size" equals "underpowered," and if indeed the study did meet power, then a non-significant finding for an appropriately powered outcome would not indicate a type II error

12-A study measures the effects of 3 levels of drug dosing on feelings of well-being (as measured by a 4-point scale). Higher ratings indicate higher doses and better well-being, respectively. Correlation analysis results in r=0.8. Which one of the following is the best interpretation of these results?

• (A) Increasing doses of the drug are highly correlated with increases in feelings of well-being
• (B) Increasing doses of the drug cause increases in feelings of well-being
• (C) Increasing doses of the drug are highly correlated with decreases in feelings of well-being
• (D) Increasing doses of the drug cause decreases in feelings of well-being

An r2 value indicates both the magnitude and direction of any correlation that exists between the variables under question. Important to always remember is that rs2 is only a measure of correlation (and linear correlation at that) and is not a measure that assesses causality. Considering this, answers b and d are not correct interpretations. Answer C is incorrect because it is a negative r2 that would indicate an inverse correlation between variables. Here, the r2 is positive, meaning that as one variable increases, so does the other. Answer a is the only correct interpretation.

22-The following table illustrates the findings of a hypothetical retrospective cohort study of 100 subjects living in small villages in Ethiopia. Specifically, the study evaluated the frequency of disease after exposure to contaminated pond water. Based on this table, which statement is most accurate in describing the risk ratio (RR) comparing disease between those subjects exposed and not exposed?

• A. Compared with the subjects who were not exposed, subjects exposed to the contaminated pond water were 2.55 times more likely to develop disease.
• B. Compared with the subjects who were not exposed, those exposed to the contaminated pond water were 54% more likely to develop disease.
• C. Compared with those exposed to the contaminated pond water, subjects who were not exposed were 62.5% less likely to develop disease.
• D. Compared with those exposed to the contaminated pond water, subjects who were not exposed were 2.55 times more likely to develop disease.

22-Answer Explanation

A. Compared with the subjects who were not exposed, subjects exposed to the contaminated pond water were 2.55 times more likely to develop disease. Because the risk ratio (RR) is a ratio of proportions, calculation of the RR is accomplished by first determining the risk of disease in the exposed group and the unexposed group. Individually. This can be obtained by dividing the number of subjects who developed disease (in each of the two groups: exposed and unexposed) by the total study population (column totals; for those who were exposed and those who were unexposed, respectively). Therefore, the risk of disease in the exposed group would be 42/44 = 0.955, and the risk of disease in the unexposed group would be 21/56 = 0.375. Then, ultimately, the RR would be 0.955/0.375 = 2.55. The secondary aspect of this correct answer is the ability to know which group is the reference group and which group is being compared with the reference group. In all situations, the reference group is the unexposed group, and the comparison group is the exposed group. Therefore, the RR is the ratio of the exposed group (comparison group) to that of the unexposed group (reference group). Because of this, the summary statement of the RR is the RR of the exposed group compared with the unexposed group. This ordering of the statement is what makes Answer D incorrect because it is stated backward (wrong comparison group and wrong reference group; comparing the wrong group order).

43-A study is designed to investigate the number of adverse drug events in a group of individuals during a 1-year period. In the results, the authors report that five events occurred per 10,000 person-years. Which one of the following measures does this estimate represent?

• A. Relative risk.
• B. Incidence rate.
• C. Odds ratio.
• D. Prevalence.

43-Answer B: Incidence rate.

Incidence rate is a measure of the instantaneous rate of developing a disease. It is estimated by dividing the number of individuals who develop a disease by the total amount of time that these individuals were at risk for developing a disease The prevalence is a measure of the number of individuals who have a condition/disease at any given time. Both relative risk and odds ratio estimate the magnitude of association between exposure and disease,

In a cohort study of 1000 non-smokers and 1000 smokers, 450 of the smokers and 50 of the non-smokers developed COPD. What is the relative risk of smoking for COPD?

• A. 9
• B. 1/8
• C. 1/9
• D. 8

RR= (450 /1000)/(50/1000)= 0.45/0.05=9

The HOPE comparing the incidence in the ramipril group to placebo group) revealed that the relative risk of myocardial infarction, stroke, and death from cardiovascular causes was 0.78 (95% CI; 0.70 - 0.86).

Which of the following conclusions provides best interpretation of this result :

• A-22% of the risk for having a myocardial infarction, stroke, or death from cardiovascular causes was increased by the use of ramipril
• B-22% of the risk for having a myocardial infarction, stroke, or death from cardiovascular causes was removed by the use of ramipril
• C-78% of the risk for having a myocardial infarction, stroke, or death from cardiovascular causes was increased by the use of ramipril
• D-78% of the risk for having a myocardial infarction, stroke, or death from cardiovascular causes was removed by the use of ramipril.

B-22% of the risk for having a myocardial infarction, stroke, or death from cardiovascular causes was removed by the use of ramipril

When describing the results of a randomized, controlled clinical trial, the investigators report using a per-protocol analysis to analyze their data. The results of their investigation comparing two pharmacologic treatments for multiple sclerosis show no difference between the treatment groups regarding the number of relapses in either group. Given their method of data analysis, which one of the following is this result best described by?

• A. May be susceptible to issues regarding a lack of power.
• B. Provides a good measure of effectiveness under usual clinical conditions.
• C. Cannot provide an estimate of the method’s effectiveness.
• D. Was obtained using the most common approach to data analysis for clinical trials.

The association between low birth weight and maternal smoking during pregnancy can be studied by obtaining smoking histories from women at the time of the first prenatal visit and then subsequently assessing and assigning birth weight at delivery according to smoking histories. What type of study is this?

• a. Clinical trial
• b. Cross-sectional
• c. Prospective cohort
• d. Case-control
• e. Retrospective cohort

A research team wishes to investigate a possible association between smokeless tobacco and oral lesions among professional baseball players. At spring training camp, they ask each baseball player about current and past use of smokeless tobacco, cigarettes, and alcohol, and a dentist notes the type and extent of the lesions in the mouth. What type of study is this?

