Examkin

Question

Difficulty: Very hardEpidemiologic Study Designs

Investigators are evaluating the association between long-term use of a specific glucagon-like peptide-1 (GLP-1) receptor agonist and the risk of developing a rare subtype of medullary thyroid carcinoma. They utilize a comprehensive national health insurance database containing records for over 5 million adult patients over the last 15 years. The investigators first identify 45 patients diagnosed with this specific carcinoma. For each affected patient, they randomly select five patients without the carcinoma from the same database, matched on birth year, biological sex, and family history of endocrine neoplasms. They then query the linked pharmacy dispensing records during the 10 years prior to diagnosis to ascertain GLP-1 receptor agonist exposure. Which of the following statements regarding this study's methodology are correct?

  1. The measure of association calculated in this study will closely approximate the relative risk because the outcome being investigated is rare.Answer
  2. B
    This study is classified as a retrospective cohort design because both the exposure and the outcome occurred prior to the initiation of the investigation.
  3. C
    The matching strategy was primarily implemented to evaluate if a family history of endocrine neoplasms acts as an effect modifier.
  4. The reliance on pharmacy dispensing records rather than patient interviews to ascertain medication exposure effectively minimizes recall bias.Answer
  5. E
    This epidemiological design allows the investigators to directly calculate the true incidence of the medullary thyroid carcinoma among GLP-1 receptor agonist users.

Answer

The correct statements are that the measure of association will approximate the relative risk due to the rare disease assumption, and that using objective pharmacy records minimizes recall bias.
The study design described is a case-control study, as participants were initially selected based on their disease status (presence or absence of the carcinoma). In such designs, the odds ratio is the calculated measure of association. Because medullary thyroid carcinoma is extremely rare, the rare disease assumption applies, which means the calculated odds ratio will mathematically approximate the true relative risk. Furthermore, utilizing objective, electronic pharmacy dispensing records instead of relying on patient questionnaires to ascertain prior drug exposure effectively eliminates recall bias—a classic vulnerability of case-control studies.

Step-by-Step Solution

1
Identify the primary study design based on how participants were selected.
The investigators selected patients based on the presence or absence of the disease (medullary thyroid carcinoma) and then looked backward to determine exposure. This confirms a case-control study design.
Identifying the study type is crucial for understanding which statistical measures can be applied and what biases are most likely.
2
Evaluate the statistical limitations and assumptions of the identified design.
Case-control studies report odds ratios (OR) rather than relative risks (RR) and cannot measure true incidence. However, because the studied cancer is extremely rare, the rare disease assumption holds, allowing the OR to approximate the RR.
Applying epidemiological principles to the specific context of the disease reveals which statistical inferences are mathematically valid.
3
Assess the impact of the data collection methods and matching criteria on study validity.
Using existing pharmacy records prevents differential memory reporting, thereby mitigating recall bias. Matching patients on specific characteristics (like family history) neutralizes those characteristics as confounders, though it does not test for effect modification.
Methodological choices directly impact a study's susceptibility to bias and confounding, which are frequently tested concepts.

Key Concept

Case-control study methodology: participant selection, the rare disease assumption, and methods for mitigating recall bias and confounding.
Rate this question

Topics

Epidemiologic Study Designs

More questions from this topic