Final Exam - Weeks 3 & 4

Question 1

What is bias?

Answer 1

Bias refers to any factor that systematically affects the outcomes of a study in a way that deviates from the true or expected results.

Question 2

Is bias related to study design?

Answer 2

Yes, bias can arise from various sources, such as the design of the study, the selection of participants, the measurement tools used, or the interpretation of the data.

Question 3

There is more bias in observational studies, true or false?

Answer 3

True, observational studies are more prone to bias than randomized controlled trials because they rely on the observation of existing situations and the collection of data from them, rather than the manipulation of variables in a controlled environment.

Question 4

Bias is independent of both sample size and statistical significance, true or false?

Answer 4

True, bias is related to study design.

Question 5

Explain a “spurious” finding.

Answer 5

Spurious is a term used to describe a statistical relationship between two variables that would, at first glance, appear to be causally related, but upon closer examination, only appear so by coincidence or due to the role of a third, intermediary variable.

Question 6

Explain a type I error.

Answer 6

Finding an association or significant finding when none exists.

Question 7

Explain type II error.

Answer 7

Finding no association when one truly does exist.

Question 8

Explain selection bias.

Answer 8

When the selection of study participants is not random or representative of the population being studied, leading to a distorted or inaccurate sample.

Question 9

Explain information bias.

Answer 9

When there are errors or inaccuracies in the measurement or collection of data, leading to incorrect or biased study results.

Question 10

Explain confounding bias.

Answer 10

When there is a third variable or factor that is related to both the exposure and outcome being studied, leading to a spurious or false association between the two. In other words, the observed association between the exposure and outcome may be explained by the confounding variable rather than a true causal relationship.

Question 11

Explain the general population.

Answer 11

The general population is the entire population of individuals with a characteristic of interest, such as a particular disease or condition of clinical concern.

Question 12

Explain the target population.

Answer 12

The target population of a study is the broad group of people that researchers are examining. In clinical trials the target population is often the group of people researchers believe might benefit from a particular experimental drug or treatment.

Question 13

Explain the difference between the general population and the target population.

Answer 13

The general population refers to the entire group of individuals or entities that share a common characteristic or live in a specific geographic area, while the target population is a specific subset of the general population that is the focus of a research study.

Question 14

If a study is investigating the prevalence of hypertension in the United States, the _____________________ would be all individuals who live in the country, while the _________________ would be those individuals who are at risk of developing hypertension or who have been diagnosed with the condition.

Answer 14

general population, target population.

Question 15

What is the available population?

Answer 15

The available population refers to the group of individuals or entities that are accessible and available for recruitment into a research study. This population may be a subset of the general population or the target population, depending on the study’s inclusion and exclusion criteria and the feasibility of accessing the desired population.

Question 16

The available population is always the same size as the target population, true or false?

Answer 16

False, the available population and the target population may be the same size, but this is not always true. The available population may be smaller than the target population, as not all individuals who meet the study’s criteria may be accessible or willing to participate in the study.

Question 17

Explain the sample population.

Answer 17

The sample population is a subset of the target population that is selected to participate in a research study. The sample is a smaller representation of the target population, and the selection of individuals for the sample is typically done using some form of sampling technique.

Question 18

If the sample population does not represent the target population, what type of bias is this?

Answer 18

Selection bias.

Question 19

To avoid selection bias, the sample population should be selected from the same source within the available population, true or false?

Answer 19

False! In order to avoid selection bias, the sample population should be selected from multiple sources within the available population.

Question 20

Volunteer bias, referral bias and non-response bias are all examples of ___________ bias.

Answer 20

Selection

Question 21

Recall bias, observer bias, and respondent bias are all examples of ___________ bias.

Answer 21

Information

Question 22

Use of a non-standardized or incorrect measure, and administering a measure incorrectly are examples of what type of bias?

Answer 22

Information bias.

Question 23

A ____________________ is an extraneous variable whose presence affects the variable being studied so that the results do not reflect the actual relationship between the variables under study.

Answer 23

confounder

Question 24

What is a confounder?

