Study Design

Study Design US Medical PG Question 1: A study is funded by the tobacco industry to examine the association between smoking and lung cancer. They design a study with a prospective cohort of 1,000 smokers between the ages of 20-30. The length of the study is five years. After the study period ends, they conclude that there is no relationship between smoking and lung cancer. Which of the following study features is the most likely reason for the failure of the study to note an association between tobacco use and cancer?

• A. Late-look bias
• B. Latency period (Correct Answer)
• C. Confounding
• D. Effect modification
• E. Pygmalion effect

Study Design Explanation: ***Latency period*** - **Lung cancer** typically has a **long latency period**, often **20-30+ years**, between initial exposure to tobacco carcinogens and the development of clinically detectable disease. - A **five-year study duration** in young smokers (ages 20-30) is **far too short** to observe the development of lung cancer, which explains the false negative finding. - This represents a **fundamental flaw in study design** rather than a bias—the biological timeline of disease development was not adequately considered. *Late-look bias* - **Late-look bias** occurs when a study enrolls participants who have already survived the early high-risk period of a disease, leading to **underestimation of true mortality or incidence**. - Also called **survival bias**, it involves studying a population that has already been "selected" by survival. - This is not applicable here, as the study simply ended before sufficient time elapsed for disease to develop. *Confounding* - **Confounding** occurs when a third variable is associated with both the exposure and outcome, distorting the apparent relationship between them. - While confounding can affect study results, it would not completely eliminate the detection of a strong, well-established association like smoking and lung cancer in a properly conducted prospective cohort study. - The issue here is temporal (insufficient follow-up time), not the presence of an unmeasured confounder. *Effect modification* - **Effect modification** (also called interaction) occurs when the magnitude of an association between exposure and outcome differs across levels of a third variable. - This represents a **true biological phenomenon**, not a study design flaw or bias. - It would not explain the complete failure to detect any association. *Pygmalion effect* - The **Pygmalion effect** (observer-expectancy effect) refers to a psychological phenomenon where higher expectations lead to improved performance in the observed subjects. - This concept is relevant to **behavioral and educational research**, not to objective epidemiological studies of disease incidence. - It has no relevance to the biological relationship between carcinogen exposure and cancer development.

Study Design US Medical PG Question 2: A research team develops a new monoclonal antibody checkpoint inhibitor for advanced melanoma that has shown promise in animal studies as well as high efficacy and low toxicity in early phase human clinical trials. The research team would now like to compare this drug to existing standard of care immunotherapy for advanced melanoma. The research team decides to conduct a non-randomized study where the novel drug will be offered to patients who are deemed to be at risk for toxicity with the current standard of care immunotherapy, while patients without such risk factors will receive the standard treatment. Which of the following best describes the level of evidence that this study can offer?

• A. Level 1
• B. Level 3 (Correct Answer)
• C. Level 5
• D. Level 4
• E. Level 2

Study Design Explanation: ***Level 3*** - A **non-randomized controlled trial** like the one described, where patient assignment to treatment groups is based on specific characteristics (risk of toxicity), falls into Level 3 evidence. - This level typically includes **non-randomized controlled trials** and **well-designed cohort studies** with comparison groups, which are prone to selection bias and confounding. - The study compares two treatments but lacks randomization, making it Level 3 evidence. *Level 1* - Level 1 evidence is the **highest level of evidence**, derived from **systematic reviews and meta-analyses** of multiple well-designed randomized controlled trials or large, high-quality randomized controlled trials. - The described study is explicitly stated as non-randomized, ruling out Level 1. *Level 2* - Level 2 evidence involves at least one **well-designed randomized controlled trial** (RCT) or **systematic reviews** of randomized trials. - The current study is *non-randomized*, which means it cannot be classified as Level 2 evidence, as randomization is a key criterion for this level. *Level 4* - Level 4 evidence includes **case series**, **case-control studies**, and **poorly designed cohort or case-control studies**. - While the study is non-randomized, it is a controlled comparative trial rather than a case series or retrospective case-control study, placing it at Level 3. *Level 5* - Level 5 evidence is the **lowest level of evidence**, typically consisting of **expert opinion** without explicit critical appraisal, or based on physiology, bench research, or animal studies. - While the drug was initially tested in animal studies, the current human comparative study offers a higher level of evidence than expert opinion or preclinical data.

