Biostatistics & Pharmacoeconomics I

Review terms: What is alpha in statistics?

Alpha is usually expressed as a proportion. Thus, if the confidence level is 95%, then alpha would equal 1 - 0.95 or 0.05. With respect to hypothesis tests , alpha refers to significance level , the probability of making a Type I error .

Review terms: What is type 1 error

Type I error is the incorrect rejection of a true null hypothesis (a "false positive"), while

a type II error is incorrectly retaining a false null hypothesis (a "false negative").

Review terms: What is a null hypothesis

In a statistical test the hypothesis that there is no significant difference between specified populations, any observed difference being due to sampling or experimental error.

Review of terms: What is P-value

The P value, or calculated probability, is the probability of finding the observed, or more extreme, results when the null hypothesis (H 0) of a study question is true.

Review of terms: When the P value is less than 0.05?

In the majority of analyses, an alpha of 0.05 is used as the cutoff for significance. If the p-value is less than 0.05, we REJECT the null hypothesis that there's no difference between the means and conclude that a significant difference DOES exist

Review of terms: Types of data

Ordinal - Ordinal data is a categorical statistical data type where the variable has natural, ordered categories and exist on an ordinal scale. An ordinal variable is a categorical variable for which the possible values are ordered. Ordinal variables can be considered "in between" categorical and quantitative variables.

Example: Educational level might be categorized as

1: Elementary school education

2: High school graduate

3: College graduate

4: Graduate degree

Nominal - nominal scale, it provides a name or category for each object

Continuous

Random

Interval - interval scale, the same difference at two places on the scale has the same meaning.

In a small, randomized, double-blind trial of a new treatment in patients with acute myocardial infarction, the mortality in the treated group was half that in the control group, but the difference was not significant. We can conclude that:

E) The clinical trial may not have adequate statistical power.

NOTE: The trial is small and some benefit was seen with the new treatment. Many times smaller studies are done to evaluate initial effects (many times referred to as pilot studies). Performing a larger trial will increase statistical power and may then show a statistically significant benefit.

A report of a clinical trial of a new antipyretic (Feverstop) versus placebo, concluded that the new drug gave a higher proportion of success than did the placebo. The study indicated an alpha of 0.05. The report ended with the statement that the trial was statistically significant (p-value < 0.05). In light of this information, we may conclude:

C) The probability of the reported finding or one more extreme by chance is less than 1 in 20.

NOTE: The p-value refers to the null hypothesis which is Group 1 = Group 2 (meaning there is no difference between the drug and the placebo). When the p-value is < alpha, the probability of obtaining a test stastic as extreme as the one actually observed is < alpha. In this case (with alpha set at 0.05), p < 0.05 means that the probability of getting the result obtained is < 5%, under the conditions of the null hypothesis. This means that the null hypothesis should be rejected, because the result is unlikely to occur (< 5% of the time) under that condition.

What is the term used for the inactive or inert substance administered to a patient that serves as a reference to determine the effects of a real drug or other intervention?

D) The placebo

A very large study has measured the body weight of everyone living west of the Mississippi River. If the values of all the weights were plotted, the graph would resemble this shape:

A) bell-shaped curve

NOTE: If many data points are plotted from a large population group, the distribution would be normal (a normal distribution), or Gaussian, or bell-shaped curve.

A researcher gathered all vitamin E studies from the past ten years. Vitamin E was used for a variety of conditions. The populations studied as well as the vitamin E formulations and doses were all different. However, the researcher did the best he could and compared the incidences of cardiovascular-related mortality in those taking vitamin E supplements versus those that did not. Describe this type of study:

E) Meta analysis

A meta analysis is a systematic review of many different but related studies in order to integrate the results. These are often done and can be useful to pool smaller trials into a larger group for analysis (e.g., Cochrane reviews).

The pharmacist is presenting a clinical trial to the medical team. The trial includes a 95% confidence interval for the primary endpoint. If the confidence interval was greater (98% or 99%), what effect will be seen with the interval?

A) The interval will be wider.

