Flashcards
Observational Studies
Cards in this deck (32)
- 1Case report/Case seriesDescriptive
No comparison group, only the case being described
Susceptible to bias
Cannot use to make treatment decisions - 2How to evaluate case reportsRare or unusual?
Denominator?
Case-fatality rate?
Is association with putative exposure real? - 3Cross Sectional studiesDescriptive
Prevalence studies in exposed vs. unexposed
Individual is unit of observation and analysis - 4WeaknessNo cause - effect
Prevalent cases are survivors
Measures at one point in time - No temporality - 5Content of 2 by 2 tableExposure (diseased and no disease) vs non exposure (disease and no disease)
Compare proportions
Can use Chi squared test - 6Point EstimatesThe effect size observed in a particular study and best estimate from the data, of the true but unknown effect size in the population
- 7Odds RatioPoint estimate used in case control study
- 8Relative risk and attributable riskPoint estimate used in Cohort studies
- 9RRR, ARR, NNTPoint estimate used in Randomized clinical trials
- 10Confidence intervalExpresses statistical precision, how sure we are that the point estimate is included in a range of values
- 11No effect meansRR and OR = 1
Measured variables (AR and average) - 0 - 12Case controlAnalytical
Outcome study
Disease vs. No disease
Take case with disease and match with control who have everything except disease
Focus on outcomes then look for exposure - 13Bias for case controlOutcome presence may affect patient recollection of exposure
So ensure all subjects are asked same questions - 14Case control strengthExcellent for rare disease
Smaller in size so quick and cheaper
Can evaluate multiple risk factors for one disease
Can use secondary data
Can test hypothesis - 15Case control studies weaknessCannot determine incidence directly so estimate RR with Odds Ratio
Selecting cases
Selecting controls
Uncertainty in exposure - disease time relationship
Bias potential - 16Content of 2 by 2 table for case controlDiseased (exposed and non exposed) vs. Not diseased (exposed and non exposed)
- 17Odds ratio formulaodds of exposure if case/odds of exposure if control
ad/bc - 18OR = 1No association between exposure and likelihood of disease
- 19OR > 1Odds of exposure in cases is greater than odds of exposure in controls
- 20OR < 1Odds of exposure in cases is less than odds of exposure in controls
- 21Cohort StudyRisk factor vs No risk factor
Longitudinal study
Analytical
Cause to effect
Exposure cannot be assigned (observational study) cos it's dangerous - 22Cohort Studies - StrengthsDirectly determines risk
Can make hypothesis
Assess relationship between an exposure and many diseases
Good method when exposure is rare - 23Cohort study weaknessRequires large sample size
Inefficient for rare disease
Takes a long time to complete
Expensive
Subject attrition affects Validity - 24Cohort study measureRisk ratio - Relative risk (incidence in exposed/ incidence in non exposed)
Atrributable risk - Incidence in exposed - incidence in non exposed - 25Relative risk (Risk ratio) (cumulative incidence){a/(a+b)}/{c/(c+d)}
- 26Relatie risk (risk ratio) Incidence density and person time(a/PT in exposed) / (c/PT in unexposed)
- 27RR = 1No association between exposure and risk of disease
- 28RR > 1Exposure is associated with increased risk of disease
- 29RR < 1Exposure is associated with decrease in risk of disease
- 30AR = 0No association between exposure and risk of disease
- 31AR > 0Exposure is associated with excess amount of disease in exposed
- 32AR < 0Exposure is associated with decreased amount of disease in the exposed
Information
- Description
- Flashcards on Observational Studies, created by sotee4u on 03/08/2015.
- Created by
- sotee4u
- Views
- 0
- Rating