Study Design Overview Flashcards
When Observation > RCT
RCT might be unethical…
- In vulnerable populations (pregnant)
- Address research questions (use lithium and risk of motor vehicle accidents)
- Examine the risk/benefit of the medication which is considered a standard of care (unethical to deny medicine in RCT)
Study Design
- Descriptive
- Analytic: Observational or Experimental
Descriptive Studies
- Correlation (ecologic)
- Case reports
- Case series
- Cross-sectional analysis
Observational Studies
- Case-cohort
- Cohort
Experimental Studies
-Clinical Trials
Descriptive Studies used for…
- Describing novel or unusual signs, symptoms, or events (case reports/series)
- Describing the distribution of a disease in relation to person, place and time (cross-sectional, correlation)
- Hypothesis generation (all descriptive studies)
Cross-Sectional
“Snapshot” of information on disease outcomes and exposures at a single point of time
Analytic Studies used for….
-Testing research hypotheses formulation from descriptive studies
Observational Research Design
- Investigator simply observes the natural course of events
- Notes who is exposed/unexposed has or has not developed the outcome
Experimental Research Design
- Investigate allocate the exposure
- Follow subjects for subsequent development of disease
Cohort
- Observational study
- Start with people free of disease
- Identify exposed and unexposed subjects
- Follow them up until disease develops
- Compares the incidence of disease among exposed and unexposed (RR)
Case-Control
- Start with people who have a disease (cases)
- Identify appropriate controls
- Look back in time to identify exposure status among cases and controls
- Compare the odds of exposure among cases and controls (OR)
Comparative Effectiveness Studies
- Compare two+ agents or interventions that are considered therapeutic alternatives
- Examination of effects in actual practice
- Conventional approach for CER-observational studies
Determining Causality - Case Reports
- Drug adverse event algorithm
- Naranjo Algorithm
Evaluating Drug Safety- Post Market
- Inherited limitations
- New ADRs are discovered postmarket in most drugs
Safety Signals
Concern about an excess of AE comparedto what would be expected to be associated with product’s use
Adverse Events
Undesirable outcome which may or may not be due to the drug itself (causality unknown)
Adverse Drug Reaction
Reasonable possibility of a causal relationship between drug use and the event exists
Serious ADE
- Results in death
- Life-threatening
- Hospitalization, disability, or congenital abnormality
- Required intervention to prevent permanent impairment or damage
Evaluating Safety Signals
Spontaneous Reporting
- Health professionals can report to FDA or manufacturer
- Manufacturer reports to FDA (within 15 days if serious)
- Patients may also report directly to FDA
Active Surveillance
-Integrated administration claims databases
Pharmacoepidemiology methods are used to evaluate safety signals
FDA Evaluation of Signals
- Factors which determine whether a signal requires further investigation
- SNIP Criteria
SNIP Criteria
S - Strength
N - New?
I - Importance clinically
P - Preventative measures
FDA Responses to Signals
- Label changes
- Name change - to avoid confusion (baby aspirin)
- Package change
- Issue “Dear Health Professional” Letter
- Restrict use of drug
- Work with manufacturer regarding possible withdrawal from US market