Having a higher confidence interval will widen the interval. For example, 99% CIs are wider than 95% CIs. 95% CIs are wider than 90% CIs.

A pharmacist is considering which beta blocker should be preferred at his institution. He has narrowed his search down to two agents. Each drug provides similar health benefits, has similar tolerability and is dosed once daily. The pharmacist will base his decision on the drug that can be purchased at the lower cost. He will use the following analysis to choose the beta blocker for his institution's formulary:

A) A cost minimization analysis

Once the benefit of various drugs is considered equivalent (such as choosing which ACEIs or ARBs to have on the formulary), the next step is to find which one is less expensive.

A pharmacy intern has been asked by his preceptor to gather 200 discharged patient charts from the chart room. He has been told that half the charts should be patients who received proton pump inhibitor (PPI) therapy while hospitalized. The other half should be patients with similar conditions and length of stay but who did not receive PPI therapy. The pharmacist wishes to conduct a study to see if there is any difference in the incidence of nosocomial infection in the PPI group versus the non-PPI group. Which of the following describes this type of study? (Select ALL that apply.)

A) Cohort study

E) Observational study

A cohort study is a type of observational study. This study describes a retrospective cohort (the cohort is discharged patients). These patients are then evaluated to determine if they were receiving PPIs, and then the outcome of nosocomial pneumonia is assessed.

To be a case control study, one would need to start out by identifying the outcome (the patients who developed pneumonia) which was NOT done at the onset of the study.

A trial compared Drug A to placebo and found that Drug A lowered LDL by an additional 5%, on average. If the alpha is set to 0.05 and the p-value was reported to be 0.03. Which of the following is/are true regarding this trial? (Select ALL that apply.)

A) Drug A is stastically significantly better than the placebo

C) The chance (read probability) of a Type I error is less than 5%

E) The investigator should fail to accept the null hypothesis

A measure of statistical significance is not the same as "clinical significance". For example, if a drug lowers LDL by 5%, it may be statistically significant versus the placebo, but clinically it probably would not be used much since other drugs lower LDL to a much greater degree.

Marital status (married, single, divorced) can be described as this type of data:

B) Nominal - provides name (don't mix it up with a number)

Discrete data can be divided into groups or categories (e.g., marital status, gender, ethnicity). These are called nominal discrete values.

What is the definition of a Type I error?

B) The null hypothesis is true, but is rejected in error.

A Type I error means that the null hypothesis is true but is rejected in error. For example, if the p-value is 0.05, it means thatless than 5% of the time the null hypothesis will be rejected in error, or a type I error will have occurred.

CASE: Strongheart Study for Patients with NYHA Functional Class II-IV

BACKGROUND: Strongheart is a new drug being investigated for chronic heart failure patients.

Question:

From the abstract above, what is the relative risk of the primary endpoint in the Strongheart group versus the placebo group? (Round to the nearest HUNDREDTH. Enter the relative risk as a decimal, not a percentage).

0.67

The relative risk is the probability of the event occurring in the exposed group versus the non-exposed group.

Calculation:

AR (absolute risk) = the number of events (good or bad) in treated or control groups, divided by the number of people in that group

ARC = the AR of events in the control group

ART = the AR of events in the treatment group

RR (relative risk) = ART / ARC

(146/1990) : (219/2001) = 0.67

Strongheart Study for Patients with NYHA Functional Class II-IV

Question :

From the abstract above, what is the relative risk reduction of the primary endpoint of all-cause mortality in patients randomized to the Strongheart group? (Round to the nearest WHOLE number. Put your answer in percentage form. Do not include the percentage sign when you type in the number.)

Answer: 33 %

RRR (relative risk reduction) = RRR = 1 - RR

Calculation:

RRR = 1-RR

Relative risk reduction is defined as 1 minus relative risk. The relative risk is 0.67, so the relative risk reduction is 0.33 (1-0.67), or 33%.

