Study design overview Flashcards
Cross-Sectional Study
When to Use: To measure the prevalence of a disease or condition at a specific point in time and to examine associations between exposure and outcome at that point.
**Pros: ** Quick and relatively inexpensive, provides a snapshot of the population’s health, useful for generating hypotheses.
**Cons: ** Cannot establish causation, can be subject to recall bias, may not capture changes over time, temporal sequence cannot be determined, unsuitable for rare outcomes, underrepresent short druation disease and over represent long duration disease, Prevelance incidence bias cases are survivors
**USES ODDS RATIO **- Odds of developing the disease among the exposed /odds of developing the disease among non-exposed
Cohort Study
**When to Use: **To measure the incidence of diseases and assess the effects of exposure over time. Useful for rare exposures or long-term health outcomes.
Pros: Allows for the assessment of temporal relationships, can estimate risk and relative risk, useful for investigating rare exposures or outcomes, incidence rates can be calculated, recall bias is minimized, multiple outcomes can be studied
Cons: Expensive and time-consuming, attrition and loss to follow-up may occur, may not be suitable for rare outcomes.
A study is conducted to investigate the long-term impact of a high-fiber diet on the development of colorectal cancer in a cohort of individuals over 10 years.
Case-Control Study:
**When to Use: **To identify risk factors for a specific disease or condition. Ideal for studying rare diseases or outcomes, disease has long induction/latency period, evaluate many risk factor for same disease.
**Pros: **Efficient for studying rare diseases, quicker and less costly, useful for identifying risk factors.
**Cons: **Cannot establish causation, selection bias, information bias (recall), not suitable for studying rare exposures, cannot ascertain temporal relationship between exposure and disease as we are going back in time
Researchers aim to investigate the risk factors associated with the development of lung cancer in a group of patients diagnosed with the disease and compare them with a group of individuals without lung cancer.
In the early 1940s, Alton Ochsner, a surgeon in New Orleans, observed that virtually all ofthe patients on whom he was operating for lung cancer gave a history of cigarette smoking.
OUTCOME (LUNG CANCER)–> EXPOSURE (SMOKING)
In utero rubella & cataract: In the 1940s, Sir Norman Gregg, an Australian ophthalmologist, observed a number ofinfants and young children in his practice who presented with an unusual form of cataract. Gregg noted that these children had been in utero during the time of a rubella (German measles) outbreak.
OUTCOME (CATARACT)–> EXPOSURE (IN UTERO RUBELLA)
**USES ODDS RATIO **- Odds of developing the disease among the exposed /odds of developing the disease among non-exposed
Randomized Controlled Trial (RCT)
When to Use: To assess the efficacy of an intervention or treatment. Ideal for testing the impact of a new treatment or preventive measure.
Pros: Provides strong evidence for causation, randomization helps control for confounding, results can be generalized to a broader population.
Cons: Expensive and time-consuming, may not be ethical or feasible for certain research questions, requires a controlled environment
A pharmaceutical company wants to evaluate the effectiveness of a new medication in reducing the frequency of migraine attacks among a group of patients with a history of frequent migraines.
Repeated Cross-sectional studies
Same information collected (snapshot) in different samples over time.
**Use: ** examine changes in population over time
pros: Not impacted by aging
Better estimate of prevalence
Cons: cannot tell us about incidence as it is a snapshot
Case Cross over studies
Type of case control study where **cases serve as their own control. **
When to use: Acute outcomes (severe intense, short), transient (temporary) exposure
Commonly used in environmental epidemiology
Multi-level studies
Exposure at community level
Outcome at individual level
e.g a program is distributed across schools and the outcome would be improved scores of a student following that program in that particular school
has a hierarchiel structure (nested structure) of observation.
interest in studying school level,
Nested case control
Control is selected everytime a case develops, matched on the duration of follow-up ; follow up a bunch a people there is a case then select a control
Case-cohort
all controls are selected from the cohort when the study started
pros: no recall bias, as selected from the same cohortt selected control from the very begining recall bais will not be differntial to cases and control.
logistically easier and less costly. follow them up there is one, two cases… enough cases then select controls at baseline. controls are selected from the subset of the cohort from the very begining.
Natural experiment
researchers take advantage of a naturally occurring event or situation that mimics the conditions of intervention study. Unlike randomized and controlled interventions, a natural experiment observing the effects of an exposure or intervention in a ‘real world’ setting
Data linkage study
combining data from
multiple sources, often collected for different purposes
Randomized control trial
pros -
equalize characters of the control and treatment group reduce confounding
double blind RCT gold standards for evaluating clinical treatments
Randomization decreases the chance of confounding
Temporailty intact
cons
expensive
might or might not be generalizable
time consumign
may expose patients to harm or delay effective interventions.
clinical equipoise
genuine uncertainity of the effects of treatment whether good or bad.
Intention to treat
retains original group assignment, ignores deviation in actual treatment obtained
**Retains randomization **
When participants drops out or cross over groups underestimates the intervention effectiveness
Once randomized always analysed
Per-protocol
Used in efficacy trials
- Analyse differences between those randomized to and actually receiving experimental and alternative treatment.
-
only analyze those who adhered to their assigned treatment; patinets are basically choosing their own groups