• a. Case-control
• b. Cross-sectional
• c. Prospective cohort
• d. Clinical trial
• e. Retrospective cohort

In a recently published clinical trial comparing rivaroxaban versus placebo in patients with recent acute coronary syndrome (ACS), the authors reported that the risk of myocardial infarction (MI) in the rivaroxaban group was 3.76% (384 MI’s out of 10,225 patients) in the rivaroxaban-treated patients and 4.47% (229 MIs of 5125 patients) in the placebo-treated patients. What is the number of patients with recent ACS that would need to be treated (NNT) to reduce the risk of one MI?

• A. 70
• B. 85
• C. 140
• D. 195

Answer is C

ARR=0.0447 - 0.0376 = 0.0071                       NNT = 10000/71 =140

In a study, the odds of disease in the group who smoke is 0.25. This means which one of the following:

• a) Smoking causes disease
• b) For every 4 who smoke, 1 has the disease
• c) For every 5 who smoke, 1 has the disease
• d) The disease occurs 4 times more often in those who smoke
• e) A larger sample is needed

What type of study design is best fit to answer the following research question: Does smoking associated with low birth weight baby in pregnant women?

• a. Randomized clinical trial.
• b. Case-control study.
• c. Prospective cohort study.
• d. Cross-sectional study

A new ARB was being studied and compared to telmisartan. The manufacturer wants to show that the new drug is similar to or no worse than telmisartan. The manufacturer does not expect the drug to be better than telmisartan. Which study type should be used to study this problem?

• A. Superiority trial
• B. Equivalence trial
• C. Non-inferiority trial
• D. None of the above

A new enzyme immunoassay for HCV RNA has a reported sensitivity of 95% and a specificity of 92%. If the prevalence of HCV in a cohort of 500 patients is 40%, which best represents the positive predictive value of this new test?

• A. 75%.
• B. 89%.
• C. 92%.
• D. 96%.

Answer is B

PPV= 190/(190+24)=88.7 %

While resuscitating a patient with anaphylaxis from a bee sting, a physician issues a verbal order for "epinephrine 1 amp." This is incorrectly interpreted by the nurse, and epinephrine 1 mg is administered intravenously. On discovery of this error, an electrocardiogram and cardiac toponins are ordered. Because several similar cases have occurred within the past year at this hospital, a retrospective observational study of epinephrine medication errors is proposed.

Which is the best design for this proposed study?

• A Case report.
• B Randomized controlled.
• C Cohort.
• D Cross-sectional.

Correct Answer: C

Answer: C. Cohort studies are best suited for determining the association between exposures/factors and disease/condition development. A case report would only be useful when describing a single event, which would not apply to this case. A randomized controlled study, although ideal in determining cause-and-effect, would not be practical in this single-center setting and might be unethical if the control group were to receive anything other than the standard of care (intramuscular epinephrine). A cross-sectional analysis of data is collected from a population, or a representative subset, at one specific point in time. The question at hand would require data collected over an extended period.

You have designed a trial to assess the efficacy of the health care-associated pneumonia (HCAP) classification first mentioned in the 2005 Infectious Diseases Society of America hospital-acquired pneumonia/ventilator-associated pneumonia (HAP/VAP) guidelines and subsequently removed from the 2016 HAP/VAP guidelines. You are comparing patients who meet the HCAP criteria and are treated with HCAP recommendations with patients who meet the HCAP criteria and are treated with community-acquired pneumonia guidelines. The primary outcome is to determine the number of patients with a composite outcome of treatment failure (including escalation of level of care, antibiotic adjustment, and mortality) in each group.

Which is the most appropriate statistical test to assess your primary outcome?

• A Chi-square test.
• B Student t-test.
• C Analysis of variance.
• D Wilcoxon rank.

Correct Answer: A.

Explanation: Answer: A The primary outcome is a categorical variable. Answer B is incorrect because the data are not parametric or continuous. Answer C is incorrect because there are only two groups, and the data are nonparametric. Answer D is incorrect because the groups are not related

A double-blind, placebo-controlled clinical trial is designed to evaluate the effect of early steroid administration on mortality in patients with acute respiratory distress syndrome (ARDS). After a 28-day follow-up, the ARDS patient mortality rate is 20.6% in the steroid group compared with 42.6% in the placebo group.

Which best depicts the relative risk reduction and number needed to treat (NNT)?

• A 52%; 2.
• B 52%; 5.
• C 22%; 2.
• D 22%; 5.

Correct Answer: B

Answer: B. Relative risk reduction equals (42.6 ~ 20.69/42.6, which equals 52%. The NNT equals 100/ absolute risk reduction, which equals 1002

A pooled analysis of results from seven randomized controlled trials compares pregabalin's different dose sizes (strengths) in the treatment of painful diabetic neuropathy. Dose regimens compared included pregabalin 150 mg, 300 mg, and 600 mg doses administered two or three times daily for 5–13 weeks. The primary outcome was patients reporting at least or greater than 50% reduction in pain from baseline. Results were: placebo: 121 of 560; 150 mg strength: 47 of 175 patients; 300 mg strength: 103 of 265 patients; 600 mg strength: 238 of 507 patients.

Which best represents the odds of achieving a greater than 50% reduction in pain from baseline between the 300-mg group and the 600-mg group?

• A 8%
• B 12%
• C 72%
• D 83%

Odds ratio = (103/162)/(238/269)=0.72

Correct Answer: C

A pooled analysis of results from seven randomized controlled trials compares pregabalin at different dose sizes (strengths) in the treatment of painful diabetic neuropathy. Dose regimens compared included pregabalin 150 mg, 300 mg, and 600 mg doses administered two or three times daily for 5–13 weeks. The primary outcome was patients reporting at least or greater than 50% reduction in pain from baseline. Results were: placebo: 121 of 550; 150 mg strength: 47 of 175 patients; 300 mg strength: 103 of 265 patients; 600 mg strength: 238 of 507 patients..