Risk Terms:

AR (absolute risk) = the number of events (good or bad) in treated or control groups, divided by the number of people in that group

ARC = the AR of events in the control group

ART = the AR of events in the treatment group

ARR (absolute risk reduction) = ARC - ART

RR (relative risk) = ART / ARC

RRR (relative risk reduction) = (ARC - ART) / ARC

RRR = 1 - RR

NNT (number needed to treat) = 1 / ARR

Strongheart Study for Patients with NYHA Functional Class II-IV

Results: All-cause mortality was lower in Strongheart group than in the placebo group (146 patient-deaths versus 219 patient-deaths; p = 0.00009, [95% CI 0.53-0.81]). There were fewer sudden deaths in the Strongheart group than in the placebo group (79 vs. 132; p = 0.0002) and deaths from worsening heart failure (30 vs 58; p = 0.0023).

Question:

From the abstract above, what is the absolute risk reduction of the primary endpoint of all-cause mortality in patients randomized to the Strongheart group? (Round to the nearest TENTH. Put your answer in percentage form. Do not include the percentage sign when you type in the number.)

ARR = ARC - ART

ARC - absolute risk of events in the control group

ART - absolute risk of events in the treatment group

ARR = (219/2001) - (146/1990) = 3.6

The absolute risk reduction is the absolute difference in outcome rates between two groups. 146/1,990 = 0.073, or 7.3%. 219/2,001 = 0.109, or 10.9%. 10.9% - 7.3% = 3.6%.

Strongheart Study for Patients with NYHA Functional Class II-IV

RESULTS: The study was stopped early on the recommendation of the independent safety committee. Mean follow-up time was 1 year. All-cause mortality was lower in Strongheart group than in the placebo group (146 patient-deaths versus 219 patient-deaths; p = 0.00009, [95% CI 0.53-0.81]). There were fewer sudden deaths in the Strongheart group than in the placebo group (79 vs. 132; p = 0.0002) and deaths from worsening heart failure (30 vs 58; p = 0.0023).

FROM: From the abstract above, what is the number needed to treat with Strongheart to prevent 1 death? Round up to the nearest WHOLE number.

ANSWER: 28

NNT (number needed to treat) = 1 / ARR

NNT = 1/3.6% (0.036) = 28

Note: The number needed to treat is defined as 1/ARR, or 27.7 rounded up to 28. Therefore, 28 patients need to be treated for 1 year to prevent 1 death. Be careful not to confuse the secondary outcomes (sudden death and deaths from worsening heart failure) with the primary outcome of the study, which was all-cause mortality.

A new drug for stroke prevention, CLOTBUST, was studied. The trial enrolled 2,546 patients. In the CLOTBUST arm, 39 out of 1,281 patients had a stroke compared with 64 patients out of 1,265 patients in the aspirin arm, p-value = 0.025. What is the number needed to treat to prevent one stroke in this trial?

NNT = 1/ARR

ARR = ARC - ART = (64/1265) - (39/1281) = 0.050- 0.030 = 0.02

NNT = 1/0.02 = 50

Number needed to treat (NNT) is 1/ARR. Therefore, 1/0.02 = 50 patients. Fifty patients will need to be treated to prevent one stroke

A clinical trial is published about a new drug, NOPAINZ, for pain reduction. There are 200 patients randomized; 100 patients to NOPAINZ and 100 patients to placebo. The study found that pain was alleviated in 90 patients taking NOPAINZ and in 50 patients taking placebo. Calculate the relative risk in this trial. (Enter the relative risk as a decimal, not a percentage.)

0.2

NOTE: Relative risk is the likelihood of an unfavorable event (in this case, pain) occurring in the exposed group versus in the non-exposed group. Therefore, 10 of the 100 patients had pain in the Nopainz group, and 50 of the 100 patients had pain in the placebo group. Therefore, 0.1/0.5 is 0.2.

A clinical trial is conducted on a new drug, LoSod. LoSod is found to reduce serum sodium to 131 mEq/L, with a 95% confidence interval of 128.7 mEq/L to 133.3 mEq/L. What is the correct interpretation of this 95% confidence interval?

A) There is a 95% chance that the interval contains the true population value.

