Lecture 7 Flashcards
Study design 2 (23 cards)
What is the main feature of a cross-sectional study?
It measures exposure and outcome at the same point in time
What type of population is used in cross-sectional studies?
A defined population or sample at a specific point or period
What are common sampling methods used in cross-sectional studies?
Simple random sampling
Stratified sampling
Cluster sampling
Systematic sampling
How is data typically collected in cross-sectional studies?
Via surveys, questionnaires, interviews, or physical/clinical assessments
What are the main limitations of cross-sectional studies?
Cannot establish causal direction
Temporality between exposure and outcome is unclear
What types of outcomes can cross-sectional studies measure?
Prevalence of disease, risk factors, or health behaviours
What statistical methods are commonly used in cross-sectional analysis?
Prevalence ratios
Chi-square tests
Logistic regression
What is a classic use of cross-sectional studies in epidemiology?
National health surveys like NHANES (US) or Health Survey for England (HSE)
What are routine data studies?
Studies using pre-collected, administrative or registry data for health research
What are key data sources for individual-level routine data in cancer research?
Cancer registration systems
Death certification records
Hospital episode statistics
What individual-level information is typically collected in routine data?
Demographics (age, sex, location)
Diagnosis (ICD codes, cancer site, stage)
Clinical info (treatment, comorbidities, survival time)
What measures are used to ensure data quality in routine health datasets?
Completeness
Accuracy
Timeliness
Consistency and standardisation
What is one strength of using routine data?
Large sample sizes and national representativeness
What is a major limitation of routine data studies?
Lack of control over variable collection; potential for coding errors or missing data
Name a landmark UK study that used routine cancer registry data
Allemani et al. CONCORD Study – global comparison of cancer survival using registry data
What characterises ecological studies using routine data?
Use of aggregated group-level data (e.g., region, country), not individual-level data
What is the main type of analysis used in ecological studies?
Correlation analysis
Linear regression between exposure and outcome variables across populations
What is the main limitation of ecological studies?
Ecological fallacy – assuming group-level associations apply to individuals
Give an example of an ecological study using routine data
Comparing lung cancer mortality and smoking prevalence across countries (e.g., WHO datasets)
What is an appropriate use case for ecological studies?
Studying population-level trends (e.g., effects of screening programs or tobacco policies)
What registry-based system tracks all cancer cases in the UK?
The National Cancer Registration and Analysis Service (NCRAS)
How is death certification data used in cancer epidemiology?
For calculating mortality rates, identifying cause of death, and analyzing trends over time
What famous global project used both individual and aggregated cancer survival data across countries?
The CONCORD Programme (Allemani et al.)