Which best represents the absolute risk reduction (ARR) between the 300-mg group and the 600-mg group?

• A 5%
• B 7%
• C 8%
• D 12%

Correct Answer: C

Answer C. The ARR i the difference between the risk of not achieving a greater than 50% reduction in pain between the 300 mg and the {600-mg treatment groups. Mathematically, the ARR is (238/507) ~ (103/265) = 0.081 or 8%.

A Cochrane review of corticosteroid use in adults with bacterial meningitis caused by Streptococcus pneumoniae reported a relative risk of 0.84 (95% confidence interval, 0.72–0.98) for reduction in mortality. The overall outcomes in this patient subset are as follows: 168 of patients in the corticosteroid group died, and 393 survived; 203 patients in the placebo group died, and 368 survived.

Which best depicts the number needed to treat (NNT) to prevent one death?

• A 2.
• B 6.
• C 8.
• D 17.

CorrectAnswerD

Answer: D The NNT can be calculated by 1/(absolute risk reduction [ARR]). The ARR is calculated by the control event rate minus the experimental event rate. Therefore, in this example, the ARR is 0.36 0.20 = 0.08 and NNT is 1/0.06 = 167, which rounds up to 17 pallens. Answer A is incorrect because this is a calculation of the relative risk reduction. Answer B is incorrect because this is a calculation of just the ARR. Answer C is incorrect because this was calculated using 168/393 instead of 1681561 for the event incidence in the experimental group and 208/368 instead of 203/571 in the control group

The study “Beta-Blockers to Prevent Gastroesophageal Vances in Patients with Cirrhosis” followed 213 patients with cirrhosis and portal hypertension for an average of 55 months. The primary outcome of the study was the development of varices or variceal hemorrhage after receiving either timolol or a placebo. Although no difference in the rate of the primary end point was found, moderate to serious adverse events occurred in S2 of 108 patients in the timolol group and 34 of 105 patients in the placebo group (p=0.02).

Given this study, which best denotes the number of people who would need to be treated with timolol for an average of 55 months for one of them to have a moderate or serious adverse event?

• A 6.
• B 15.
• C 50.
• D 127.

Correct Answer: A

‘Answer: A This question asks for the number needed to harm (NNH), which can be calculated by dividing 1 by the absolute risk increase (ARI). n this study, the ARI was calculated by first finding the percent of patients in each group who experienced harm (52/108 = 48% for timolol and 34/109 = 32% for placebo). The absolute percentages are subtracted (48% ~ 32% = 16%) to determine the ARI The NNH is 1 divided by 16% = 6.

You are part of a pharmacist research team evaluating a pharmacy-directed patient care program to diagnose and treat streptococcal pharyngitis in community pharmacies. A journal article evaluating a new rapid group A streptococcal antigen test compared with a throat culture (gold standard) in patients presenting with “sore throat” is identified during a literature search. The article provides the following 2 x 2 table comparing the results obtained from screening 1500 patients.

Which option best depicts the specificity of the new streptococcal antigen test?

• A 86.5%.
• B 91.7%.
• C 98.2%.
• D 98.9%.

Correct Answer: D

Answer: D. Specificity represents the proportion of those without the disease who test negative. In the question above, it is necessary to identify the true negatives (TNs) and false positives (FPs). Specificity is calculated using the formula [TN/(TN + FP)] x 100%.

As the decentralized pharmacist responsible for the neurology floor, you are asked by your clinical coordinator to evaluate the results of the SOCRATES trial, which compared ticagrelor with aspirin for the prevention of recurrent stroke. For the primary outcome of time to occurrence of stroke, myocardial infarction, or death within 90 days, the hazard ratio (HR) for ticagrelor versus aspirin was 0.89 (85% confidence interval [CI] 0.78–1.01).

Which statement best interprets this result?

• A Ticagrelor decreased the risk of a primary outcome within 90 days by 11%, but it was statistically nonsignificant.
• B Ticagrelor increased the risk of a primary outcome within 90 days by 11%, but it was statistically nonsignificant.
• C Ticagrelor decreased the risk of a primary outcome within 90 days by 11%, and it was statistically significant.
• D Ticagrelor increased the risk of a primary outcome within 90 days by 11%, and it was statistically significant.

Correct Answer: A

Answer A: Because the HR is less than 1, the benefits in favor of the intervention group—in this case, ticagrelor subtracting the HR ‘from 1 and multiplying the result by 100% gives the percent benefit—in this case, 11%. Thus, ticagrelor decreased the risk of a primary event by 11%, making options B and D incorrect. In considering the statistical significance of an HR, the 85% CI must not include 1 in order to be considered significant. Because the 95% Cl in this example includes 1, the benefit is not statistically significant. Therefore, [Online Comes! end Ops AEG.

A recent controlled clinical trial published in the New England Journal of Medicine evaluated the use of “intensified” antituberculosis therapy versus standard therapy in adults with tuberculosis meningitis. The study was randomized, double-blind, and placebo-controlled and enrolled adults with a diagnosis of tuberculosis meningitis with or without HIV, admitted to one of two Vietnamese hospitals. Patients in the intensified therapy arm received higher doses of rifampin (15 mg/kg body weight per day) as well as levofloxacin for the first 8 weeks of treatment, followed by standard therapy of rifampin for 9 months. Patients in the placebo arm received 9 months of rifampin 10 mg/kg body weight per day for 9 months. The authors found no significant difference in deaths at 9 months post-randomization between the intensified group and the standard group.

You are rounding with the infectious diseases consult team in the intensive care unit on a patient with HIV and recently diagnosed tuberculosis meningitis. Your attending recounts the recent study noted above and asks for your thoughts on the validity of this study in the context of this patient. Which type of validity would you be most concerned about when applying this study to your patient?

• A. Internal validity because the study was conducted only in Vietnamese hospitals.
• B. External validity because the study was conducted only in Vietnamese hospitals.
• C. Internal validity because the study design did not include the standard of care for treatment of tuberculosis meningitis in patients with HIV.
• D. External validity because the study design did not include the standard of care for treatment of tuberculosis meningitis in patients with HIV.