NOTE: A confidence interval gives an estimated range of values which is likely to include an unknown population parameter; the estimated range being calculated from a given set of sample data.

In biostatistics, what is meant by the mode in a group of values?

C) The value that occurs most frequently

A hospital pharmacist presents an article at journal club. She reports that the average weight of the subjects in the study is 172 pounds, with a standard deviation of 10. Choose the correct statement:

B) Most of the subjects had a weight within 10 pounds of the mean.

The standard deviation (SD) shows how much variation there is from the average. For example, if the average weight in a study population is 172 pounds, with a SD of zero, then all subjects weigh the same (172 pounds). If the SD is 10, then roughly 68% of the subjects have a weight within 10 pounds of the mean (from 162 to 182 pounds).

A clinical trial is published about a new drug, NOPAINZ, for pain reduction. There are 200 patients randomized; 100 patients to NOPAINZ and 100 patients to placebo. The study found that pain was alleviated in 90 patients taking NOPAINZ and in 50 patients taking placebo. Calculate the relative risk reduction (RRR) in this trial. (Enter the relative risk reduction as a decimal, not a percentage.)

0.8

Relative risk reduction (RRR) = 1-RR (relative risk). The relative risk is 0.2. Therefore, 1-0.2 = 0.8.

A clinical trial is published about a new drug, NOPAINZ, for pain reduction. There are 200 patients randomized; 100 patients to NOPAINZ and 100 patients to placebo. The study found that pain was alleviated in 90 patients taking NOPAINZ and in 50 patients taking placebo. Calculate the absolute risk reduction (ARR) in this trial. (Enter the absolute risk reduction as a decimal and not a percentage.)

0.4

Absolute risk reduction (ARR) = 0.5 - 0.1 = 0.4

Choose the best description of the purpose of a pharmacoeconomic analysis:

A) To identify and measure the costs and the consequences of drugs.

Pharmacoeconomics takes into account broader costs beyond just drug acquisition costs and is an important tool to ensure resources are used wisely.

BACKGROUND: Vitamin K antagonists are highly effective in preventing stroke in patients with atrial fibrillation but have several limitations. Apixaban is a novel oral direct factor Xa inhibitor that has been shown to reduce the risk of stroke in a similar population in comparison with aspirin.

RESULTS: The median duration of follow-up was 1.8 years. The rate of the primary outcome was 1.28% per year in the apixaban group, as compared with 1.60% per year in the warfarin group

From the abstract above, what is the relative risk reduction of the primary outcome in patients receiving apixaban compared to warfarin? Round to the nearest WHOLE number. (Enter your answer as a percentage but do not enter the percentage sign.)

20

Relative risk reduction is defined as 1 minus relative risk. The relative risk is calculated as 0.8, so the relative risk reduction is 0.20 or 20% (1-0.8).

BACKGROUND: Vitamin K antagonists are highly effective in preventing stroke in patients with atrial fibrillation but have several limitations. Apixaban is a novel oral direct factor Xa inhibitor that has been shown to reduce the risk of stroke in a similar population in comparison with aspirin.

RESULTS: The median duration of follow-up was 1.8 years. The rate of the primary outcome was 1.28% per year in the apixaban group, as compared with 1.60% per year in the warfarin group

From the abstract above, what is the absolute risk reduction of the primary outcome in patients receiving apixaban compared to warfarin? (Enter your answer as a percentage but do not enter the percentage sign.)

0.32

The absolute risk reduction is the absolute difference in outcome rates between two groups. 1.6% - 1.28% = 0.32%.

BACKGROUND: Vitamin K antagonists are highly effective in preventing stroke in patients with atrial fibrillation but have several limitations. Apixaban is a novel oral direct factor Xa inhibitor that has been shown to reduce the risk of stroke in a similar population in comparison with aspirin

From the abstract above, what is the number needed to treat with apixiban compared to warfarin to prevent one incidence of the primary outcome?

NNT (number needed to treat) = 1 / ARR

313

The number needed to treat is 1/ARR. ARR = 1.6 - 1.28 = 0.32%; therefore, 1/0.0032 = 312.5, or 313.