Answer 24

A confounder is a variable that is associated with both the exposure and the outcome in a research study, and that may distort or confound the relationship between the two.

Question 25

In a study investigating the relationship between smoking and lung cancer, age could be a ___________________. This is because age is associated with both smoking behavior and the risk of developing lung cancer. If age is not taken into account when analyzing the data, it may appear that smoking is a stronger risk factor for lung cancer than it actually is.

Answer 25

Confounding variable

Question 26

A _______________________ is associated with exposure (independent of outcome) and outcome (independent of exposure), is not in causal pathway.

Answer 26

confounding variable

Question 27

If you believe a variable is a confounder, would you expect it to fit in the causal pathway?

Answer 27

No, confounders (confounding variables) are not in the causal pathway, they affect both the exposure and the outcome separately.

Question 28

Related to both the exposure and the outcome BUT NOT in the causal pathway = __________________ variable.

Related to both the exposure and the outcome BUT IN the causal pathway = ____________________ variable.

Answer 28

Confounding

Intermediate

Question 29

How can confounding variables be controlled for in research?

Answer 29

Stratification: Stratification involves dividing study participants into subgroups based on a confounding variable and then analyzing the data separately for each subgroup. This can help to control for the confounding variable and uncover any differences in the association between the exposure and the outcome.

Matching: Matching is a method of controlling for confounding by selecting study participants who are similar with respect to potential confounders. This can be done by matching on a single variable or by using multiple variables.

Study design: Researchers can control for confounding at the design stage of the study. For example, they can use randomization to ensure that the study groups are similar with respect to potential confounders.

Question 30

You want to assess a program for improving the eating habits of shift workers. You put up flyers where many work night shifts and invite them to participate. What type of bias does this represent?

Answer 30

Selection bias (sample population not representative of available/target population).

Question 31

You are conducting a multi-center study. One of your colleagues in a clinic is not comfortable using the outcome measure provided in the study protocol. They decided to use an outcome measure they have used before instead. What type of bias does this represent?

Answer 31

Information bias (the wrong information was collected/ information was collected using the wrong outcome measure).

Question 32

You test 200 volunteers (100 men and 100 women), you find that lack of exercise leads to weight gain. What type of bias is represented in this case?

Answer 32

Confounding bias (sex was controlled for however other possible confounders were not such as age).

Question 33

What are the four main categories of quantitative designs?

Answer 33

Non-experimental
Pre-experimental
Quasi-experimental
True experimental

Question 34

Designs that do not randomize the sample, or mechanically manipulate the independent variable but may have a control group.

Answer 34

Non-experimental designs.

Question 35

Survey designs, and passive observation are examples of what type of quantitative study design.

Answer 35

Non-experimental.

Question 36

Explain cross-sectional study design.

Answer 36

A cross-sectional study design is a type of observational research study that collects data from a population or a representative subset of the population at a single point in time. The study design is called cross-sectional because it examines the relationship between the outcome of interest and the exposure or risk factors of interest at a specific point in time.

Question 37

Cross-sectional study design is observational, true or false?

Answer 37

True.

Question 38

Cross-sectional study design is an example of what type of quantitative study design?

Answer 38

Non-experimental (observational). The investigator does not alter the exposure status. The investigator measures the outcome and exposures in the population and may study their association.

Question 39

___________ is the proportion of persons with a particular condition or attribute in a given population at a point in time.

Answer 39

Prevalence.

Question 40

Cross-sectional study designs may be used for population-based surveys, true or false?

Answer 40

True.

Question 41

Cross-sectional studies may be used for estimating the prevalence in clinic-based studies, true or false?

Answer 41

True.

Question 42

Cross-sectional studies are often used in epidemiological research to estimate…

Answer 42

the prevalence of a disease or condition in a population, or to examine the distribution of risk factors among different populations.

Question 43

One of the main advantages of a cross-sectional study design is _______________________________________________________________________________. However, cross-sectional studies have some limitations, such as the inability to establish causality or temporal relationships between the outcome and exposure variables.

