Epidemiology study designs

Description

Mind Map on Epidemiology study designs, created by jillian.kowalchuk on 01/05/2014.
jillian.kowalchuk
Mind Map by jillian.kowalchuk, updated more than 1 year ago
jillian.kowalchuk
Created by jillian.kowalchuk over 10 years ago
365
2

Resource summary

Epidemiology study designs
  1. Ecological
    1. Features
      1. Advantages
        1. If data is difficult to measure at an individual level
          1. To study group-sepcific events
            1. When data are readily available
              1. Quick
                1. Less costly
              2. Disadvantages
                1. Cannot infer causality
                  1. Confounders likely
                  2. Population-based
                    1. Prevalence or incident cases
                      1. Data Collection
                        1. Routine data sources
                          1. Vital registration data
                            1. Demographic data
                              1. Disease registries
                            2. Analysis
                              1. Standardization
                                1. Direct
                                  1. Standardized Rate
                                  2. Indirect
                                    1. SMR
                                2. Interpretation
                                  1. Ecological fallacy
                                    1. Bias
                                      1. Mixing/Migration
                                        1. Confounding
                                        2. Study Design
                                          1. Time-trend design
                                            1. Mixed studies
                                              1. Multi-group study
                                          2. Cross-sectional
                                            1. Features
                                              1. Advantages
                                                1. Rapid
                                                  1. Easy to collect/analyse
                                                    1. Relatively cheap
                                                      1. Can help policy to see how many ppl need treatment
                                                        1. May be possible to assume causality if exposure constant (i.e. genetics)
                                                          1. May be only plausible study design to answer question
                                                          2. Disadvantages
                                                            1. Did exposure precede outcome (information bias)?
                                                              1. Medium probability of selection bias
                                                                1. Medium confounding
                                                                  1. Causality not possible in exposures that change (i.e. diet)
                                                                    1. Can't see time relationship or incidence rate
                                                                    2. Study Design
                                                                      1. Descriptive
                                                                        1. Estimate the frequency of outcomes or exposure without the intention of investigating causal associations.
                                                                          1. Individual level data on outcome, exposure or both
                                                                          2. Secular
                                                                            1. Seasonal
                                                                            2. Analytical
                                                                              1. Collects data on both outcome and exposure at the same time-point for a given study subject. Good for testing hypothesize of association where exposure status does not change over time.
                                                                            3. Sampling
                                                                              1. Questionnaires
                                                                                1. Pilot study
                                                                              2. Data collection
                                                                                1. Period Prevalence
                                                                                  1. Point Prevalence
                                                                                  2. Interpretation
                                                                                    1. Reverse causality
                                                                                    2. Analysis
                                                                                      1. **Prevalence Ratio
                                                                                        1. Outcome rare= odds ratio
                                                                                      2. Cohort Studies
                                                                                        1. Features
                                                                                          1. Advantages
                                                                                            1. Low selection bias
                                                                                              1. Low recall bias
                                                                                                1. Low confounding
                                                                                                2. Disadvantages
                                                                                                  1. Loss to follow up
                                                                                                    1. Logistically difficult
                                                                                                      1. Expensive
                                                                                                      2. Study Design
                                                                                                        1. Descriptive
                                                                                                          1. Observational
                                                                                                          2. Analytical
                                                                                                            1. Prospective
                                                                                                              1. Retrospective
                                                                                                              2. Sampling
                                                                                                                1. Occupational cohort
                                                                                                                2. Analysis
                                                                                                                  1. Descriptive
                                                                                                                    1. Risk or rate
                                                                                                                    2. Analytical
                                                                                                                      1. Risk, rate, attributable risk, attributable fraction, population attributable risk, population attributable fraction
                                                                                                                      2. Time-series analysis
                                                                                                                      3. Data collection
                                                                                                                        1. Retrospective= routine sources
                                                                                                                          1. Prospective = exposures that aren't going to change
                                                                                                                        2. Interpretation
                                                                                                                          1. Confounding/bias
                                                                                                                            1. Loss to follow up
                                                                                                                              1. Observer bias
                                                                                                                              2. Screening method
                                                                                                                                1. Lead-time bias
                                                                                                                                  1. Length-time bias
                                                                                                                              3. Case Control
                                                                                                                                1. Features
                                                                                                                                  1. Sampling
                                                                                                                                    1. Case definition
                                                                                                                                      1. Inclusion/exclusion criteria
                                                                                                                                    2. Data Collection
                                                                                                                                      1. Almost always retrospective
                                                                                                                                        1. Reporting bias
                                                                                                                                        2. Advantages
                                                                                                                                          1. Low loss-to follow up
                                                                                                                                            1. Rapid
                                                                                                                                            2. Disadvantages
                                                                                                                                              1. Selection bias
                                                                                                                                                1. Information bias
                                                                                                                                                  1. Confounding
                                                                                                                                                  2. Study Design
                                                                                                                                                    1. Matching
                                                                                                                                                      1. Nested case-control
                                                                                                                                                        1. Overmatching
                                                                                                                                                          1. Individual matching
                                                                                                                                                            1. Frequency matching
                                                                                                                                                              1. Control
                                                                                                                                                          2. Analysis
                                                                                                                                                            1. Odds ratio of exposure
                                                                                                                                                              1. Outcome rare measure approximates risk ratio
                                                                                                                                                              2. Interpretation
                                                                                                                                                                1. Watch for discussed biases and confounders
                                                                                                                                                                  1. Reverse causality
                                                                                                                                                                  2. Screening method
                                                                                                                                                                    1. Selection bias
                                                                                                                                                                      1. Information bias
                                                                                                                                                                  3. Intervention
                                                                                                                                                                    1. Advantages
                                                                                                                                                                      1. Low selection bias
                                                                                                                                                                        1. Low information bias
                                                                                                                                                                          1. Low Confounding
                                                                                                                                                                            1. Infer causality
                                                                                                                                                                            2. Disadvantages
                                                                                                                                                                              1. Loss to follow up
                                                                                                                                                                                1. More complex
                                                                                                                                                                                  1. Expensive
                                                                                                                                                                                  2. Features
                                                                                                                                                                                    1. Study Design
                                                                                                                                                                                      1. Ethical approval
                                                                                                                                                                                        1. Declaration of Helsinki
                                                                                                                                                                                        2. Efficacy
                                                                                                                                                                                          1. Effectiveness
                                                                                                                                                                                            1. Plausability studies
                                                                                                                                                                                              1. Historical controls
                                                                                                                                                                                                1. Geographical controls
                                                                                                                                                                                                  1. Opportunistic controls
                                                                                                                                                                                                2. RCT
                                                                                                                                                                                                  1. Cluster-randomized trials
                                                                                                                                                                                                    1. Factorial design
                                                                                                                                                                                                      1. Crossover design
                                                                                                                                                                                                      2. Non-randomized trial
                                                                                                                                                                                                        1. selection bias
                                                                                                                                                                                                      3. Sampling
                                                                                                                                                                                                        1. Representative population
                                                                                                                                                                                                          1. Choice of Control
                                                                                                                                                                                                            1. Allocation
                                                                                                                                                                                                              1. Simple randomization
                                                                                                                                                                                                                1. Systematic randomization
                                                                                                                                                                                                                  1. Blocked Randomization
                                                                                                                                                                                                                    1. Stratified randomization
                                                                                                                                                                                                                      1. Matched-pair randomization
                                                                                                                                                                                                                        1. Envelopes (allocation concealment)
                                                                                                                                                                                                                          1. Minimization
                                                                                                                                                                                                                          2. Blinding
                                                                                                                                                                                                                            1. Single
                                                                                                                                                                                                                              1. Double blinded
                                                                                                                                                                                                                                1. Gold standard: Double-blinded randomized placebo-controled trial
                                                                                                                                                                                                                                2. ascertainment bias
                                                                                                                                                                                                                              2. Data Collection
                                                                                                                                                                                                                                1. Same as Cohort and Cross-sectional
                                                                                                                                                                                                                                2. Analysis
                                                                                                                                                                                                                                  1. Intervention allocation
                                                                                                                                                                                                                                    1. Interim analysis
                                                                                                                                                                                                                                      1. Intention-to-treat v.s per-protocol
                                                                                                                                                                                                                                        1. Intervention efficacy
                                                                                                                                                                                                                                        2. Interpretation
                                                                                                                                                                                                                                          1. Bias and confounders
                                                                                                                                                                                                                                          2. Screening methods
                                                                                                                                                                                                                                            1. Information bias
                                                                                                                                                                                                                                              1. Misclassification bias
                                                                                                                                                                                                                                          Show full summary Hide full summary

