Biostats exam 2 quiz questions and answers

In a normally distributed dataset of systolic blood pressures (mean=130 mmhg, SD=10mmHg), approximately what percentage of patients would be expected to have systolic BP between 110 and 150 mmhg?

(B) Approximately 95%, capturing patients within two standard deviations of the mean

A clinical trial of a new antiplatelet agent reports a statistically significant reduction in stroke risk (p=0.03). A pharmacist evaluating the study notes that the ARR is 0.3%. What conclusion is most appropriate?

(C) The result statistically significant but may lack meaningful clinical significance

A researcher conducts a hypothesis test and concludes that a new antiemetic is superior to ondansetron. In truth, the two drugs are equally effective. Which type of error has occurred?

(B) Type I error-incorrectly rejecting the null hypothesis when it is actually true

A researcher uses a 95% CI to assess whether two anticoagulants differ in mean INR. The CI for the difference is (-0.15, 0.42). What is the correct statistical conclusion?

(B) Fail to reject H0 because the CI includes the null value of zero

A hospital formulary committee wants to determine whether a new oral anticoagulant produces fewer bleeding events than warfarin. Phase II data strongly suggests the new drug is safer. What test type is most appropriate?

(C) Test of superiority (one-sided) because prior data supports new drug being better

A researcher fails to find a statistically significant difference in cure rates between a 5-day and 10 day course of azithromycin for community acquired pneumonia (p=0.38). A colleague concludes the two regimens are therefore equivalent.

Which statement BEST evaluates the conclusion?

(C) The conclusion is incorrect bc equivalence requires a specific TOST or CI procedure, not a significant difference test

A research team is testing whether a new antiseptic hand gel reduces MRSA transmission rates in a hospital setting. They hypothesize that transmission rates with the new gel will be lower than rates with standard gel. Which type of alternative hypothesis is most appropriate for this study?

(B) A directional (one tailed) alternative hypothesis, because the researchers predict a specific direction of effect

A pharmacy researcher reports: "Using a two-sample t-test, the mean blood glucose reduction was 28 mg/dL in the treatment arm versus 14 mg/dL in the placebo arm (p=0.008)." How should this p-value be correctly interpreted?

(B) There is a 0.8% probability that a difference this large would occur by chance if the null hypothesis is true

A clinical pharmacist at a large academic medical center reviews a phase III trial of a novel anticoagulant versus warfarin (n=12400 per arm). The study reports a statistically significant reduction in stroke incidence (p=0.04, absolute risk reduction=0.3%). The pharmacist concludes the result is statistically significant but clinically unimportant. Which combination of principles best supports this assessment?

Large sample sizes increase statistical power, enabling detection of very small effects that may lack meaningful clinical impact

A clinical pharmacist wants to compare fasting blood glucose levels in the same group of patients before and after initiating metformin therapy. Which statistical test is most appropriate?

(B) Paired samples t-test using each patient as their own control

A pharmacy researcher states: "Our study comparing warfarin INR before and after pharmacist-led counseling yielded p=0.03. This means there is a 3% probability that the null hypothesis is true." Which statement best corrects this interpretation?

(B) The p-value reflects the probability of results this extreme if H0 were already true

SELECT ALL THAT APPLY. Which of the following correctly describe why a 95% confidence interval provides more clinically useful information than a p-value alone when evaluating a pharmacist-led diabetes management program?

(A) The CI reveals the magnitude and direction of the treatment effect on HbA1c (B) The CI quantifies the precision of the estimate through the width of the interval (D) The CI enables clinicians to judge whether the effect size is therapeutically important

A pharmacy researcher wants to compare mean HbA1c reductions across four different diabetes medication classes: biguanides, sulfonylureas, GLP-1 agonists, and SGLT-2 inhibitors. Which statistical test is most appropriate?

(B) One-way ANOVA using medication class as their single factor

Which of the following study designs satisfy ALL four assumptions required for one-way ANOVA? SELECT ALL THAT APPLY.

(A) Systolic BP reduction (mmHg) in 90 patients randomized to three antihypertensive drugs with equal group sizes and similar group variances (C) Total cholesterol (mg/dL) in three independent patient cohorts, each normally distributed, with comparable standard deviations

A pharmacy school implements three different teaching modalities (lecture-only, flipped classroom, and simulation-based) for pharmacokinetics. An ANOVA yields F=4.87, p=0.011. What is the correct interpretation?

(B) At least one teaching modality produces a significantly different mean exam score from the others

A pharmacy student wants to compare the average blood pressure readings of 45 hypertensive patients before and after a new antihypertensive medication. A shapiro-wilk test confirms the data is normally distributed with no significant outliers. Which test type and specific test are most appropriate?

(B) Parametric; paired samples t-test

An investigator is studying whether LDL cholesterol levels differ among patients assigned to one of four treatment groups (placebo, low-dose statin, high-dose statin, and combination therapy). The sample includes 120 patients (30 per group), and the data is approximately normally distributed with equal variances across groups. Which test type and specific test are most appropriate?

(C) Parametric; one-way ANOVA

A clinical trial enrolls 500 patients and compares pain relief scores between an experimental NSAID and a placebo. Each group has 250 patients. The data is continuous. A Shapiro-Will test shows statistically significant non-normality (p=0.03), but a histogram reveals the distribution is only slightly skewed with no extreme outliers. Which statement best guides the choice of statistical test?

(B) The Central Limit Theorem supports using a parametric independent samples t-test given the large sample size

A clinical pharmacist is reviewing a study that evaluated whether vancomycin trough concentration (measured in mg/L) can be predicted from creatinine clearance (mL/min). The outcome variable is a continuous laboratory value. Which regression method is most appropriate for this analysis?

(B) Linear regressions modeling trough level as a continuous outcome

A study uses linear regression to examine the effect of metformin dose (mg/day) on fasting plasma glucose (mg/dL) in patients with type 2 diabetes. The regression coefficient (B) for metformin dose is -0.04. Which of the following is the correct interpretation?

(C) For each 1 mg increase in dose, glucose falls by 0.04 mg/dL on average

A retrospective cohort study uses logistic regression to examine the association between fluoroquinolone use and clostridioides difficile infection. After adjusting for age and comorbidities, the odds ratio (OR) is 3.4 with a 95% confidence interval of 1.9 to 6.1. Which interpretation is correct?

(D) Fluoroquinolone users have 3.4 times the adjusted odds of C. Diff infection