Correct Answer: B

Internal validity pertains to how well a study was conducted. This includes ensuring that outcomes were evaluated accurately with minimal bias and confounding variables. External validity addresses how well a study is employed and applied to a broader population.

Answer B is correct (and Answer A is incorrect) because this study was only conducted in Vietnamese hospitals and, therefore, may be less generalizable to a U.S. population, limiting its external validity. That this study did not use the standard meningitis tuberculosis treatment consisting of isoniazid, rifampin, pyrazinamide, and ethambutol is a possible limitation, but this would not affect its validity (Answer C and Answer D are incorrect).

A study is developed to evaluate pain management in patients with endometriosis. After 6 months, 95 of the 120 participants in group A and 64 of the 113 participants in group B achieved adequate pain control, as described by a pain score of 7/10 or greater (p = 0.001).

Which best depicts the number needed to treat (NNT) for this outcome?

• A 4.
• B 6.
• C 8.
• D 10.

Correct Answer: A

Answer A: Because p=0.001 (significant), the NT can be calculated. The NNT is determined by calculating the reciprocal of the absolute risk reduction. To calculate the isk reduction, subtract the absolute risk in group A (85/120) from the absolute risk in group 8 (64/113). The absolute risk is 0.23. The reciprocal of 110.23 = 4.34, making Answer A the best answer

Chronic spontaneous urticaria is characterized by recurrent itchy wheals. First-line management is with H1-antihistamines. In a Cochrane review of H1-antihistamines in the treatment of chronic spontaneous urticarial, investigators conducted a systematic search of major databases for randomized controlled trials. Investigators included 73 studies with 9758 participants; 34 studies provided outcome data for 23 comparisons. Compared with placebo, centrizing 10 mg daily in the short and intermediate term, relative risk (HR) 2.72, 95% confidence interval [CI, 1.51–4.91] led to complete suppression of urticaria. Levocetirizine 20 mg daily was effective for short-term use (RR 20.97, 95% CI, 1.37–37.7,60), as was 5 mg for intermediate-term use (RR 52.98, 95% CI, 3.31–843.81). Desforatability 20 mg was effective for the short-term (RR 45.97, 95% CI, 1.04–245.04), as was 5 mg in the intermediate term (RR 37.00, 95% CI, 2.31–563.70). Some methodological limitations were observed; among these were that few studies had been included for each comparison of reported outcome data that could be incorporated into meta-analyses.

Which is the best conclusion that can be drawn from the systematic review described?

• A Several antihistamines are effective in the complete suppression of chronic spontaneous urticaria.
• B No antihistamines are effective in the complete suppression of chronic spontaneous urticaria.
• C Several antihistamines are safe and effective in the complete suppression of chronic spontaneous urticaria.
• D No antihistamines are safe and effective in the complete suppression of chronic spontaneous urticaria.

Correct Answer: A

Answer A. Statistical significance was shown for cetirizine, evocetirizine, and desloratadine because each RR value provided was positive, and each related Cl did not cross 1. No safety data were provided; thus, a conclusion regarding the safety of each agent can not be determined.

You are analyzing data from a heart failure medication use evaluation. You are comparing two groups of patients: those who received a certain heart failure drug and those who did not. Ejection fraction (EF) is recorded as a percentage.

When comparing baseline EFs between groups, which type of data and test would be best to use in your comparison?

• A Continuous t-test.
• B Continuous, chi-square.
• C Dichotomous, t-test.
• D Dichotomous, chi-square.

Correct Answer: A

Election fraction is a continuous variable (Answers C and D are incorrect), which can be quantifiable on an infinite scale. Dichotomous data are not continuous, and a patient can be in only one of two categories (dead, alive, black/white, EF greater than 4DIEF less than 40, fc). Continuous variables are more powerful and are appropriately evaluated by the t-test (Answer A is correct, Answer B is incorrect). Dichotomous data are evaluated appropriately by the chi-square test (Answer C is incorrect)

A randomized controlled trial assesses the effects of the treatment of heart failure on global functioning in three groups of adults after 6 months of treatment. Investigators wanted to assess global functioning with the New York Heart Association (NYHA) functional classification, an ordered scale from I to IV, and to compare the patient classification after 6 months of treatment.

Which statistical test is most appropriate to assess differences in functional classification between the groups?

• A. Kruskal-Wallis test.
• B. Wilcoxon signed-rank test.
• C. Analysis of variance (ANOVA).
• D. Analysis of covariance (ANCOVA).

Answer: A

The NYHA functional class is an ordinal scale from I (no symptoms) to IV (severe symptoms). Neither ANOVA nor ANCOVA is appropriate for ordinal or noncontinuous data (Answers C and D are incorrect). The Wilcoxon signed-rank test is an appropriate nonparametric test to use for paired ordinal data, such as the change in NYHA functional class over time on the same person (Answer B is incorrect). The Kruskal Wallis test is the nonparametric analog of a one-way ANOVA and is appropriate for this analysis (Answer A is correct).

2. You are evaluating a randomized, double-blind, parallel-group controlled trial that compares four antihypertensive drugs for their effect on blood pressure. 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 investigators used an unpaired (independent samples) t-test to test the hypothesis that each drug was equal to the other.

Which statement is most appropriate?

• A. Investigators 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.

2. Answer: C

You cannot determine which finding is more important (in this case, the best drug) on the basis of the p-value (i.e., a lower p-value does not mean more important) (Answer B is incorrect). All statistically significant results are interpreted as significant without respect to the size of the p-value. This trial had four independent samples, and use of the unpaired (independent samples) t-test is not appropriate because it requires several unnecessary tests and increases the chances of making a type I error (Answer A is incorrect). In this setting, ANOVA is the correct test (Answer C is correct), followed by a multiple-comparisons procedure to determine where the actual differences between groups lie. A paired t-test is inappropriate because this is a parallel- group trial (Answer D is incorrect). Use of ANOVA in this case assumes a normal distribution and equal variance in each of the four groups.

