ABFM Research Methods & Evidence-Based Practice 3 — Questions and Answers
Question 1: Which element of PICO is represented by 'forensic DNA analysis' in a research question comparing identification methods?
- Population
- Intervention/Index test (Correct answer)
- Comparison
- Outcome
Correct answer: Intervention/Index test
In PICO, the 'I' represents the Intervention or Index test being evaluated, which in this case is forensic DNA analysis.
Question 2: A forensic researcher wants to assess the accuracy of a new bite mark analysis technique. Which statistic simultaneously reflects both sensitivity and specificity across all thresholds?
- Positive predictive value
- Likelihood ratio
- Area under the ROC curve (AUC) (Correct answer)
- Cohen's kappa
Correct answer: Area under the ROC curve (AUC)
The AUC of a receiver operating characteristic curve summarizes diagnostic accuracy across all possible cut-off thresholds into a single index.
Question 3: In a forensic toxicology study, inter-rater reliability between two examiners is best quantified using which statistic?
- Pearson correlation
- Cohen's kappa
- Cronbach's alpha
- Intraclass correlation coefficient (Correct answer)
Correct answer: Intraclass correlation coefficient
Intraclass correlation coefficient (ICC) is preferred for quantitative or ordinal ratings by multiple raters because it accounts for both agreement and consistency.
Question 4: A forensic study with n=12 autopsy cases uses a non-parametric test for group comparison. Which assumption was likely violated?
- Independence of observations
- Normal distribution of the outcome variable (Correct answer)
- Equal group sizes
- Random sampling
Correct answer: Normal distribution of the outcome variable
Non-parametric tests are chosen when the normality assumption is violated, which is common with small samples like n=12.
Question 5: Which research framework is most appropriate for investigating the lived experiences of families following a forensic autopsy in a medicolegal context?
- Randomized controlled trial
- Prospective cohort study
- Qualitative phenomenological study (Correct answer)
- Cross-sectional survey
Correct answer: Qualitative phenomenological study
Phenomenological qualitative research is designed to explore subjective human experiences and meaning, making it ideal for understanding families' experiences.
Question 6: Publication bias in a meta-analysis of forensic evidence reliability studies is best visualized using which tool?
- Forest plot
- Funnel plot (Correct answer)
- Kaplan-Meier curve
- Box-and-whisker plot
Correct answer: Funnel plot
Funnel plots display effect size against study precision, and asymmetry in the plot suggests potential publication bias.
Question 7: In evidence-based forensic medicine, which GRADE category applies to evidence from well-designed observational studies with consistent, precise results?
- High
- Moderate (Correct answer)
- Low
- Very low
Correct answer: Moderate
GRADE classifies well-designed observational studies as 'Moderate' quality evidence by default, which can be upgraded under certain conditions.
ABFM Research Methods & Evidence-Based Practice 3
Which element of PICO is represented by 'forensic DNA analysis' in a research question comparing identification methods?