Answer 43

that it is relatively quick and inexpensive to conduct.

Question 44

Cross-sectional study design can establish cause and effect, true or false?

Answer 44

False, cross-sectional study design can not establish cause and effect.

Question 45

Cross-sectional study design is unable to establish temporal relationships between the outcome and exposure variables, explain.

Answer 45

Establishing a temporal relationship is important in determining causality between an exposure and an outcome. In order for causality to be established, it is necessary for the exposure to precede the outcome in time. Cross-sectional study design is unable to establish temporal relationships because it assesses the exposure and outcome at the same time, without the ability to determine which event occurred first.

Question 46

Cross-sectional research design is prone to ________________ bias.

Answer 46

Selection bias. Cross-sectional study designs are prone to selection bias because they rely on a sample of participants who are recruited at a single point in time. The sample may not be representative of the target population, leading to biased results.

Question 47

What is case-control study design?

Answer 47

A case-control study design is a type of observational study that compares individuals with a specific disease or condition (cases) to individuals without the disease or condition (controls). The study design is retrospective, meaning that it looks back in time to identify differences in exposure to risk factors or potential causes of the disease or condition between the cases and controls.

Question 48

Case-control study design is observational, true or false?

Answer 48

True. A case-control study design is a non-experimental study design, meaning that it does not involve intervention or manipulation of the exposure or risk factors.

Question 49

Case-control study design is always prospective, true or false?

Answer 49

False! Case control study design is always retrospective (starts with an outcome then traces back to investigate exposure)

Question 50

In case-control study design _____________ precedes ____________.

Answer 50

In case-control study design the exposure precedes the outcome.

Question 51

Explain “case” and “control” in case-control study design?

Answer 51

Case refers to individuals with the outcome, condition, or event.

Control refers to individuals without the outcome, condition, or event.

Question 52

Cases and controls need to be similar on all variables except the ____________ and the ________________.

Answer 52

Exposure and the outcome.

Question 53

Cases and controls must arise from the same __________________.

Answer 53

general population.

Question 54

Controls should be selected as a representative sample of the ___________________________________.

Answer 54

population which produced the cases.

Question 55

Name some common sources of bias in case-control studies.

Answer 55

Recall bias: cases and controls may recall past exposure differently

Interviewer/observer bias: the recording of exposure information may vary depending on the investigator’s knowledge of an individual’s disease status.

Selection bias: controls are unrepresentative of the population that produced the cases.

Question 56

Name some weaknesses of case-control studies.

Answer 56

• Ascertainment of past exposures can be difficult.
• Subject to biases.
• Selection of controls can be challenging.
• Difficult to reference records and data if record keeping is either inadequate or unreliable.

Question 57

What type of study design is:
- Not expensive
- Efficient for studying rare diseases
- Efficient for studying diseases with a long latency period
- Able to study multiple exposures
- Good with study conditions where would be unethical to expose the participants (e.g., smoking)

Answer 57

Case-control study design.

Question 58

The odds ratio (OR) is used in case-control studies to estimate…

Answer 58

the strength associated between exposure and outcome.

Question 59

Explain odds ratio.

Answer 59

The odds ratio is a statistical measure used to quantify the strength of association between two binary variables, or outcomes. It is calculated as the ratio of the odds of the outcome occurring in one group compared to the odds of the outcome occurring in another group.

To calculate the odds ratio, first, we need to construct a 2x2 contingency table that shows the number of individuals in each group who either have or do not have the outcome of interest.

Question 60

Explain an odds ratio of >1, <1, and =1

Answer 60

If the odds ratio is greater than 1, it indicates that the outcome is more likely to occur in the first group.

If the odds ratio is less than 1, it indicates that the outcome is more likely to occur in the second group.

An odds ratio of 1 indicates that there is no difference in the odds of the outcome occurring between the two groups.

Question 61

If OR >1 and 95% CI does not include 1, then the exposure…

Answer 61

significantly increases odds of outcome/disease.

Question 62

If OR<1 and 95% CI does not include 1, then the exposure….

Answer 62

significantly decreases odds of outcome/disease.

