Study Design graph types and Kaplan-Meier curves Flashcards
What type of study is a follow-up in a prospective manner where exposure is determined at the outset?
Cohort study
Name 2 types of DESCRIPTIVE STUDIES
1) Case reports/ Case series
2) Cross-sectional (prevalence studies) including surveys and census
Name 3 types of ANALYTICAL STUDIES
1) Case control studies
2) Cohort studies
3) Ecological studies
Name 2 main categories of EXPERIMENTAL STUDIES (Not study designs)
1) controlled studies
2) uncontrolled studies
When describing a research study, what are the two aspects of design related to recording of exposure and outcome?
1) identification (GROUPING) of subjects i.e. exposed vs unexposed, or cases vs controls
2) the TIME-COURSE i.e. prospective, retrospective or cross-sectional (exposure and outcome measured together)
What is a prospective study vs retrospective?
In a prospective study, a group of subjects is assembled who have been exposed to a risk/ intervention, and the researchers await an outcome (exposure-outcome) i.e it goes FORWARD in time.
In a retrospective study, the exposure has already happened so it is outcome-exposure) e.g case control study
What category of study has no testing of causal hypotheses or comparisons to other groups?
Descriptive studies
What type of study can SUGGEST a punitive relationship between variables?
“How common is a condition? What is the nature of this sample?”
Cross-sectional studies
This is a study that examines the relationship between diseases (or other health-related characteristics) and other variables of interest as they exist in a defined population at one particular time (ie exposure and outcomes are both measured at the same time). Best for quantifying the prevalence of a disease or risk factor, and for quantifying the accuracy of a diagnostic test.
What type of study:
Both exposure and disease have occurred before the onset of the study. It is often the first mode of studying a suspected association of causality. It is mostly useful in diseases where exposure is common, but the disease occurs only in a small proportion of those exposed. Defining a case is crucial – strict diagnostic criteria and eligibility criteria must be followed
Case Control
In a case control study:
How are controls selected?
Controls by definition must be identical to
cases except for the presence or absence of disease and recruited independently of exposure status.
Up to a ratio of 4:1, when the number of controls is increased, the power of a study increases. But not after exceeding this ratio.
To ensure comparability the cases and controls are often matched – i.e. certain selected variables such as age, sex, etc. are matched
when controls and patients are recruited. Matching can be done in groupwise or pairwise fashion, the latter being more time
consuming.
Overzealous matching may not only make recruitment of controls difficult, but can also reduce the risk difference between the two
groups. Overzealous matching may inadvertently result in matching for putative causal factors as well.
What is the odds ratio, and how is this used in case control studies?
This is the ratio of the odds of the event happening in an exposed group versus a non-exposed group.
To estimate the risk of exposure in case-control studies, a cross-product ratio called odds ratio is used. This is because relative risk
cannot be calculated as incidence rates are not available (no longitudinal observation to see new cases developing as in cohort
studies
In what type of study are observations repeated in the same population over a prolonged period of time?
Prospective longitudinal studies
Note: costly and time consuming but good for studying the course of an illness, incidence rates and risk factors
What category of studies involve a comparative analysis of 2 groups to measure the association between variables of interest?
Analytical studies
What type of study involves a group of people with specific characteristics/ exposure who are followed up over a period of time to detect an outcome?
“What are the effects of this risk factor/ exposure?”
Cohort study
What is a case-control study?
When there are groups of patients including those who have experienced an outcome (cases) and those who haven’t (controls) and then the researcher hypothesises about the causal relationship between the outcome, and an exposure of interest.
Patients with a certain outcome or disease and an appropriate group of controls without the outcome or disease are selected (usually with careful consideration of appropriate choice of controls, matching, etc) and then information is collected retrospectively on whether the subjects have been exposed to the factor under investigation.
Note: a variation of this is the nested case-control study which is the same but draws cases and controls from an existing cohort.
1) What is an ecological study?
2) What is ecological fallacy?
1) A specific type of analytical study where data describes what happens to a group rather than an individual.
Ecological studies consider where and when diseases are occurring in populations. The presence / absence of specific risk factors is not related to individuals with or without disease, but rather to populations. Ecological studies are often considered to be weak study types as they cannot prove anything, however they are used to raise questions for further research.
2) falsely ascribing the characteristics of a group to all individuals of that group
What is an experimental study?
Experimental studies are generally prospective studies where the exposure is experimentally assigned to the groups in a variable fashion to observe the effect. Unlike the cohort studies, here
deliberate manipulation of exposure (intervention) takes place.
What are 4 subtypes of EXPERIMENTAL studies i.e where an intervention is used:
Randomised trial
Non-randomised trial
Case study*
Case series*
- can be both experimental and observational
What are 5 types of OBSERVATIONAL studies i.e no intervention, observation only:
Cohort
Case-control*
Cross-sectional
Ecological (Correlation) study
Case series*
- can be both experimental and observational
What is the best study design for the following clinical enquiries?
1) Treatment effectiveness
2) Treatment efficacy
3) Causation (aetiology)
1) Pragmatic RCT (or systematic reviews)
2) Experimental RCT
3) Cohort or case control
What is the best study design for the following clinical enquiries?