BACKGROUND: Vitamin K antagonists are highly effective in preventing stroke in patients with atrial fibrillation but have several limitations. Apixaban is a novel oral direct factor Xa inhibitor that has been shown to reduce the risk of stroke in a similar population in comparison with aspirin.

The rate of hemorrhagic stroke was 0.24% per year in the apixaban group, as compared with 0.47% per year in the warfarin group

From the abstract above, what is the absolute risk reduction of hemorrhagic stroke? (Enter answer as a percentage but do not enter the percentage sign.)

0.23

ARR = ARC - ART

0.47-0.24=0.23

NOTE: The absolute risk reduction is the absolute difference in outcome rates between two groups. 0.47% - 0.24% = 0.23%. This question specifically asks about a secondary endpoint. If the question referred to the primary endpoint, the data for the composite endpoint (ischemic or hemorrhagic stroke or systemic embolism) must be used.

A new drug for stroke prevention, CLOTBUST, was studied. The trial enrolled 2,546 patients. In the CLOTBUST arm, 39 out of 1,281 patients had a stroke compared with 64 patients out of 1,265 patients in the aspirin arm, p-value = 0.025. What was the relative risk reduction of stroke in this trial?

C) 0.4

Relative risk reduction = 1-RR. Since the RR is calculated to be 0.6, the relative risk reduction is 1 - 0.6 = 0.4.

The JUPITER trial studied the benefits of rosuvastatin 20 mg PO daily versus placebo. The primary outcome of the trial was the occurrence of a first major cardiovascular event, defined as nonfatal myocardial infarction, nonfatal stroke, hospitalization for unstable angina, an arterial revascularization procedure, or confirmed death from cardiovascular causes. At the time of study termination, 142 first major cardiovascular events had occurred in the rosuvastatin group, as compared with 251 in the placebo group, p-value = 0.00001. Each arm had 8,901 patients. What was the relative risk reduction if randomized to rosuvastatin in this trial?

A) 43%

Relative risk reduction = 1-RR. Event rates per arm: 142/8,901 = 1.6% in rosuvastatin arm and 251/8,901 = 2.8% in placebo arm. 1.6/2.8 = 0.57. Therefore, 1-0.57 = 0.43, or 43%.

A new drug for stroke prevention, CLOTBUST, was studied. The trial enrolled 2,546 patients. In the CLOTBUST arm, 39 out of 1,281 patients had a stroke compared with 64 patients out of 1,265 patients in the aspirin arm, p-value = 0.025. What was the relative risk of stroke in this trial?

0.6

The relative risk of stroke in this trial is calculated by taking the risk of stroke in each arm: 39/1,281 = 0.03 and 64/1,265 = 0.05. Therefore, 0.03/0.05 = 0.6.

Blood pressure, hemoglobin A1C and LDL cholesterol can each be described as this type of data:

C) Continuous

Continuous variables can take an infinite number of possible values (within a range), such as height, weight, A1C, blood pressure. For example, a patient's A1C can be 7.2 or 7.3 or 7.4 and so on.

A pharmacist is presenting the Helsinki trial which investigated gemfibrozil 1,200 mg daily versus control. The results showed that CHD events were 2.7% in the gemfibrozil arm versus 4.1% in the control group, with a p-value of < 0.02. What is the relative risk reduction of CHD events in this trial?

B) 34%

Relative risk reduction = 1-RR. Since RR = 0.66, then relative risk reduction is 1 - 0.66 = 0.34.

In biostatistics, what is meant by the median of a group of values?

B) The value in the middle of a ranked list

The median is found by arranging all the observations in numerical order (lowest to highest or highest to lowest, or ranked) and selecting the value in the middle of the list.

A pharmaceutical company wished to show that their antiplatelet agent worked better than placebo. They enrolled 12,000 patients at many different research sites. They worked with the physicians to ensure that the patients were randomly assigned to the antiplatelet agent or to a placebo. Both the patients and researchers did not know which patients received the active drug. Choose the correct statement:

C) The trial was double-blinded

Double-blinding, the best type of controlled clinical study, means that neither the subjects nor the researchers know which subjects received the active drug and which received the placebo.