Study Design US Medical PG Question 3: A researcher is conducting a study to compare fracture risk in male patients above the age of 65 who received annual DEXA screening to peers who did not receive screening. He conducts a randomized controlled trial in 900 patients, with half of participants assigned to each experimental group. The researcher ultimately finds similar rates of fractures in the two groups. He then notices that he had forgotten to include 400 patients in his analysis. Including the additional participants in his analysis would most likely affect the study's results in which of the following ways?

• A. Wider confidence intervals of results
• B. Increased probability of committing a type II error
• C. Decreased significance level of results
• D. Increased external validity of results
• E. Increased probability of rejecting the null hypothesis when it is truly false (Correct Answer)

Study Design Explanation: ***Increased probability of rejecting the null hypothesis when it is truly false*** - Including more participants increases the **statistical power** of the study, making it more likely to detect a true effect if one exists. - A higher sample size provides a more precise estimate of the population parameters, leading to a greater ability to **reject a false null hypothesis**. *Wider confidence intervals of results* - A larger sample size generally leads to **narrower confidence intervals**, as it reduces the standard error of the estimate. - Narrower confidence intervals indicate **greater precision** in the estimation of the true population parameter. *Increased probability of committing a type II error* - A **Type II error** (false negative) occurs when a study fails to reject a false null hypothesis. - Increasing the sample size typically **reduces the probability of a Type II error** because it increases statistical power. *Decreased significance level of results* - The **significance level (alpha)** is a pre-determined threshold set by the researcher before the study begins, typically 0.05. - It is independent of sample size and represents the **acceptable probability of committing a Type I error** (false positive). *Increased external validity of results* - **External validity** refers to the generalizability of findings to other populations, settings, or times. - While a larger sample size can enhance the representativeness of the study population, external validity is primarily determined by the **sampling method** and the study's design context, not just sample size alone.

Study Design US Medical PG Question 4: You are reading through a recent article that reports significant decreases in all-cause mortality for patients with malignant melanoma following treatment with a novel biological infusion. Which of the following choices refers to the probability that a study will find a statistically significant difference when one truly does exist?

• A. Type II error
• B. Type I error
• C. Confidence interval
• D. p-value
• E. Power (Correct Answer)

Study Design Explanation: ***Power*** - **Power** is the probability that a study will correctly reject the null hypothesis when it is, in fact, false (i.e., will find a statistically significant difference when one truly exists). - A study with high power minimizes the risk of a **Type II error** (failing to detect a real effect). *Type II error* - A **Type II error** (or **beta error**) occurs when a study fails to reject a false null hypothesis, meaning it concludes there is no significant difference when one actually exists. - This is the **opposite** of what the question describes, which asks for the probability of *finding* a difference. *Type I error* - A **Type I error** (or **alpha error**) occurs when a study incorrectly rejects a true null hypothesis, concluding there is a significant difference when one does not actually exist. - This relates to the **p-value** and the level of statistical significance (e.g., p < 0.05). *Confidence interval* - A **confidence interval** provides a range of values within which the true population parameter is likely to lie with a certain degree of confidence (e.g., 95%). - It does not directly represent the probability of finding a statistically significant difference when one truly exists. *p-value* - The **p-value** is the probability of observing data as extreme as, or more extreme than, that obtained in the study, assuming the null hypothesis is true. - It is used to determine statistical significance, but it is not the probability of detecting a true effect.

Study Design US Medical PG Question 5: Which of the following study designs would be most appropriate to investigate the association between electronic cigarette use and the subsequent development of lung cancer?

• A. Subjects with lung cancer who smoke and subjects with lung cancer who did not smoke
• B. Subjects who smoke electronic cigarettes and subjects who smoke normal cigarettes
• C. Subjects with lung cancer who smoke and subjects without lung cancer who smoke
• D. Subjects with lung cancer and subjects without lung cancer
• E. Subjects who smoke electronic cigarettes and subjects who do not smoke (Correct Answer)