3. In the results of a randomized, double-blind, controlled clinical trial, the difference in hospital readmission rates between the intervention group and the control group is 6% (p=0.01), and it is concluded that there is a statistically significant difference between the groups.

Which statement is most consistent with this finding and conclusions?

• A. The chance of making a 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 a type I error is 1 in 100.

3. Answer: D

The typical a priori α error (type I) rate is 5% (i.e., when the study was designed, the error rate was designed to be 5% or less). The actual type I error rate is reported in the question as 0.01 (1%) (Answer A is incorrect). Answers B and C are related; the study did have enough power because a statistically significant difference was observed. Similarly, a type II error was not made because this error has to do with not finding a difference when one truly exists. In this question, the type I error rate is 1%, the value of the p-value (Answer D is correct).

4. You are reading a manuscript that evaluates the impact of obesity on enoxaparin pharmacokinetics. The authors used an unpaired t-test to compare the baseline values of body mass index (BMI) in normal subjects and obese subjects. You are evaluating the use of an unpaired t-test to compare the BMI between the two groups.

Which choice best represents the most appropriate criteria to be met to use this parametric test?

• A. The sample sizes in the normal and obese subjects should be equal to allow the use of a t-test.
• B. A t-test is not appropriate because BMI data are ordinal.
• C. The variance of the BMI data has to be similar in each group.
• D. The pre-study power should be at least 90%.

4. Answer: C

Sample sizes need not be equal to use a t-test (Answer A is incorrect). Body mass index data are not ordinal but continuous; thus, a t-test is appropriate (Answer B is incorrect). The assumption of equal variances is required to use any parametric test (Answer C is correct). A specific value for power is not required to use a test (Answer D is incorrect).

5. You are evaluating the results and discussion of a journal club article to present to the pharmacy residents at your institution. The randomized, prospective, controlled trial evaluated the efficacy of a new controller drug for asthma. The primary end point was the morning forced expiratory volume in 1 second (FEV.) in two groups of subjects (men and women). The difference in FEV. between the two groups was 15% (95% confidence interval [CI], 10%-21%).

Which statement is most appropriate, given the results?

• A. Without the reporting of a p-value, it is not possible to conclude whether these results were statistically significant.
• B. There is a statistically significant difference between the men and women (p<0.05).
• C. There is a statistically significant difference between the men and women (p<0.01).
• D. There is no statistically significant difference between the men and women.

5. Answer: B

The reporting of the mean difference and CI is thought by many to be a superior means of presenting the results from a clinical trial because it describes both precision and statistical significance, as compared with a p-value, which distills everything into one value, making Answer A incorrect. The presentation of the data in this manner clearly shows all the necessary information for making the appropriate conclusion. To assess statistical significance by use of CIs, the 95% CI (corresponding to the 5% type I error rate used in most studies) may not contain zero (signifying no difference between men and women) for the mean difference, making Answer D incorrect. Answer B is correct because the p-value of less than 0.05 corresponds to the 95% CI in that item. To evaluate Answer C, we would need to know the 99% CI.

6. An early-phase clinical trial of 40 subjects evaluated a new drug known to increase high-density lipoprotein cholesterol (HDL) concentrations. The objective of the trial was to compare the new drug’s ability to increase HDL with that of lifestyle modifications (active control group). At the beginning of the study, the mean baseline HDL 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 for the control group was 44 mg/dL and for the new drug group, 49 mg/dL. The p-value for the comparison at 3 months was 0.08.

Which statement provides the best interpretation of these results?

• A. An a priori α 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 concentrations.
• C. The new drug is better than lifestyle modifications because it increases HDL concentrations to a greater extent.
• D. This study is potentially underpowered.

6. Answer: D

Answer A is incorrect because it uses unconventional approaches to determine statistical significance. Although this can be done, it is unlikely to be accepted by other readers and investigators. This study observed a nonsignificant increase in HDL concentration between the two groups. With a small sample size, such as the one used in this study, there is always concern about adequate power to observe a difference between the two treatments. A difference may exist between these two drugs, but the number of subjects studied may be too small to detect it statistically. Answer D is correct because, with the lack of information provided in this narrative, it is not possible to estimate power; thus, more information is needed. Answer B may be correct, but without first addressing the question of adequate power, it would be an inappropriate conclusion to draw. Answer C is incorrect because even though the new drug increased HDL concentration more than the other treatment, it is inappropriate to conclude that it is better because, statistically, it is not.

7. Researchers planned a study to evaluate the percentage of subjects who achieved less than a target blood pressure (less than 140/90 mm Hg) when initiating therapy with two different doses of amlodipine. In the study of 100 subjects, the amlodipine 5-mg group (n=50) and the amlodipine 10-mg group (n=50) were compared. The investigators used a blood pressure goal as their primary end point, defined as the percentage of subjects who successfully achieved the blood pressure goal at 3 months.

Which is the most appropriate statistical test to answer such a question?

• A. Independent samples t-test.
• B. Chi-square or Fisher exact test.
• C. Wilcoxon signed-rank test.
• D. One-sample t-test.

7. Answer: B

The primary end point in this study, the percentage of subjects at or below the target blood pressure, is nominal data. Subjects at target blood pressure (less than 140/90 mm Hg) are defined as having reached the target. This type of data requires either a chi-square test or a Fisher exact test (depending on the sample size or, more accurately, the number of counts in the individual contingency table cells) (Answer B is correct). An independent samples t-test is not appropriate because actual blood pressure values are not being compared (at least not in this question or this end point) (Answer A is incorrect). If we were comparing the actual blood pressure between the two groups, the test might be appropriate if parametric assumptions were met. The Wilcoxon signed-rank test is the appropriate nonparametric test for comparing paired samples (usually in a crossover trial) (Answer C is incorrect). Finally, a one-sample t-test is used to compare the mean of a single group with the mean of a reference group. This is also incorrect in this situation because two groups are being compared (Answer D is incorrect).

