Case-control study Martin van der Esch, PhD Amsterdam Rehabilitation Research Center | Reade Case-control Study – Design Select subjects on the basis of disease status Disease + - Exp + a b Exp - c d d1 d0 Amsterdam Rehabilitation Research Center | Reade 1 • Odds of exposure among cases = a / c • Odds of exposure among controls = b / d Disease case control Exp + a b Exp - c d d1 d0 Amsterdam Rehabilitation Research Center | Reade 2 RR OR Amsterdam Rehabilitation Research Center | Reade Fletcher RH et al; Clinical epidemiology; LWW; 2014: pg 87 Analysis of CCS: The OR as a measure of association • The only valid measure of association for the CCS is the Odds Ratio (OR) • Under reasonable assumptions (– the rare disease assumption) the OR approximates the RR. • OR = Odds of exposure among cases (disease) Odds of exposure among controls (non-dis) • Odds of exposure among cases = a / c • Odds of exposure among controls = b / d • Odds ratio = a/c = axd [= cross-product ratio] b/d bxc Amsterdam Rehabilitation Research Center | Reade 4 Example CCS - Smoking and Myocardial Infarction Study: Desert island, population = 2,000 people, prevalence of smoking = 50% [but this is unknown], identify all MI cases that occurred over last year (N=40), obtain a random sample of N=40 controls (no MI). What is the association between smoking and MI? MI + - Smk + 30 20 Smk - 10 20 40 40 OR = a . d = 30 . 20 = 3.0 (same as the RR!) c.b 10 . 20 5 Amsterdam Rehabilitation Research Center | Reade Odds Ratio (OR) Similar interpretation as the Relative Risk OR = 1.0 (implies equal odds of exposure - no effect) ORs provide the exact same information as the RR if: • controls represent the target population • cases represent all cases • rare disease assumption holds (or if case-control study is undertaken with population-based sampling) Remember: • • • OR can be calculated for any design but RR can only be calculated in RCT and cohort studies The OR is the only valid measure for CCS Publications will occasionally mis-label OR as RR (or vice versa) Amsterdam Rehabilitation Research Center | Reade 6 Controlling extraneous variables (confounding) Exposure of interest may be confounded by a factor that is associated with the exposure and the disease i.e., is an independent risk factor for the disease B A C Amsterdam Rehabilitation Research Center | Reade 7 How to control for confounding At the design phase • Randomization • Restriction • Matching At the analysis phase • Age-adjustment • Stratification • Multivariable adjustment (logistic regression modeling, Cox regression modeling) Amsterdam Rehabilitation Research Center | Reade 8 Matching is commonly used in CCS Control an extraneous variable by matching controls to cases on a factor you know is an important risk factor or marker for disease • Example: • Age (within 5 years) • Sex • Neighbourhood If factor is fixed to be the same in the cases and controls then it can’t confound Amsterdam Rehabilitation Research Center | Reade 9 Design of a case-control study Fletcher RH. Clinical epidemiology 2014: pg 82 Amsterdam Rehabilitation Research Center | Reade Case-control design Identifying basicpopulation collecting Measuring exposition factor(s) and covariables patients effect? Measuring exposition factor(s) and covariables controls selecting time Amsterdam Rehabilitation Research Center | Reade Steps to follow in case-control study Amsterdam Rehabilitation Research Center | Reade Case-control study: steps to follow 1. Selecting study population 1a. Selection patients 1b. Selection controls 2. Measuring exposition factors (blinded) - Central determinant - Covariables (confounders and moderators) 3. Analysis - Unadjusted ‘Rough’ effect - Adjusted effect (controlled for covariables) Amsterdam Rehabilitation Research Center | Reade 1a. Selecting patients SOURCE: ‘counters’ in healthcare Hospital (department) Practice of physician Depending of illness definition screeningsprogram INCIDENT or PREVALENT cases? Prevalent cases - Too many less severe cases (especially in lethal diseases) - Disease influences exposition-status (especially in ‘variable’ determinants) Incident cases - Long ‘intake’-duration (especially in rare diseases) - Severe diseases: ethical + practical objections - Lethal diseases: † cases † controles ? - Clear beginning disease, diagnostics ? Amsterdam Rehabilitation Research Center | Reade Selection-bias Information-bias 1b. Selection controles • Illness-free (or not having the disease of study) Medical history? Assessments • Biomedical confirmation? Source of origin Theory Study population = Population of all persons who could have been at the hospital, and could have been diagnosed with the disease of study and having the disease Private/primary pratice Hospital Open population • Inclusion- and en exclusion criteria • Matchen ? Amsterdam Rehabilitation Research Center | Reade Area 1b. Selection controles (continued) SOURCE: Hospital (hospital-based case-control study) - Same mental state - Comparable interview conditions - High respons No information-bias - Which disease(s) ? selection-bias exclude etiological simularity ! Open population (population-based case-control study) - Population register - ‘random digit dialing’ matching Area controles - aselect, random sample No selection-bias - Different interview conditions information-bias - Bad respons, selective Still selection-bias ? - ‘overmatching’ Amsterdam Rehabilitation Research Center | Reade 1b. Selection controles (continued) IN- en EXCLUSION criteria: ‘at risk’? Historia morbi? Same as for cases Great danger: selection bias MATCHEN: Matching to reduce the chance of supposed confounders Consequences for validity en efficiency? Amsterdam Rehabilitation Research Center | Reade 2. Measurement of exposition-factors 1. Central determinant 2. Potential confounders / effectmodificators Danger: • Measurements incorrect / not comparable • Measurements incomplete Solution: • ‘Blinded’ measurement • Independent sample of participants Problem: • Timing of measurements (relevant timeframe?) Amsterdam Rehabilitation Research Center | Reade Information-bias Analysis Calculation of measurement of risk: Odds Ratio (OR) (no incidence can be calculated:no RR calculation is possible) 1. Calculation of unadjusted risk 2. Calculation ‘adjusted’ risk Stratified analysis Multivariate models Elimination of confounding Demonstrating effectmodification Amsterdam Rehabilitation Research Center | Reade Advantages and disadvantages Amsterdam Rehabilitation Research Center | Reade Advantages case-control design • Relative short duration • Relatively small sample size (same precision of the risk estimation of the effect parameter as in a cohort design) Disadvantages case-control design • Validity under pressure Comparability? Might be a problem, then…. not patients and controls are studied, but exposed and non-exposed Selection-bias Information-bias Amsterdam Rehabilitation Research Center | Reade Confounding Cohort Case Control Amsterdam Rehabilitation Research Center | Reade Fletcher RH et al; Clinical epidemiology; LWW; 2014: pg 81 COHORT vs CASE-CONTROL DESIGN Cohort • • • • • • • • • • • • • • • correct, ‘natural’ sequence insight in risk actors observations at level of persons studying multiple outcomes studying multiple determinants completeness in follow-up disease(s) with low incidence-rates disease(s) with low preclinical stage, duration of incubation, latency duration sensitive for confounding sensitive for information-bias sensitive for information-bias sensitive for selection-bias size of study population cost duration Amsterdam Rehabilitation Research Center | Reade Case-control ? Summary Amsterdam Rehabilitation Research Center | Reade Summary 1 verzamelen Measuring exposition-factor(s) + other determinants Cases 3 - exposed - non-exposed identification Basic population effect? Measuring exposition-factor(s) + other determinants 2 time Amsterdam Rehabilitation Research Center | Reade Controls 3 - exposed - non-exposed selecting Questions? Amsterdam Rehabilitation Research Center | Reade
© Copyright 2026 Paperzz