                                                                                                                                                                                                                                          Similar

                                                                                                                                                                                                                                          Subh Milis le Seamus O Neill
                                                                                                                                                                                                                                          l.watters97
                                                                                                                                                                                                                                          Computer Hardware
                                                                                                                                                                                                                                          cscutt
                                                                                                                                                                                                                                          Bayonet Charge flashcards
                                                                                                                                                                                                                                          katiehumphrey
                                                                                                                                                                                                                                          An Inspector calls Techniques
                                                                                                                                                                                                                                          anya14
                                                                                                                                                                                                                                          PE AQA GCSE REVISION FLASHCARDS
                                                                                                                                                                                                                                          ellie.baumber
                                                                                                                                                                                                                                          Aparatos y sistemas del cuerpo humano
                                                                                                                                                                                                                                          Mai Sin Más
                                                                                                                                                                                                                                          Hitlers Germany
                                                                                                                                                                                                                                          gursharonkaur15
                                                                                                                                                                                                                                          GCSE Biology B2 (OCR)
                                                                                                                                                                                                                                          Usman Rauf
                                                                                                                                                                                                                                          AQA GCSE Physics Unit 3 Mindmap
                                                                                                                                                                                                                                          Gabi Germain
                                                                                                                                                                                                                                          AQA A2 Biology - Chapter 8 Inheritance and Selection
                                                                                                                                                                                                                                          Charlotte Lloyd
                                                                                                                                                                                                                                          Frankenstein Key Quotes
                                                                                                                                                                                                                                          michelle.lau