Over 35,000 nurses were studied in a longitudinal study based in Framingham, Massachusetts. Each year, the nurses were followed up and asked to report on any incidence of heart disease. The researchers wanted to study incidence of heart disease in the subjects using hormone therapy versus those who did not use hormone therapy. Describe this type of study:

D) Cohort study

A study group (the cohort) is followed over time and outcomes (the disease, or heart disease in this example) are compared to a subset of the group who were not exposed to an intervention, such as a drug. This example is prospective in design.

Choose the name of the term used to determine statistical significance in a trial:

C) The p-value

In most clinical trials, the statistical significance is defined as a p-value less than 0.05.

A pharmaceutical company wished to show that their antiplatelet agent worked better than placebo. They enrolled 12,000 patients at many different research sites. They worked with the physicians to ensure that the patients were randomly assigned to the antiplatelet agent or to a placebo. Both the patients and physicians did not know which patients received the active drug. Describe this type of trial:

B) Controlled clinical trial

A study group is compared to one or more control (comparison) groups in a controlled setting.

BACKGROUND:

Angiotensin-converting-enzyme inhibitors improve the outcome among patients with left ventricular dysfunction, whether or not they have heart failure.

METHODS:

A total of 9,297 high-risk patients who had evidence of vascular disease or diabetes plus one other cardiovascular risk factor and who were not known to have a low ejection fraction or heart failure were randomly assigned to receive ramipril or matching placebo daily for a mean of five years. The primary outcome was a composite of myocardial infarction, stroke, or death from cardiovascular causes.

RESULTS:

A total of 4,645 patients were assigned to receive ramipril 10 mg daily and 4,652 were assigned to receive placebo. Of the patients taking ramipril, 650 patients reached the primary endpoint as compared with 826 patients who were assigned to receive placebo (p < 0.001).

Q: From the abstract above, calculate the absolute risk reduction of the composite primary endpoint of myocardial infarction, stroke, or death from cardiovascular causes. (Round to the nearest TENTH. Put your answer in percentage form. Do not include the percentage sign when you type in the number.)

3.8

The absolute risk reduction is the absolute difference in outcome rates between two groups.

ARR = ARC - ART

риск абсолютный, бери по модулю, модно отнимать друг от друга

Same case as above,

Question:

From the abstract above, what is the number needed to treat to prevent the composite endpoint of myocardial infarction, stroke, or death from cardiovascular causes? Round up to the nearest WHOLE number.

NNT = 1/ARR

ARR = /ART-ATC/

NNT= 1/3.8 =26.3

27

The number needed to treat is defined as 1/ARR, or 26.3 rounded up to 27. Therefore, 27 patients need to be treated for 5 years to prevent 1 composite endpoint.

See problem above

Question:

From the abstract above, calculate the relative risk of the primary endpoint in the ramipril group compared to the placebo group.

RR= ART/ARC

E) 0.79

The relative risk is the probability of the event occurring in the exposed group versus the non-exposed group. The risk of the primary outcome is 14% in the ramipril group versus 17.8% in the placebo group. 14%/17.8% gives a relative risk of 0.79.

Same problem from above

Question:

From the abstract above, what is the relative risk reduction of the primary endpoint of myocardial infarction, stroke, or death from cardiovascular causes?

D) 21%

RRR = 1-RR

Relative risk reduction is defined as 1 minus relative risk. The relative risk is calculated as 0.79, so the relative risk reduction is 0.21 (1-0.79).

A pharmacist is presenting the Helsinki trial which investigated gemfibrozil 1,200 mg daily versus control. The results showed that coronary heart disease (CHD) events were 2.7% in the gemfibrozil arm versus 4.1% in the control group, with a p-value of < 0.02. What is the relative risk of CHD events in this trial?