Study Design Explanation: ***Subjects who smoke electronic cigarettes and subjects who do not smoke*** - This design represents a **cohort study**, which is ideal for investigating the **incidence** of a disease (lung cancer) in groups exposed and unexposed to a risk factor (electronic cigarette use). - By following these two groups over time, researchers can directly compare the **risk of developing lung cancer** in e-cigarette users versus non-smokers. *Subjects with lung cancer who smoke and subjects with lung cancer who did not smoke* - This option incorrectly compares two groups both with lung cancer, where the exposure to smoking can either be **electronic or traditional cigarettes,** but does not provide a control group without lung cancer to assess the association. - This design would not allow for the calculation of an **incidence rate** or a **relative risk** of lung cancer development specific to electronic cigarette use. *Subjects who smoke electronic cigarettes and subjects who smoke normal cigarettes* - This design compares two different types of smoking, which might be useful for comparing their relative risks but doesn't include a **non-smoking control group** to establish the absolute association with electronic cigarettes. - While it could show if e-cigarettes are "safer" than traditional cigarettes, it wouldn't directly answer whether e-cigarettes themselves **cause lung cancer**. *Subjects with lung cancer who smoke and subjects without lung cancer who smoke* - This describes a **case-control study** but focuses on smoking in general rather than specifically electronic cigarettes, which is the independent variable of interest. - While valuable for identifying risk factors, it would need to specifically differentiate between **electronic cigarette smokers** and other smokers to answer the question adequately. *Subjects with lung cancer and subjects without lung cancer* - This general description of a **case-control study** is too broad; it does not specify the exposure of interest, which is electronic cigarette use. - To be relevant, the study would need to gather data on **electronic cigarette use** in both the lung cancer group and the non-lung cancer control group.

Study Design US Medical PG Question 6: The APPLE study investigators are currently preparing for a 30-year follow-up evaluation. They are curious about the number of participants who will partake in follow-up interviews. The investigators noted that of the 83 participants who participated in the APPLE study's 20-year follow-up, 62 were in the treatment group and 21 were in the control group. Given the unequal distribution of participants between groups at follow-up, this finding raises concerns for which of the following?

• A. Volunteer bias
• B. Reporting bias
• C. Inadequate sample size
• D. Attrition bias (Correct Answer)
• E. Lead-time bias

Study Design Explanation: ***Attrition bias*** - **Attrition bias** occurs when participants drop out of a study, especially if the dropout rate differs between the intervention and control groups, which can lead to a **skewed comparison** of outcomes. - The unequal distribution of participants (62 vs. 21) between the treatment and control groups at the 20-year follow-up suggests that a disproportionate number of participants may have dropped out of one group, thus leading to attrition bias. *Volunteer bias* - **Volunteer bias** occurs when individuals who volunteer for a study differ significantly from the general population or those who decline to participate, potentially affecting the study's **generalizability**. - This scenario describes differences in retention *after* initial participation, not differences in initial willingness to join. *Reporting bias* - **Reporting bias** refers to the selective reporting of study findings, where positive or statistically significant results are more likely to be published or emphasized than negative or non-significant ones, which can distort the overall evidence base. - This bias relates to how results are disseminated, not to differential dropout rates or participant retention in a study. *Inadequate sample size* - **Inadequate sample size** means that the number of participants in a study is too small to detect a statistically significant effect if one truly exists, leading to a lack of **statistical power**. - While the overall number of participants at follow-up might be small, the primary concern here is the *unequal distribution* between groups, indicating a problem with participant retention rather than just a low total count. *Lead-time bias* - **Lead-time bias** occurs when early detection of a disease (e.g., through screening) makes survival appear longer than it actually is, without necessarily prolonging the patient's life, by advancing the **point of diagnosis**. - This bias is relevant to screening programs and disease detection, not to the differential dropout rates observed in a longitudinal study.

Study Design US Medical PG Question 7: You are interested in studying the etiology of heart failure reduced ejection fraction (HFrEF) and attempt to construct an appropriate design study. Specifically, you wish to look for potential causality between dietary glucose consumption and HFrEF. Which of the following study designs would allow you to assess for and determine this causality?

• A. Cross-sectional study
• B. Case series
• C. Cohort study (Correct Answer)
• D. Case-control study
• E. Randomized controlled trial

