Flashcards
Study designs
Cards in this deck (15)
- 1What is a case control study?A retrospective observational study design that enrolls cases (e.g. men aged 18-45) with a certain outcome (e.g. diseased) and those without the outcome (control group).
It is important that the control group has similar profile to the group of interest (e.g. also men aged 18-45) - 2What is a prospective cohort study?A longitudinal study design that recruits cases without the outcome (e.g. lung cancer). The cases may have different levels of exposure (e.g. smokers and non-smokers) to a suspected correlate (e.g. smoking).
They are then followed over time to identify incidence differences between the groups. Any other exposures can be identified and compared in real time.
As the sequence of events is available, cohort studies can infer causality. - 3What is a retrospective cohort study?A longitudinal study design that recruits cases with an exposure that have been identified as a risk factor for the development of an outcome. Their histories are checked and the outcomes are compared to a control group.
As the sequence of events is available, cohort studies can infer causality. - 4What is a cross sectional survey?An observational study design that takes a "snapshot" of the current state to estimate a statistic at that specific point in time (e.g. prevalence of migraines on 22nd Nov 2019)
- 5What are Randomised Control Trials (RCTs)?The highest quality clinical trials with multiple layers of strict methodology to minimise the effects of any confounding variables or bias. Involves "blinding" the elements (e.g. patients), the providers (e.g. nurses) and the researchers.
- 6What is a case study?A descriptive account of a specific occurrence or outcome (e.g. Patient A suffering from cysticercosis)
- 7What are systematic reviews?Highly organised reviews of the highest quality evidence available. Collates an exhaustive collection of all available information on a research topic, systematically appraises them and selects the highest quality studies for analysis.
- 8What are the pros of case control studies?They are ideal for rare outcomes or those with long latency periods between exposure and outcome.
They are less resource intensive.
They are better suited to high migration populations where follow up is difficult. - 9What are the cons of case control studies?They are prone to selection bias (not representative of the target population).
They require accurate categorisation into outcome groups.
Possible exposures must be identifiable.
They are prone to recall bias.
They are not ideal for rare exposures.
They generally are not able to calculate incidence. - 10What are the pros of prospective cohort studies?There are no ethical issues regarding which group gets treatment.
Incidence and prevalence can be calculated.
As the sequence of events is available, cohort studies can infer causality.
Multiple diseases and outcomes can be studied at the same time.
Exposures can be identified in real time.
There is no recall bias. - 11What are the cons of prospective cohort studies?They are resource intensive and often involve very high sample sizes.
Attrition (drop out) can be high.
Attrition can introduce bias (e.g. certain types of individuals may be more likely to drop out)
Follow up can be difficult (i.e. migration and death)
Cases may have to be reclassified during the study period. - 12What are the pros of retrospective cohort studies?There are fewer ethical issues as only the histories of cases are studied.
They can identify exposures.
As the sequence of events is available, cohort studies can infer causality.
Incidence can be measured.
Attrition is not a factor. - 13What are the cons of retrospective cohort studies?They can be resource intensive.
Prone to recall bias.
Requires understanding of possible correlates.
Requires accurate record keeping. - 14What are the pros of cross sectional surveys?Less resource intensive.
Can collect a large variety of variables (e.g. prevalence).
Facilitates further and more accurate study designs. - 15What are the cons of cross sectional surveys?Sample sizes may be very large.
Does not have a time aspect so rates (e.g. incidence) cannot be calculated.
Causation cannot be inferred, only correlation.
Prone to selection bias.
Information
- Description
- Public Health Flashcards on Study designs, created by Ralfh Pulmones on 09/11/2019.
- Created by
- Ralfh Pulmones
- Views
- 45
- Rating