Study Designs Beyond RCTs Flashcards
observational studies
researcher is documenting a natural occurring relationship
observational studies can be
descriptive or analytical
interventional studies
researcher actively performs an intervention in some or all participants
descriptive studies
describing data on one or more characteristics of a group
does not focus on reasons for occurrence
focus on novel/unusual signs, symptoms or events
analytical studies
tests a hypothesis and establishes causal relationships between two variables
prospective studies
the outcome has not occurred when the study starts
participants followed over a period of time
retrospective studies
the outcome has already occurred in each individual by the time they are enrolled in the study
data collected from records
what type of study:
investigator assigns exposure?
when they don’t?
yes - experimental
no - observational
what type of study:
if the study describes the characteristics of a sample?
yes - descriptive study
what type of study:
if the study analyzes relationship between 2 variables?
analytical study
for observational trials:
sampling determined based on outcome?
case control
for observational trials:
sampling determined based on exposure?
cohort
for observational trials:
sampling without regard to outcome/exposure?
cross sectional
which study is the first step into a new area of research?
descriptive studies
methods of collecting descriptive studies
surveys
cross sectional studies
observation
case report
case series
strengths of descriptive studies
quick and cheap
allows for analysis of facts, in depth understanding
both qualitative and quantitative methods
weaknesses of descriptive studies
cannot use statistical tools
observer effect
bias
cannot establish causative relationships
descriptive statistics
measures of central tendency
measures of dispersion
measures of frequency distribution
tables and charts
causal inference
drawing conclusion that a specific treatment was the cause of the effect
observational analytical trials only establish _______________ while interventional analytical trials only establish _________________
association
causation
case report observes ______________ while case series observes _________________
a single person
multiple patients
advantages of case report/series
detailed
identify undocumented rare/new diseases or adverse effects
disadvantages of case report/series
bias
no information on disease frequency
small number of patients
cross sectional studies
identify prevalence of exposure/outcome at a snapshot in time
retrospective and observational
survey research
cross sectional study
study the prevalence, distribution, and relationships of sociologic and psychological variables
observational analytical studies
no randomization
analyze associations between exposures and outcomes by observing subjects in real world settings
theory
a set of statements/principles devised to explain group of facts/phenomena
types of analytical observational studies
cohort
cross sectional
case control
case control studies
association between disease states and previous exposure to risk factors
appropriate use for case control studies
risk factors for rare conditions
contact trace original point in outbreak
case control studies cannot establish a
casual relationship
cohort studies
group of subjects, initially free of study outcomes, followed over time from an exposure to one or more outcomes
cohort studies determine association between
risk factors and subsequent development of disease states
cohort studies are useful in investigation of
multiple outcomes after a single exposure and in study of rare exposures
prospective cohort studies
subjects are followed prospectively to assess for development of disease
advantages of prospective cohort studies
strongest type of observational study
evaluate temporal relationship
better control of data collection
disadvantages of prospective cohort studies
cannot establish casual relationship
expensive
loss to follow up
retrospective cohort studies
association of risk factors and subsequent development of disease states
advantages of retrospective cohort studies
evaluate temporal relationship
quick and cheap
can calculate incidence and risk
real world evidence
Narrative review
summary of evidence on specific topic, qualitative
“review articles”
Systematic review
answers a defined research question
summarize data from multiple studies, qualitative
Meta-analysis
statistical methods that aggregate results of multiple studies, quantitative
Meta-analysis types
pair wise: compare intervention to control (direct)
network: multiple interventions, but one common (indirect)
Publication bias
positive results more likely to be published
Language bias
negative results less likely to be published in english
Citation bias
positive results cited more frequently
Clinical heterogeneity
variability in participants, interventions and outcomes
Methodological heterogeneity
variability in study design, outcome measurement tools and risk of bias
Statistical heterogeneity
variability in intervention effects being evaluated across different studies
Poor overlap of confidence intervals indicates _____
statistical heterogeneity
Forest Plot
visual of effect estimates and CIs for individual studies and meta-analyses
area of block = weight
PRISMA statement
evidence-based min set of items for reporting in systemic and meta-analyses
PRISMA Flow diagram
describes how investigators went from all records identified via search strategy to most relevant studies
% of variability
<25% = low inconsistency
50-75% = moderate
>75% = high inconsistency
Limitations of SLRs and meta-analyses
one number cannot summarize entire field
file drawer problem
important studies ignored