4) Prognosis
5) Diagnostic assessment
(evaluating a new tool)
4) cohort
5) Cross-sectional comparison to gold standard
What is the best study design for the following clinical enquiries?
6) health economics
7) ‘meaning’ or health ‘experience.’
6) Cost-effectiveness study
7) Qualitative Studies
What are the advantages and disadvantages of a Case Control study?
7x advantages
5 x disadvantages
♣ Advantages:
o Easy to carry out
o Less time-consuming
o Less expensive
o Suitable for investigating rare diseases
o Subjects are not exposed to any new risks
o Several different etiological factors for
single disease can be studied
o No attrition problems
♣ Disadvantages:
o Highly prone to selection and recall bias
o Control group selection may be difficult
o Incidence cannot be measured - so odds
ratio only – not relative risk – can be measured
o Cannot prove causality
o Temporality is difficult to determine
retrospectively
For the following types of questions, what would the best study type(s) be?
1) Therapy
2) Diagnosis
1) RCT, cohort, case control. case series
2) Cohort (PREFERRED!) with comparison to gold standard test (cross sectional studies may be used when assessing the accuracy of a diagnostic test as they measure exposure and outcome simultaneously)
For the following types of questions, what would the best study type(s) be?
3) Prognosis
4) Etiology/ Harm
3) Cohort studies, case control, case series
4) RCT, cohort, case control, case series
For the following types of questions, what would the best study type(s) be?
5) Prevention
6) Cost
5) RCT, cohort, case control, case series
6) Economic Analysis
Cohort studies can be RETROSPECTIVE or PROSPECTIVE.
What do each of these look like?
What are 2 advantages of a retrospective study?
What is a relative advantage of a prospective study?
In a prospective cohort study a study population WITHOUT disease is separated into cohorts which differ by exposure to a risk factor. These cohorts are then followed up to see if disease / an outcome occurs. This enables calculation of the relative risk.
Risk factors can be binary (e.g. diabetic yes / no) or continuous (e.g. serum phosphate level).
A retrospective cohort study is quicker to perform and less costly than a prospective one. In this type of cohort study, the data are collected from records. Thus, the outcomes have occurred in the past. Even though the outcomes have occurred in the past, the basic study design is essentially the same. Thus, the investigator starts with the exposure and other variables at baseline and at follow-up and then measures the outcome during the follow-up period.
In a prospective study, all the assessments of the exposures are done before the study begins. This gives the researchers more control over the risk factors.
The following advantages and disadvantages refer to which type of study?
Advantages:
- Unbiased distribution of confounders
- Blinding more likely
- Randomisation facilitates statistical analysis
Disadvantages:
- Expensive
- Time consuming
- Volunteer bias
- Ethically problematic at times
RCT
The following advantages and disadvantages refer to which type of study?
Advantages:
- Ethically safe
- Subjects can be matched
- Can establish timing and directionality of events
- Eligibility criteria and outcome assessments can be standardised
- Administratively easier and cheaper than RCT
Disadvantages:
- Controls may be difficult to identify
- Exposure may be linked to a hidden confounder
- Blinding is difficult
- Randomisation not present
- For rare disease, large sample sizes or long follow-up necessary
Cohort Study
The following advantages and disadvantages refer to which type of study?
Advantages:
- Quick and cheap
- Only feasible method for very rare disorders or those with long lag between exposure and outcome
- Fewer subjects needed than cross-sectional studies
Disadvantages:
- Reliance on recall or records to determine exposure status
- Confounders
- Selection of control groups is difficult
- Potential bias: recall, selection
Case-Control Study
The following advantages and disadvantages refer to which type of study?
Advantages:
- Cheap and simple
- Ethically safe
Disadvantages:
- Establishes association at most, not causality
- Recall bias susceptibility
- Confounders may be unequally distributed
- Neyman bias
- Group sizes may be unequal
Cross-sectional Survey
The following advantages and disadvantages refer to which type of study?
Advantages:
- Cheap and simple
- Ethically safe
Disadvantages:
- Ecological fallacy (when relationships which exist for groups are assumed to also be true for individuals)
Ecological Study
What is a Kaplan-Meier curve also known as?
Survival curve
What are the axes on a Kaplan-Meier curve?
Survival curves show, for each time plotted on the X axis, the portion of all individuals surviving as of that time.
What type of graph?
Graphical representation of the sample minimum, lower quartile, median, upper quartile and sample maximum
Box and Whisker
What type of graph?
Used to demonstrate the existence of publication bias in meta-analyses
Funnel plot
What type of graph?
A graphical display of continuous data that shows data by way of bars to display the frequency of numerical data
Histogram
What type of graph?
A graphical display used to compare the frequency, total count, sum, or an average of data in different categories by using horizontal or vertical bars. It is also known as a column chart
Bar chart
What type of graph?
Usually found in meta-analyses and provide a graphical representation of the strength of evidence of the constituent trials
Forest plot
What type of graph?
Graphical representation using Cartesian coordinates to display values for two variables for a set of data
Scatter Plot
What type of graph?
Estimate of the survival function showing decreasing survival with time
Kaplan-Meier