Question 63

If OR=1 and 95% CI does not include 1, then exposure…

Answer 63

does not affect odds of disease/outcome.

Question 64

If the 95% confidence includes 1, what does this mean?

Answer 64

There is no statistical significance.

Question 65

Association between gender and fractures due to falls among patients in a rehabilitation hospital.

OR=1.71
CI=0.99, 2.97

Interpret these results.

Answer 65

The OR>1 meaning that the exposure increases the odds of the outcome. In this case this could mean that males had a 1.71 times higher odds or 71% higher odds of fracture due to falls compared to female participants.

The CI does not include 1 meaning that these results are not statistically significant.

Question 66

Association between smoking and lung cancer.

OR=8.25
CI=2.52-71.37

Interpret these results.

Answer 66

The OR>1 meaning that the exposure increases the odds of the outcome. In this case the exposure (smoking) increases the odds of the outcome (lung cancer).

The CI does not include 1, meaning that these results are statistically significant.

Question 67

Association between exercise and stroke.

OR=0.40
CI=0.20-0.91

Interpret these results.

Answer 67

The OR<1, meaning that the exposure decreases the odds of the outcome. In this case, the exposure (exercise) decreases the outcome (stroke). The findings indicate that people who exercise have 0.40 times the odds of having a stroke or have 60% lower odds of having a stroke than those who don’t exercise.

The CI does not include 1, meaning that these results are statistically significant.

Question 68

What is a cohort study design?

Answer 68

A cohort study is a type of observational study where a group of individuals, or a “cohort,” is identified based on a shared characteristic, such as age, gender, occupation, or exposure to a particular risk factor. The cohort is then followed over time to observe the occurrence of health outcomes of interest, such as the development of a disease or the occurrence of a particular event.

Question 69

Cohort studies can be retrospective or prospective, true or false?

Answer 69

True.

Question 70

What is the best study design for determining incidence and natural history of a condition?

Answer 70

Cohort study.

Question 71

In a ________________ cohort study, the cohort is identified at the beginning of the study and followed over time into the future, with data collected at regular intervals.

Answer 71

prospective

Question 72

In a ___________________cohort study, the cohort is identified based on exposure status or other characteristics that have already occurred in the past, and data are collected retrospectively.

Answer 72

retrospective

Question 73

What is relative risk/risk ratio?

Answer 73

The ratio of the probability of developing the condition if exposed to a certain variable compared with the probability if not exposed.

Question 74

Relative risk = ____________/_______________

Answer 74

Relative risk= risk of outcome if exposed/ risk of outcome if unexposed

Question 75

Association between smoking and myocardial infarction.

RR=2.45
CI=1.15-5.47

Interpret these results.

Answer 75

RR>1, meaning the exposure (smoking) increases the outcome (myocardial infarction). People who smoke are at 2.5 times greater risk than non-smokers of having an MI.

This finding is statistically significant because the CI does not include 1.

Question 76

Association between regular exercise and stroke.

RR=0.85
CI=0.61-1.06

Interpret these results.

Answer 76

RR<1, meaning the exposure (regular exercise) decreases odds of the outcome (stroke). People who exercise regularly are at 0.85 times less risk of stroke (15% less likely).

This finding is not statistically significant because the CI includes 1.

Question 77

Cohort study design can establish temporal sequence, true or false?

Answer 77

True, in a cohort study, researchers identify a group of people who are free of the outcome of interest at the beginning of the study and then follow them over time to see whether they develop the outcome. By doing so, the temporal sequence of exposure and outcome can be established, as the exposure occurs before the outcome.

Question 78

A study is conducted to investigate the relationship between smoking and lung cancer. Researcher identified a group of people who are smokers (the exposed group) and a group of people who are non-smokers (the unexposed group). Then, they followed both groups over time and observe the incidence of lung cancer in each group. If the incidence of lung cancer is higher in the exposed group than in the unexposed group, it suggests that smoking is a risk factor for lung cancer. What type of study design is this?

Answer 78

Cohort study design.