Study Design Explanation: ***Cohort study*** - A **cohort study** observes a group of individuals over time to identify risk factors and outcomes, allowing for the assessment of **temporal relationships** between exposure (dietary glucose) and outcome (HFrEF). - This design is suitable for establishing a potential **causal link** as it tracks participants from exposure to outcome, enabling the calculation of incidence rates and relative risks. *Cross-sectional study* - A **cross-sectional study** measures exposure and outcome simultaneously at a single point in time, making it impossible to determine the **temporal sequence** of events. - This design can only identify **associations** or correlations, not causation, as it cannot establish whether high glucose consumption preceded HFrEF. *Case series* - A **case series** describes characteristics of a group of patients with a particular disease or exposure, often to highlight unusual clinical features, but it lacks a **comparison group**. - It cannot assess causality because it does not provide information on the frequency of exposure in healthy individuals or the incidence of the disease in unexposed individuals. *Case-control study* - A **case-control study** compares individuals with the outcome (cases) to those without the outcome (controls) to determine past exposures, which makes it prone to **recall bias**. - While it can suggest associations, it cannot definitively establish a temporal relationship or causation as the outcome is already known when exposure is assessed. *Randomized controlled trial* - A **randomized controlled trial (RCT)** is the gold standard for establishing causation by randomly assigning participants to an intervention or control group, but it may not be ethical or feasible for studying long-term dietary exposures and chronic diseases like HFrEF due to the long follow-up period and complexity of diet. - While ideal for causality, directly controlling and randomizing dietary glucose intake over decades to observe HFrEF development might be practically challenging or unethical.

Study Design US Medical PG Question 8: A population is studied for risk factors associated with testicular cancer. Alcohol exposure, smoking, dietary factors, social support, and environmental exposure are all assessed. The researchers are interested in the incidence and prevalence of the disease in addition to other outcomes. Which pair of studies would best assess the 1. incidence and 2. prevalence?

• A. 1. Prospective cohort study 2. Cross sectional study (Correct Answer)
• B. 1. Prospective cohort study 2. Retrospective cohort study
• C. 1. Cross sectional study 2. Retrospective cohort study
• D. 1. Case-control study 2. Prospective cohort study
• E. 1. Clinical trial 2. Cross sectional study

Study Design Explanation: ***1. Prospective cohort study 2. Cross sectional study*** - A **prospective cohort study** is ideal for measuring **incidence** (new cases over time) because it follows a group of individuals forward in time to observe who develops the disease. - A **cross-sectional study** is suitable for measuring **prevalence** (existing cases at a specific point in time) as it surveys a population at one moment to determine the proportion with the disease. *1. Prospective cohort study 2. Retrospective cohort study* - A **retrospective cohort study** assesses past exposures and outcomes and can measure incidence, but it is not the primary choice for prevalence. - While a prospective cohort study is appropriate for incidence, a retrospective cohort study is less suited for determining current prevalence. *1. Cross sectional study 2. Retrospective cohort study* - A **cross-sectional study** measures prevalence, not incidence, as it captures disease status at a single point in time. - A **retrospective cohort study** looks back in time to identify past exposures and subsequent outcomes, which is not the best method for current prevalence. *1. Case-control study 2. Prospective cohort study* - A **case-control study** compares exposures between individuals with a disease (cases) and those without (controls) and is best for studying rare diseases and estimating odds ratios, not incidence or prevalence directly. - A **prospective cohort study** is suitable for incidence, but a case-control study is not for incidence or prevalence. *1. Clinical trial 2. Cross sectional study* - A **clinical trial** is an experimental study designed to test the efficacy of interventions and is not primarily used to measure disease incidence or prevalence in a general population. - While a cross-sectional study is appropriate for prevalence, a clinical trial is not designed for incidence measurement.

Study Design US Medical PG Question 9: Many large clinics have noticed that the prevalence of primary biliary cholangitis (PBC) has increased significantly over the past 20 years. An epidemiologist is working to identify possible reasons for this. After analyzing a series of nationwide health surveillance databases, the epidemiologist finds that the incidence of PBC has remained stable over the past 20 years. Which of the following is the most plausible explanation for the increased prevalence of PBC?

• A. Improved quality of care for PBC (Correct Answer)
• B. Increased availability of diagnostic testing for PBC
• C. Increased exposure to environmental risk factors for PBC
• D. Increased awareness of PBC among clinicians
• E. Increased average age of the population at risk for PBC