8. An investigational drug is being compared with an existing drug for the treatment of anemia in patients with chronic kidney disease. The study is designed to detect a minimum 20% difference in response rates between the groups, if one exists, with an a priori α of 0.05 or less. The investigators are unclear whether the 20% difference between response rates is too large and think a smaller difference might be more clinically meaningful. In revising their study, they decide they want to be able to detect a minimum 10% difference in response.

Which change to the study parameters is most appropriate?

• A. Increase the sample size.
• B. Select an α of 0.001 as a cutoff for statistical significance.
• C. Select an α of 0.10 as a cutoff for statistical significance.
• D. Decrease the sample size

8. Answer: A

Detecting the smaller difference between the treatments requires more power. Power can be increased in several different ways. Answer A is correct because the most common approach is to increase the sample size, which is expensive for the researchers. Answer D is incorrect because smaller sample sizes diminish a study’s ability to detect differences between groups. Power can also be increased by increasing α, but doing so increases the chances of a type I error. Answer B decreases α, thus making it more difficult to detect differences between groups. Answer C certainly makes it easier to detect a difference between the two groups, but it uses an unconventional α value and is thus not the most appropriate technique.

9. You are designing a new computer alert system to investigate the impact of several factors on the risk of corrected QT interval (QTc) prolongation. You want to develop a model to predict which patients are most likely to experience QTc prolongation after the administration of certain drugs or the presence of certain conditions. You plan to assess the presence or absence of several different variables.

Which technique will be most useful in completing such an analysis?

• A. Correlation.
• B. Kaplan-Meier curve.
• C. Regression.
• D. Confidence intervals.

9. Answer: C

Regression analysis is the most effective way to develop models to predict outcomes or variables (Answer C is correct). There are many different types of regression, but all share the ability to evaluate the impact of multiple variables simultaneously on an outcome variable. Correlation analysis is used to assess the association between two (or more) variables, not to make predictions (Answer A is incorrect). Kaplan-Meier curves are used to graphically depict survival curves or time to an event (Answer B is incorrect). Confidence intervals are not used to make predictions (Answer D is incorrect)

Q1: If the study team did not use a structured definition for measurable disease, what type of bias may be occurring?

• A. Allocation
• B. Compliance
• C. Misclassification
• D. Selection

The correct answer is C.

Misclassification bias occurs if structured definitions are not used. This leads to possible issues with including or excluding incorrect patients into the study.

Q2: A study is evaluating a new therapy for prostate cancer. The study excluded patients with a Gleason score of 7 or higher. What type of bias is the study team trying to prevent?

• A. Allocation
• B. Confounding
• C. Measurement
• D. Selection

The correct answer is B.

To try to prevent confounding bias, patients who are sicker may be excluded from a study. This exclusion helps remove bias from any other conditions or the role disease severity plays in contributing to how well the study treatment works.

ARS Q1: A study evaluated at least a 25% reduction in pain as the primary outcome. What type of data is being evaluated for this outcome?

• A. Continuous
• B. Descriptive
• C. Nominal
• D. Ordinal

The correct answer is C - nominal.

You need to decide the type of data described for this specific question. At least a 25% reduction in pain is a type of nominal data because it depicts a “yes or no” situation – at least a 25% reduction in pain occurred or did not occur (meaning less than 25% reduction in pain).

ARS Q1. A study included participants with ECOG performance status of 0 or 1 only. What type of data is ECOG performance status?

• A. Continuous
• B. Descriptive
• C. Nominal
• D. Ordinal

The correct answer is D - ordinal.
You need to decide the type of data described for this specific question. ECOG performance status is ordinal because there is an implied rank or order between numbers in this type of status.

Exemestane for Breast-Cancer Prevention in Postmenopausal Women. 

Outcome = invasive ER positive breast cancer

A study (n = 9,989) was conducted to evaluate a new test developed to improve screening for colorectal cancer. The new test identified colorectal cancer in 379 patients. Colonoscopy identified colorectal cancer in 823 patients. The positive predictive value is 23.6%.

The coefficient of determination is 0.96, and the correlation coefficient is 0.98 for a study evaluating the occurrence of alopecia with a new chemotherapy agent for breast cancer.

What statement is accurate based on these data?

• A. More alopecia as the dose increases
• B. Less alopecia as the dose increases
• C. Alopecia is not affected by increasing the dose
• D. Need more information to answer

The correct answer is A = more alopecia as the dose increases. You use regression concepts to determine this answer. Regression shows strength of a relationship so when the coefficient of determination is close to 1 that means all variation in Y (alopecia) is attributed to X (increasing dose)

A phase 1 clinical trial is being conducted at your institution. The protocol is using a standard 3 + 3 titration design.

• Dose level 1: 0.5 mg/kg – 0 / 3 patients experience DLT
• Dose level 2: 1 mg/kg – 0 / 3 patients experience DLT
• Dose level 3: 1.5 mg/kg – 1 / 3 patients experience DLT, then 3 more patients are enrolled and 0 / 3 experiences DLT
• Dose level 4: 2 mg/kg – 1 / 3 patients experience DLT, then 3 more patients are enrolled and 1 / 3 experiences DLT

Which of the dose levels above would be considered the MTD?

Answer: Dose level 3 (1 out of 6 patients experienced DLT) would be considered the maximum tolerated dose.

A new study (n = 9,989) was conducted to evaluate a new test developed to improve screening for colorectal cancer. The specificity of the new test was 84%, and the sensitivity is 92%.

Which statement is most accurate regarding this new test?

• A. High likelihood to miss a diagnosis
• B. Expect fewer false negatives
• C. Expect many false positives
• D. Low likelihood to make a positive diagnosis

The correct answer is B

Expect fewer false negatives. The sensitivity of the test is high (92%); therefore, one would expect fewer false negatives and less likely to miss a diagnosis. The specificity (84%) is higher for this test. One would expect fewer false positive diagnoses and less likely to identify patients as having the condition when they really do not have it.