Question 79

In a _____________________ study the investigator measures the outcome and the exposure in the study participants at the same time.

Answer 79

Cross-sectional study

Question 80

Participants are selected for the study based on their outcome status. The investigator then assesses the exposure in both these groups.

Answer 80

Case-control study

Question 81

Participants are selected based on the exposure status of the individual. They are then followed over time to evaluate for the occurrence of the outcome of interest.

Answer 81

Cohort study

Question 82

Are case-control studies or cohort studies typically more expensive?

Answer 82

Cohort studies are typically more expensive.

Question 83

If the objective of the study is to determine prevalence, what study design would be the most appropriate?

Answer 83

Cross sectional

Question 84

If the objective of the study is to determine incidence, what study design would be the most appropriate?

Answer 84

Cohort

Question 85

If the objective of the study is to determine prognosis, what study design would be the most appropriate?

Answer 85

Cohort

Question 86

Explain pre-experimental study design.

Answer 86

Pre-experimental designs are research designs that lack some of the key features necessary to establish a causal relationship between an independent variable and a dependent variable. They are called “pre-experimental” because they are often used as a preliminary step before conducting a more rigorous experimental design.

Question 87

Pre-experimental designs are characterized by the absence of ___________________________ and the lack of random assignment of participants to groups. This means that researchers are unable to rule out alternative explanations for the observed relationship between the independent and dependent variables.

Answer 87

of a comparison or control group

Question 88

Pre-experimental designs do not randomize the sample, but do include a control group, true or false?

Answer 88

False, pre-experimental designs do not randomize the sample, OR have a control group.

Question 89

In pre-experimental designs the independent variable is manipulated, true or false?

Answer 89

True, the independent variable is manipulated by the researcher to observe its effects on the dependent variable. However, pre-experimental designs lack some of the key features necessary to establish a causal relationship between the independent variable and the dependent variable. This includes the absence of a comparison or control group and the lack of random assignment of participants to groups.

Question 90

Explain one-shot design.

Answer 90

One-shot design is a type of pre-experimental design. This design involves measuring the dependent variable after the independent variable has occurred. There is no measure of outcome at baseline, but there is a measure of outcome post-intervention.

Question 91

This type of pre-experimental design measures outcome post-intervention only (no measure at baseline).

Answer 91

One-shot design.

Question 92

Explain static group comparison design.

Answer 92

Static group comparison design is a type of pre-experimental design. This design involves comparing two groups that differ in terms of the independent variable. However, the groups are not randomly assigned, which can lead to selection bias. The outcomes are compared between groups post-intervention.

Question 93

This type of pre-experimental design measures outcome post-intervention between two groups but does not include a measure at baseline for either group. On group receives the intervention and the other does not.

Answer 93

Static group comparison design.

Question 94

Explain one group pretest-posttest design.

Answer 94

One group pretest-posttest design is a type of pre-experimental study design. This design involves measuring the dependent variable before and after the independent variable has occurred. However, there is no comparison group, making it difficult to rule out alternative explanations for the observed changes.

Question 95

This type of pre-experimental design compares outcome between baseline and post-intervention time points of one group.

Answer 95

One group pretest-posttest design.

Question 96

_______________________ designs do not randomize the sample, but may have a control group or mechanically manipulate independent variable (an intervention).

Answer 96

Quasi experimental.

Question 97

Explain quasi-experimental design.

Answer 97

In quasi-experimental design, the researcher manipulates the independent variable, but groups are not formed through random assignment. Instead, groups are often pre-existing or naturally occurring, and the researcher must try to control for confounding variables through various means.

Question 98

Quasi-experimental design is a research design that shares some features with true experimental design, but lacks…

Answer 98

random assignment of participants to groups.

Question 99

Quasi-experimental designs are used when…

Answer 99

random assignment is not possible or ethical, but the researcher still wants to establish cause-and-effect relationships between variables. They are particularly useful when the independent variable is not something that can be randomly assigned to participants, such as a person’s gender, age, or medical condition.

Question 100

Non-experimental, pre-experimental, quasi-experimental and true experimental study designs are all _______________________.