Study Design Explanation: ***Improved quality of care for PBC*** - This leads to a **longer survival time** for patients with PBC. When incidence remains stable but patients live longer, the cumulative number of living cases (prevalence) naturally increases. - An increase in prevalence with stable incidence is a classic indicator of **improved patient survival** due to better management or treatment. *Increased availability of diagnostic testing for PBC* - This would primarily impact the **incidence** of PBC by detecting more cases that were previously undiagnosed. The question states that the incidence has remained stable. - While improved diagnostics might initially increase *reported* incidence, if the true incidence is stable, it wouldn't explain a sustained rise in prevalence without a corresponding change in incidence or survival. *Increased exposure to environmental risk factors for PBC* - This would directly lead to an **increase in the incidence** of PBC, as more people would be developing the disease. - Since the incidence is stable, an increase in environmental risk factors is not the most plausible explanation for increased prevalence. *Increased awareness of PBC among clinicians* - Similar to increased diagnostic testing, increased awareness would likely lead to the diagnosis of more new cases, thus **increasing the incidence** of PBC. - A stable incidence despite increased awareness means that the actual rate of new cases developing the disease has not changed, ruling this out as the primary cause of increased prevalence. *Increased average age of the population at risk for PBC* - An aging population could potentially increase the incidence of age-related diseases. However, if the **incidence has remained stable**, it implies that even with an older population, the rate of new diagnoses has not increased. - While age is a risk factor for PBC, an increase in prevalence without a change in incidence suggests a factor influencing the duration of the disease rather than its onset.

Study Design US Medical PG Question 10: A patient is in the ICU for diabetic ketoacidosis and is currently on an insulin drip. His electrolytes are being checked every hour and his potassium is notable for the following measures: 1. 5.1 mEq/L 2. 5.8 mEq/L 3. 6.1 mEq/L 4. 6.2 mEq/L 5. 5.9 mEq/L 6. 5.1 mEq/L 7. 4.0 mEq/L 8. 3.1 mEq/L Which of the following is the median potassium value of this data set?

• A. 6.05
• B. 5.10
• C. 5.16
• D. 5.45 (Correct Answer)
• E. 3.10

Study Design Explanation: ***5.45*** - To find the **median**, first arrange the potassium values in ascending order: 3.1, 4.0, 5.1, 5.1, 5.8, 5.9, 6.1, 6.2. - Since there are **eight** (an even number) values, the median is the average of the two middle values (the 4th and 5th values): (5.1 + 5.8) / 2 = 10.9 / 2 = **5.45**. *6.05* - This value might be obtained by incorrectly averaging a different pair of numbers or miscalculating the average of the sorted data set. - It is not the correct median for this particular data set of potassium values. *5.10* - While 5.1 is present twice in the data set, and is one of the middle values, it is not the **median** because the **median** for an even number of values is the average of the two middle numbers, not just one of them. - This would be the median if the values were 3.1, 4.0, 5.1, 5.1, 5.1, 5.8, 5.9, 6.1. *5.16* - This value does not correspond to any of the numbers in the data set nor does it result from the correct calculation of the **median**. - It might represent an incorrect average or a miscalculation of a percentile. *3.10* - This value is the **minimum** potassium level recorded, not the median. - The median represents the middle value in a sorted data set, while the minimum is the lowest value.

Study Design Flashcard 1 of 10

Question: _____ studies are observational studies that compare a group of people with disease to a group without disease

Answer: Case-control

Study Design Flashcard 2 of 10

Question: What type of study uses subjects as their own controls?_____

Answer: Crossover studies

Study Design Flashcard 3 of 10

Question: Confounding bias occurs when the exposure-disease relationship is muddled by the effect of an extraneous factor that is associated with both exposure and disease and _____ on the causal pathway

Answer: is NOT

Study Design Flashcard 4 of 10

Question: A method of statistical analysis that pools summary data (ex. means, RRs) from multiple studies for a more precise estimate of the size of an effect is known as _____

Answer: meta-analysis

Study Design Flashcard 5 of 10

Question: Which type of observational study is best for studying rare diseases? _____

Answer: Case-control studies

Study Design Flashcard 6 of 10

Question: What is the ANOVA test used for? (biostatistics)

Answer: Compare means of three or more groups.

Study Design Flashcard 7 of 10

Question: prevalence (epidemiology)

Answer: # existing cases/population at risk ...at a discrete point in time

Study Design Flashcard 8 of 10

Question: meta-analysis (epidemiology)

Answer: Study that pools data from other studies and analyzes them in aggregate to achieve greater statistical power

Study Design Flashcard 9 of 10

Question: incidence (epidemiology)

Answer: # new cases/population at risk ...over a specified time period (i.e. a compound rate)

Study Design Flashcard 10 of 10

Question: What kind of study compares one group of people with disease to another group of people without disease and looks for an exposure or risk factor that explains the difference?

Answer: Case-control study