The correlation coefficient is 0.98, and the coefficient of determination is 0.96 for a study evaluating a new chemotherapy agent to treat advanced pancreatic cancer.
What percentage best estimates the variation in the outcome due to the independent variable?

• A. 2%
• B. 19%
• C. 96%
• D. 98%

Answer: C = 96%. By definition the coefficient of determination estimates the variation in the outcome due to the independent variable.

The coefficient of determination is 0.98 for a study evaluating if continued smoking affects complete response (CR) of small cell lung cancer when taking a new oral chemotherapy. The alpha level is set at < 0.05. The P value was determined to be 0.052.

What is the best way to interpret these findings?

• A. Smoking positively impacts CR
• B. Smoking negatively impacts CR
• C. Smoking has no effect on CR
• D. The results are inconclusive

The correct answer is C = smoking has no effect on CR. You use regression concepts to determine this answer. Regression cannot show direction of a relationship; it only shows strength of a relationship. The P value is greater than the alpha level, so smoking has no effect on CR.

A study evaluated progression-free survival (PFS) associated with bevacizumab plus olaparib or placebo as first-line maintenance therapy in ovarian cancer. The authors concluded PFS was longer with olaparib.

What results support the authors’ conclusion?

• A. HR 0.59 (95% CI, 0.49 to 0.72)
• B. HR 0.85 (95% CI, 0.58 to 1.25)
• C. HR 1.15 (95% CI, 0.98 to 1.29)
• D. HR 1.21 (95% CI, 1.03 to 1.38)

Answer: A

The correct answer is A – HR 0.59 (95% CI, 0.49 to 0.72). The confidence interval does not cross 1 and the HR is less than 1 indicating a difference in progression-free survival between treatment groups.

A meta-analysis (n = 19 trials) evaluated PD-1 vs PD-L1 inhibitors in adults with cancer. Three individuals reviewed trials for incorporation into the analysis. The funnel plot displayed asymmetry. There was no difference in overall PFS between groups (HR 0.62; 95%CI, 0.37 to 1.05; I² = 83%, and P < 0.001).

What statement is the most appropriate?

• A. Publication bias is not present
• B. More individuals were needed to assess trial eligibility
• C. Additional tests should have been included in the analysis
• D. Heterogeneity exists between trials

The correct answer is D – heterogeneity exists between trials. The question stem states there is asymmetry per the funnel plot. The I² statistic = 83% and Q test P value < 0.001 indicating there is high heterogeneity among trials. More trials and more individuals will not necessarily be needed for analyses.

A noninferiority trial evaluated palonosetron IV infusion compared with palonosetron IV bolus for CINV (total n = 441). A total of 212 subjects in each group were needed to meet power (90%). The prespecified noninferiority margin was -15%. The authors conducted a per-protocol analysis (PPA) after the intention-to-treat analysis. For the PPA, the risk difference was -3.4% (99% CI, -12% to 5.2%). What statement is the most accurate?

• A. The incorrect noninferiority margin was used.
• B. IV infusion is noninferior to IV bolus.
• C. Trial did not meet power so conclusion is valid.
• D. Conduct PPA before intention-to-treat analysis.

The correct answer is B – IV infusion is noninferior to IV bolus. The predetermined noninferiority margin was -15%, which was determined appropriately. For noninferiority, the 99% CI cannot contain this number. Meeting power in this trial did not factor into the conclusion because the products were found to be noninferior. If the products were found to be not noninferior then we would be concerned about power because we would not know if the products were not noninferior because there were not enough individuals to determine if they were noninferior.

The correlation coefficient is 0.95, and the coefficient of determination is 0.90 for a study evaluating the occurrence of vomiting with increasing doses of a new chemotherapy agent to treat metastatic NSCLC. What statement is accurate based on these data?

• A. Less vomiting as the dose increases.
• B. More vomiting as the dose increases.
• C. Vomiting is not affected by increasing dose.
• D. Need more information to determine.

The correct answer is B – more vomiting as the dose increases. You need to use the coefficient of determination (r2) to answer this question. This value can determine a cause and effect relationship. The r² is very close to 1 indicating the variation in nausea (X variable) is attributed to increasing the dose (Y variable).

A study team asks for your advice on determining a primary outcome for a randomized, phase 3 trial evaluating overall survival in patients with breast cancer following treatment with a new oral agent. Which statement is the most correct regarding this new trial?

• A. Shorter follow-up timeframe is needed
• B. Progression-free survival is a better primary outcome
• C. Confounders during follow-up phase may affect results
• D. Overall survival is more susceptible to bias

The answer is C – confounders during follow-up phase may affect results. Confounders in follow-up may include additional therapies and new comorbid conditions. A longer time to death may occur and generally studies looking at OS have a longer follow-up phase. Overall survival is less vulnerable to bias.

Questions 1 and 2 pertain to the following case.

A recently released statin is associated with less myopathy than other currently available statins. After 2 years of use, a retrospective case-control study was undertaken by the manufacturer after 20 different reports of severe myopathy were sent to the U.S. Food and Drug Administration’s (FDA’s) MedWatch program. Risk factors for statin-induced myopathy were not assessed; however, both the cases and the controls of this study had identical diagnostic evaluations and were stratified according to the duration of statin use before the onset of myopathy.

Which type of bias is this study design most susceptible to?

• A. Confounding by indication.
• B. Recall bias
• C. Diagnostic bias
• D. Misclassification

Answer: B

Recall bias is always a potential concern for case control studies because of the amount of time that passes between the study and the drug “ingestion.” Because risk factors were not included in the study design, this is of concern (Answer B is correct). Although a study may be susceptible to many types of bias, the other choices would not pose as much risk (if any) compared with recall bias (Answers A, C, and D are incorrect).

2. Which factor will be most affected by the type of bias likely to occur in this study?

• A. External validity.
• B. Internal validity.
• C. Assessment of exposure.
• D. Number of patients needed for the study.