Answer 100

Quantitative study designs. They involve the collection of numerical data through standardized procedures and statistical analyses to investigate relationships between variables.

Question 101

Explain non-equivalent control group design.

Answer 101

Non-equivalent control group design is a type of quasi-experimental design. This design involves a treatment group and a comparison group that are not formed through random assignment. Instead, the researcher selects pre-existing groups that are as similar as possible on key varibales that might affect the outcome.

Question 102

When using a ______________study design, a researcher might compare the test scores of students in two schools that have similar demographics, but one school uses a new teaching method while the other does not without using random assignment to from the two groups.

Answer 102

Quasi-experimental (non-equivalent control group design).

Question 103

In non-equivalent control group design, the outcome of both groups is measured at baseline, true or false?

Answer 103

True.

Question 104

Explain some advantages of quasi-experimental study design.

Answer 104

• Increased external validity: often used in real-world settings and are more likely to reflect the complexity of real-life situation than laboratory experiments.
• Cost-effective: often less expensive then experimental designs, as they do not require the same level of control over extraneous variables or recruitment of large samples of participants.
• More ethical: often more ethical than experimental designs, as they do not involve random assignment of participants to groups. This can be especially important when the intervention being studied has the potential to harm participants, or when random assignment is not feasible or practical.

Question 105

Explain some disadvantages of quasi-experimental design.

Answer 105

• Threats to internal validity: susceptible to threats to internal validity, such as selection bias, history effects and maturation effects. This means that it is more difficult to establish a causal relationship between the independent and dependent variables.
• Lack of randomization: do not involve random assignment of participants to groups, which can increase the risk of confounding variables affecting the results.
• Limited control over extraneous variables: making it more difficult to attribute any observed effects solely to the manipulation of the independent variable.

Question 106

Why is there a risk of confounding bias in quasi-experimental study design?

Answer 106

Confounding variables are usually balanced by the randomization process.

Question 107

This type of quantitative study design randomizes the sample, has a control group and mechanically manipulates the independent variable.

Answer 107

True experimental.

Question 108

Explain true experimental study design.

Answer 108

True experimental study design is a research method used to establish a cause-and-effect- relationship between two or more variables. In this design, the researcher manipulates an independent variable, while controlling for other variables, and observes the effect of this manipulation on a dependent variable.

Question 109

In this type of quantitative study design, the researcher manipulates an independent variable, while controlling for other variables, and observes the effect of this manipulation on a dependent variable. The design involves random assignment of participants to groups to ensure that any observed effects are due solely to the manipulation of the independent variable, rather than any pre-existing differences between the groups.

Answer 109

True experimental study design.

Question 110

Randomized controlled trials, posttest only design, Solomon four-group designs, factorial designs and counterbalance designs are all what type of study design?

Answer 110

Quantitative (true experimental).

Question 111

The ________________________ design is considered the gold standard in research because it provides the strongest evidence for causality between the independent and dependent variables. However, this design can be challenging to implement in some research settings, and may require a large sample size and significant resources to carry out.

Answer 111

True experimental.

Question 112

Which true experimental study design do these characteristics relate to?
- The intent of the research is to evaluate a deliberate intervention which is often an innovation in treatment.
- Intervention and comparison groups are deliberately manipulated by the investigator.
- Groups are assigned by a random process.
- Groups enter at the same calendar time and are followed over the same time period.
- The groups are compared.

Answer 112

Randomized controlled trial.

Question 113

Name some advantages of RCTs.

Answer 113

• Control over extraneous variables: random assignment
• Causality: can establish causality by demonstrating a statistically significant difference between treatment and control groups. By randomizing participants and using a control group, the design can demonstrate that the treatment is responsible for observed changes.
• Reliability: use standardized methods and protocols for data collection and analysis.
• Generalizability: to a larger population.
• Acceptance in medical practice: RCTs are widely accepted in medical practice and research.
• Strongest evidence for a causal relationship between an exposure and an outcome.

Question 114

Name some challenges associated with RCT study design.