C) 66%

The relative risk of CHD events in this trial is calculated by taking the risk of each arm: 2.7%/4.1% = 0.66, or 66%.

A clinical trial reports that patients taking a new drug, NO-GO, experience less incontinence episodes. The number of incontinent episodes (the primary endpoint) was 16.4% in the NO-GO group and 28.7% in the placebo group. The trial randomized 600 patients (300 patients in each arm). What is the absolute risk reduction? (Do not include the percentage sign when you type in the number.)

ARR = /ART - ARC/

12.3

The absolute risk reduction is calculated by subtracting the event rates of the two groups. Therefore, 28.7% - 16.4% = 12.3%

Correlation in a clinical trial describes:

B) The relationship between two variables

If the data are correlated and the data points are plotted, a straight line can be drawn that will run through, or close to, most of the data points. Data can be positively or negatively correlated depending on the direction of the slope of the line. If the data points are not correlated, the data points will be scattered all over the plot.

Choose the name of the assertion that a clinical trial is designed to disprove which states there is no difference between two groups.

B) The null hypothesis

A clinical trial is testing the null hypothesis which states there is no difference between the two treatment groups.

Choose the categories that represent direct medical costs: (Select ALL that apply.)

A) Cost of a prescription

C) Cost of a surgery

D) Length of a hospital stay

The cost of drugs, the cost of surgery, hospital length of stay are all examples of direct costs.

A clinical trial evaluated the effects of a chemotherapeutic given to patients with osteosarcoma. The trial duration was three months and the primary endpoint was mortality. During this time, there were two deaths among patients receiving placebo and one death among patients receiving active drug with an equal amount of patients randomized to each arm. It could be stated that the drug decreased the risk of death by 50%. The "relative risk" of death in the placebo group was 2, or twice the amount in the drug group. The benefit sounds great, but in reality the benefit was very small. Choose the correct statements: (Select ALL that apply.)

A) The relative risk can be used to make a small benefit appear larger than is warranted.

B) Relative risk is also called the risk ratio.

C) An important feature of relative risk is that it tells you little about the actual risk outside the trial.

E) A relative risk < 1 means that fewer events are occurring in the treatment group compared to the control group.

Relative risk (RR) gives you a measure of the risk of an event in one group compared to the risk of that event in a comparison group. It does not give you an idea of how important (or large) the treatment effect really is in the population-at-large.

A pharmacist is conducting an analysis to determine the benefits of a diabetes treatment regimen in terms of both cost and clinical outcomes measured as a reduction in A1C values. Choose the type of analysis the pharmacist will perform:

C) A cost effectiveness analysis

The most common tool used in pharmacoeconomic research is the cost-effectiveness analysis, which compares the clinical effects of therapy (such as mortality, morbidity) to their net costs.

What is the definition of a Type II error?

D) The null hypothesis is false, but is accepted in error.

A Type II error means that the null hypothesis is false, yet it is accepted in error. The study authors conclude that there is no difference, when there actually is a difference.

A clinical trial reports that patients taking a new drug, NO-GO, experience less incontinence episodes. The number of incontinent episodes (the primary endpoint) was 16.4% in the NO-GO group and 28.7% in the placebo group. The trial randomized 600 patients (300 patients in each arm). What is the number needed to treat to prevent 1 incontinence episode?

9

The number needed to treat is 1/ARR. ARR = 28.7% - 16.4% = 12.3%. Therefore, NNT = 1/0.123 = 8.13, rounded up to 9 since we cannot treat a fraction of a person.

In biostatistics, what is meant by the mean of a group of values?

D) The average

The mean is found by adding up the values in a list, and dividing by the number of values.

In a small, randomized, double-blind trial of a new treatment in patients with acute myocardial infarction, the mortality in the treated group was half that in the control group, but the difference was not significant. We can conclude that:

E) The clinical trial may not have adequate statistical power.

The trial is small and some benefit was seen with the new treatment. Many times smaller studies are done to evaluate initial effects (many times referred to as pilot studies). Performing a larger trial will increase statistical power and may then show a statistically significant benefit.