2. Answer: B

Internal validity is greatly jeopardized because the study is not designed to protect against this possible bias. In a sense, this design flaw jeopardizes external validity (how well does a study apply to other patients with this condition/disease?), but a lack of internal validity is most affected (Answer B is correct). The other answers can be adequately controlled for in the design and conduct of the study (Answers A, C, and D are incorrect).

3. When describing the results of a randomized controlled clinical trial, the investigators report using an intention-to-treat analysis to analyze their data. The results of their investigation comparing two diuretics for heart failure show no difference in the number of hospitalizations for decompensated heart failure between the treatment groups. Given their method of data analysis, which statement is most appropriate?

• A. May be susceptible to issues regarding recall bias.
• B. Provides a good measure of effectiveness under typical clinical conditions.
• C. Cannot provide an estimate of the method’s effectiveness.
• D. May overestimate the actual treatment effect.

3. Answer: B

Intention-to-treat analysis generally considers the approach, which gives the best estimate of use effectiveness (use under typical clinical trial conditions), whereas per-protocol analysis gives a better estimate of method effectiveness (use under ideal conditions) (Answer B is correct; Answer C is incorrect). Intention-to- treat analysis is the most common approach to data analysis for randomized controlled trials and may underestimate the treatment effect (Answer D is incorrect). Recall bias is not a concern with randomized controlled trials (Answer A is incorrect).

4. A prospective randomized study compared once daily enoxaparin with twice-daily enoxaparin when treating patients with venous thromboembolism (VTE). One of the study endpoints was the recurrence of VTE. The following table summarizes recurrence rates in all patients.

The 95% confidence interval (CI) for the difference in recurrence rates between the two groups was –1.5% to 4.5%.

Which conclusion is most appropriate?

• A. Twice-daily enoxaparin is superior to once daily
• B. Superiority of twice-daily enoxaparin could not be established over once daily.
• C. Once-daily enoxaparin is not inferior to twice daily.
• D. No conclusion can be drawn because p-values are unavailable.

4. Answer: B

The CI of the difference in recurrence rate between the two groups includes zero; thus, there is no statistically significant difference between the two groups (Answer B is correct). Answer A is incorrect because the 95% CI contains zero and is therefore not statistically significant.
Answer C is incorrect because not enough information is provided. Answer D is incorrect because all the above information can be determined without the benefit of reported p-values.

5. According to the data in the previous question and the result obtained, which best represents the number of patients who would need to be treated with twice-daily enoxaparin to prevent the recurrence of one VTE episode?

• A. Number needed to treat (NNT) would be 2.
• B. NNT would be 67.
• C. NNT would be 152.
• D. NNT should not be calculated because the result was non-significant.

5. Answer: D

Answers A and C are incorrect calculations. Calculating the NNT to prevent one recurrence using twice-daily therapy is as follows: 0.044 - 0.029 = 0.015 and 1/0.015 =66.7...67; however, the NNT should not be calculated when the end point of interest is nonsignificant (Answer B is incorrect; Answer D is correct).

Questions 6 and 7 pertain to the following case.

A multicenter, double-blind, placebo-controlled trial randomly assigned 4837 patients to treatment with margarine supplemented with the omega-3 fatty acid α-linolenic acid (ALA) (margarine with ALA) or a placebo margarine. The primary combined end point was the rate of cardiovascular events, defined as fatal and nonfatal cardiovascular events and percutaneous coronary interventions. Data were analyzed according to intention-to-treat analysis with the use of a Cox proportional hazards model. The hazard ratio (HR) and 95%CI for the margarine with ALA group were 0.91 and 0.78–1.05, respectively. In the prespecified subgroup of women, margarine with ALA was associated with an HR of 0.73 (95% CI, 0.51–1.03).

6. Which statement is most appropriate?

• A. Margarine with ALA statistically significantly reduced the risk of cardiovascular events (p<0.05).
• B. Margarine with ALA statistically significantly reduced the risk of cardiovascular events (p<0.01).
• C. Margarine with ALA did not significantly reduce the risk of cardiovascular events (p>0.05).
• D. Without a p-value, it is not possible to determine whether margarine with ALA affected cardiovascular events.

6. Answer: C

Answers A and B are incorrect because the margarine with ALA did not significantly reduce the risk of cardiovascular events (the 95% CI includes 1 [no difference in risk]). Answer D is incorrect because the p-value is not required for interpreting statistical significance when the 95% CI is provided. Answer C is correct because the p-value corresponds to the 95% CI.

7. When the study was being designed, which choice describes the outcome for which the study was most likely to have been powered?

• A. Differences in the rate of the composite outcome, cardiovascular events.
• B. Differences in the rate of percutaneous coronary interventions.
• C. Differences in the rate of the composite outcomes in women.
• D. Differences in the rate of the composite outcomes in men.

7. Answer: A

Clinical trials are usually adequately powered to compare primary end points (Answer A is correct). Because Answer B is part of the composite outcomes, the study was likely not powered to detect this outcome independently. Similarly, even though the subgroup analysis was determined a priori, the study is not typically designed to have sufficient power to make this comparison (Answers C and D are incorrect).

8. In a meta-analysis of studies examining the effects of several antihypertensive drugs, the odds ratio (OR) for treatment with low-dose diuretics compared with calcium channel blockers for cardiovascular disease events was 0.84 (95% CI, 0.75–0.95).

Which statement is the most appropriate interpretation of these findings?

• A. Treatment of hypertension with low-dose diuretics was more effective in preventing cardiovascular disease events than treatment with calcium channel blockers.
• B. Treatment of hypertension with calcium channel blockers was more effective in preventing cardiovascular disease events than treatment with low doses of diuretics.
• C. The difference observed between treatment with calcium channel blockers and low doses of diuretics is not statistically significant.
• D. The odds of developing cardiovascular events when treating hypertension with low doses of diuretics are lower than when using calcium channel blockers.

8. Answer: D

Answers A and B are incorrect because each implies that one drug is more effective than the other. In this type of study design, neither drug is more/less effective. Answer C is incorrect because the CI of the OR does not include 1; thus, the finding is statistically significant at the 5% level, making Answer D correct.