Answer 114

• Often complex and time-consuming
• Expensive
• Require a large sample size, struct protocols, and well-trained staff
• RCTs often involve a placebo group which can be ethically challenging as it involves withholding treatment from some participants.
• Attrition or loss of participants during the study can affect the power of the study and reduce the generalizability of the results.
• Limited external validity: while RCTs can establish causality, they may not always be applicable to real-world settings. The strict protocols and highly controlled environment of RCTs may not always reflect real-world settings.

Question 115

RCTs are intended to answer questions about the ______________ or ______________ of an intervention.

Answer 115

efficacy or effectiveness

Question 116

Why would we use PICOT format not PECOT for RCT research questions?

Answer 116

PICOT is used over PECOT because RCTs include an intervention (I).

IN PECOT the E stands for Exposure: the intervention, exposure or risk factor being studied.

Question 117

PECOT is typically used in ________________________ studies, where the exposure or risk factor being studied may not necessarily be an intervention or treatment. PICOT, on the other hand, is typically used in ______________________ studies, where the focus is on a specific intervention or treatment.

Answer 117

Observational

Interventional

Question 118

The _____________________ refers to a phenomenon in which study participants modify their behavior in response to being observed or studied, rather than the intervention or treatment being tested. This can result in changes in behavior or outcomes that are not due to the intervention or treatment itself, but rather to the participants’ awareness of being studied.

Answer 118

Hawthorne effect.

Question 119

Strongest evidence for causality comes from a design where subjects are assigned ______________ to the interventions.

Answer 119

at random

Question 120

Explain the advantages of randomization.

Answer 120

• Probability of group assignment is known and equal for all (does not depend upon characteristic s of subjects)
• Limited selection bias: subjects are not assigned (or chose) intervention based on propensity to have a good outcome.
• Characteristics of subjects, known and unknown, are assigned along with the subject
• Limited confounding bias: on average, in larger sample sizes, randomization balances group on confounding factors (know and unknown)
• Limited information bias: as long as outcome assessors are blind to group assignment

Question 121

Randomization only guarantees balance at the outset of the trial, explain why.

Answer 121

As time goes on, subjects may drop out of trial, they may cross-over (to the intervention or away from it). If these individuals are excluded from the final analysis, the only people who remain are those who selected themselves to remain where they were assigned (selection bias). A solution to this bias is to ensure that for the analysis, everyone must stay in the groups to which they were assigned, and to report subjects path through the study (CONSORT diagram).

Question 122

It is not always feasible to blind the subjects and assessors based on the nature of the intervention, name a possible solution for this dilemma.

Answer 122

If possible choose an outcome not affected by observer bias, for example the 2-min timed-up-and-go test is an assessment that could be used as an outcome free of observer bias.

Question 123

Explain single and double blinding strategies in research study design and explain the difference between the two in effectiveness of reducing bias.

Answer 123

Single blinding involves masking the participants to the treatment they are receiving, while the investigators are aware of the treatment assignments.

Double blinding involves masking both the participants and the investigators to the treatment assignments.

Double blinding is generally considered to be more rigorous and effective at reducing bias compared to single blinding. This is because double blinding eliminates the potential for bias due to both the participant and the investigator knowing the treatment assignments, and it reduces the potential for placebo effects or other factors that could influence the study outcomes.

However, double blinding may not always be feasible or necessary in certain types of studies. For example, in some studies, it may be difficult or impossible to blind the participants, such as in surgical interventions. In other studies, blinding may not be necessary or may not be possible due to the nature of the intervention or treatment being studied.

Question 124

What is effect size?

Answer 124

Effect size refers to the magnitude or strength of a statistical relationship or difference between two variables. It provides a standardized measure of the size of an effect, allowing researchers to compare the strength of effects across different studies or experimental conditions. (magnitude of the effect due to the intervention)

Question 125

Smaller effect size require _______________ samples to detect.

Answer 125

Larger samples.

