/
Flashcards from PDF
Save to my account
Sign up
Flashcards from PDF
Flashcards from PDF
Study
1
Question
Why is combining evidence from multiple studies beneficial compared to relying on a single study?
Answer
Combining evidence guards against disadvantages of single studies such as inadequate sample size (risk of false negatives), variability due to chance, and subtle design or participant differences. It increases statistical power, potentially revealing significant effects missed by individual trials, and avoids the ethical issues of conducting unnecessary new research when effects are already indicated.
2
Question
What does "statistical power" mean in the context of clinical studies and why is it important?
Answer
Statistical power is the probability that a study correctly rejects the null hypothesis (detects an effect when there actually is one). It is important because insufficient power can lead to false negatives, where a real effect is missed due to too small sample size or inadequate study design.
3
Question
What are the risks of only using English-language articles or a single database when conducting a systematic review?
Answer
Using only English-language articles can bias results because relevant studies might exist in other languages, and translated trial reports tend to overstate intervention effects. Searching only one database risks missing relevant studies since not all articles are indexed there, reducing the thoroughness and potentially skewing conclusions.
4
Question
How is 'internal validity' defined, and why is it critical in clinical research?
Answer
Internal validity assesses whether observed changes in the dependent variable can confidently be attributed to manipulation of the independent variable rather than to other factors. It is critical because poor internal validity means the study's findings can't reliably establish cause and effect, undermining the truthfulness of conclusions.
5
Question
What strategies are recommended to improve the quality and reproducibility of quality assessments in systematic reviews?
Answer
Use a clear assessment strategy with predetermined scoring systems, more than one independent assessor to reduce errors, and pre-agreed checklists to increase objectivity and transparency. Addressing assessor disagreements clearly also improves reproducibility.
6
Question
What is publication bias, and how can funnel plots help detect it?
Answer
Publication bias refers to the tendency for studies showing no effect or undesirable results to be unpublished or overlooked. Funnel plots display treatment effects against study size; ideally, they look like a symmetrical pyramid. Asymmetry suggests possible publication bias or systematic differences between smaller and larger studies.
7
Question
What distinguishes clinical heterogeneity from statistical heterogeneity in meta-analysis?
Answer
Clinical heterogeneity refers to differences in the characteristics of the studies, such as variations in forms of treatment or participant populations. Statistical heterogeneity refers to variability in the estimates of treatment effect between studies that exceeds what would be expected by chance alone.
8
Question
What is the difference between fixed effect and random effect models in meta-analysis with respect to heterogeneity?
Answer
When I² statistic is less than 50%, studies are considered homogenous and a fixed effect model is used assuming one true effect size. When I² is over 50%, studies are considered heterogeneous; a random effects model is used which accounts for variability between study effects beyond chance.
9
Question
Why is distinguishing between primary and secondary outcomes important in randomized controlled trials (RCTs)?
Answer
Primary outcomes are the main measures judged to assess an intervention's effectiveness and are used to determine sample size and power. Secondary outcomes are additional measures of interest but not the main basis for conclusions. Distinguishing them helps ensure the study is appropriately powered and results interpreted correctly.
10
Question
How does randomization improve the validity of RCTs compared to observational studies?
Answer
Randomization ensures groups are likely similar in both known and unknown factors affecting outcomes, reducing selection bias and confounding. This means observed outcome differences can be more confidently attributed to the intervention. Observational studies lack randomization, raising risks of imbalance and confounding.
11
Question
What are some common randomization methods and their strengths or weaknesses?
Answer
Simple randomization assigns patients purely by chance (like flipping a coin), but can lead to chance imbalances. Stratified randomization ensures equal distribution of confounders across groups. Minimisation uses computer algorithms to balance groups. Bad methods include predictable assignments like date of birth or surname initial, which can introduce bias.
12
Question
What is allocation concealment and why is it important in clinical trials?
Answer
Allocation concealment means the person assigning participants to treatment groups does not know which treatment the next participant will get. This prevents selection bias by ensuring treatment assignments cannot be anticipated or manipulated.
13
Question
Describe the role of blinding in clinical trials and differentiate between single and double blinding.
Answer
Blinding prevents bias by hiding treatment allocation. Single blinding means the patient is unaware of their treatment group, reducing placebo or reporting bias. Double blinding means both patient and clinician (and sometimes assessors) are unaware, reducing detection and assessment biases based on expectations.
14
Question
What is intention-to-treat analysis (ITTA), and how does it help reduce bias?