Question 126

Interpret the effect size of the following:

0.0 to 0.2 =
0.2 to 0.5 =
0.5 to 0.8 =
0.8 and above =

Answer 126

0.0-0.2 = trivial effect size
0.2-0.5 = small effect size
0.5-0.8 = moderate effect size
0.8 and above = strong effect size

Question 127

While_____________________ indicate whether an effect is statistically significant (i.e., unlikely to have occurred by chance), ___________________indicates how large the effect is in practical terms.

Answer 127

p-values

effect size

Question 128

For example, a ____________ effect size may be statistically significant in a large sample size, but it may not be clinically relevant. On the other hand, a ___________ effect size may be clinically relevant even if it is not statistically significant, particularly if the sample size is small or the effect is difficult to measure accurately.

Answer 128

Small

Large

Question 129

Explain why a trial may be stopped early if an effect size occurs.

Answer 129

A trial may be stopped early if the effect size is very large, which means that the intervention being tested is having a much greater impact than expected. Stopping a trial early due to a large effect size is often done for ethical reasons, as it allows participants in the control group to receive the intervention that is proving to be effective.

For example, if a trial is testing a new drug for a life-threatening disease and the effect size is very large, it may be deemed unethical to continue the trial with a control group receiving a placebo, as it would deny them the potentially life-saving treatment. In such cases, the trial may be stopped early, and the treatment may be made available to all participants in the trial.

Question 130

Explain possible reasons for an effect size having an influence on a decision to stop a trial.

Answer 130

• Unexpected harmful effects
• Unexpected beneficial effects
• Futility in that no effect is likely ever to be observed
• Data quality issues that need to be corrected if the trial is to continue

Question 131

A Consort diagram is a useful tool to review for detecting…

Answer 131

Selection bias

Question 132

P-value tells you…

Answer 132

if the results produced by the intervention are statistically significant when compared to the results produced by the control group.

Question 133

Explain the null hypothesis

Answer 133

The null hypothesis is a statement that there is no statistically significant difference between two groups, or no statistically significant relationship between two variables.

Question 134

It is a hypothesis of no effect or no difference.

Answer 134

Null hypothesis.

Question 135

In a study comparing two treatments for a particular disease, the ______________________ would be that there is no statistically significant difference in outcomes between the two treatments.

Answer 135

Null hypothesis.

Question 136

Explain the alternative hypothesis.

Answer 136

The alternative hypothesis is a statement that contradicts the null hypothesis and suggests that there is a statistically significant difference or relationship between two groups or variables being compared.

Question 137

It is the opposite of the null hypothesis and is often denoted as H1.

Answer 137

Alternative hypothesis.

Question 138

When the p-value is lesser than or equal to alpha level…

When the p-value is greater than or equal to alpha level…

Answer 138

When the p-value is lesser than or equal to the p-value, we reject the null hypothesis (the groups are different and this difference is statistically significant)

When the p-value is greater than or equal to the p-value, we fail to reject the null hypothesis (we accept the null hypothesis) (the results are not statistically significant)

Question 139

Explain the alpha level.

Answer 139

The alpha level, also known as the significance level, is the probability level used in hypothesis testing to determine whether the null hypothesis should be rejected or not. It is denoted by the symbol α and is usually set at 0.05, 0.01, or 0.001.

Question 140

Confidence interval = _________________

Answer 140

CI = 1 - alpha level

Question 141

For RCTs the null value = _____

Answer 141

0

Question 142

The result would not be statistically significant if the confidence interval includes the _____________.

Answer 142

Null value.

Question 143

Designs that do not randomize the sample, or mechanically manipulate independent variable but may have a control group.

Answer 143

Non-experimental designs.

Question 144

Surveys are a type of __________________ study design, when used to make comparisons at a single point in time.

Answer 144

cross-sectional

Question 145

Cross-sectional study designs are __________________(what type of broader study design).

Answer 145

observational.

Question 146

We are interested to know the prevalence of ADHD among students in primary schools of city of X. We design a population-based survey to assess the prevalence of this condition. We go to all the schools that were supposed to be included in the study and examine the population. What type of study design is this?

Answer 146

cross-sectional.