Answer
ITTA analyzes participants according to the treatment group they were originally assigned, regardless of whether they completed or adhered to the treatment. This reduces attrition bias from dropouts and provides a pragmatic estimate of treatment effect reflecting real-world adherence.
15
Question
Why can attrition bias compromise the internal validity of a study?
Answer
Attrition bias occurs when participants drop out in a way that is not random and differs systematically between groups, affecting outcome measurements. This loss of participants can reduce study power and distort results, leading to false negatives or biased effect estimates.
16
Question
What is the concept of 'equipoise' in clinical trial ethics and what happens if it is lost?
Answer
Equipoise means genuine uncertainty about which treatment is better. It is the ethical basis for assigning patients randomly. If evidence clearly shows one treatment is superior, equipoise is lost, and the trial should stop to avoid withholding effective treatment.
17
Question
What key features affect the external validity of clinical trials?
Answer
External validity concerns how generalizable the results are to other populations or settings. It depends on participant characteristics (similarity to your patients), geographic location, healthcare system context, and whether economic or practical aspects of implementing the treatment have been evaluated.
18
Question
What are the four main approaches to limit confounding in observational studies?
Answer
The four approaches are: restriction (limit participants with confounders), matching (create groups matched on confounders), stratification (analyze subgroups separately to balance confounders), and multiple variable regression (statistically adjusting for confounders to model their effects).
19
Question
What are some common types of bias encountered in clinical research, and how can they affect study outcomes?
Answer
Common biases include selection bias (systematic errors in group creation), sample bias (unrepresentative participants), volunteer bias, non-responder bias, performance bias (differences in care other than interventions), expectation bias, and information bias. These can distort estimated effects away from the truth, either exaggerating or underestimating them.
20
Question
How can poor calibration of measurement instruments lead to measurement bias in quantitative studies?
Answer
If instruments are not properly calibrated (e.g., scales not zeroed), measurements can be consistently over- or underestimated, creating systematic errors that bias the data and conclusions.
21
Question
What is differential follow-up in cohort studies, and why is it important?
Answer
Differential follow-up refers to whether participants in different study groups are followed up in the same way and for a sufficiently long period. It is important because unequal or insufficient follow-up can introduce bias, affecting the study's validity and the ability to detect true associations.
22
Question
List some advantages and disadvantages of cohort studies.
Answer
Advantages: They can study large cohorts with multiple outcomes and are suitable for studying rare exposures. Disadvantages: They require a large sample size, long follow-up periods (which increase cost and potential loss to follow-up), risk of selection bias, misclassification of exposure, and biases if assessors are not blinded.
23
Question
What factors affect the internal validity of an observational study?
Answer
Internal validity is affected by selection bias, differential follow-up, accurate outcome measurement, misclassification of exposure, and confounders.
24
Question
How is external validity assessed in clinical studies?
Answer
External validity depends on whether the population and setting of the study are similar to the one you care about, if the findings can be realistically implemented (e.g., in primary or secondary care), and whether the intervention is cost-effective in that context.
25
Question
Describe the main characteristics, advantages, and disadvantages of cross-sectional studies.
Answer
Cross-sectional studies analyze participants at a single point in time, comparing those with and without disease regarding risk factors. Advantages include being cheap, quick, and able to observe multiple outcomes. Disadvantages are that they make it difficult to establish causal associations.
26
Question
What is the primary purpose of controls in a case-control study?
Answer
Controls provide a comparison group from the same population without the outcome of interest, allowing analysis of differences in exposure to causative factors between cases and controls.
27
Question
What are key sources of bias specific to case-control studies?
Answer
Recall bias, selection bias, ascertainment bias, and misclassification (especially differential misclassification) can all introduce bias in case-control studies. Controls should be selected from the same population and recruited acceptably to reduce bias.
28
Question
Explain the difference between differential and non-differential misclassification.
Answer
Differential misclassification occurs when information collected from one group is accurate but from another is inaccurate, biasing results either towards or away from the null. Non-differential misclassification occurs when information from both groups is inaccurate, generally biasing results toward the null hypothesis.
29
Question
What strategies can enhance the accuracy of outcome measurement in studies?
Answer
Use standard application of outcome measurement, validate algorithms, ensure comprehensive population coverage, establish outcomes blind to exposure status, and use multiple independent assessors.
30
Question
What errors can inaccurate outcome measurement lead to in research studies?
Answer
Inaccurate outcome measurement can cause systematic bias resulting in type 1 (false positive) or type 2 (false negative) errors, and differential misclassification can bias results toward or away from the null hypothesis.