Studied by 6 people
primary research approach
collect and analyze new data
secondary research approach
analyze existing data
tertiary research approach
review and synthesize existing literature
research question
A statement including the specific exposure, disease, or outcome, and the population which research is focused on
study objectives
takes a form of a measurable question or a 'to' statement. Represents a step towards answering the main research question
lead researcher
first author, does the majority of the work
senior researcher
last author, most experienced and guides the work of others
Coauthors
most papers have between four to dozens
International Committee of Medical Journal Editors (ICMJE): Authorship Criteria
Substantial contributions to conception and design and/or acquisition of data and/or analysis and interpretation of data
Drafting the article and/or revising it critically for important intellectual content
Final approval of the version to be published
correlational studies
uses population level (group) data to examine the relationship between exposure rates and disease rates
ecological fallacy
assuming from a population that there is a high level of something, everyone will experience that same thing
correlational analysis
the use of statistical correlation to evaluate the strength of the relations between variables
correlational analysis for continuous variables
Pearson correlation co-efficient (r)
correlational analysis for rank/categorial variables
Spearman rank-order correlation
linear regression
used when comparing more than two variables with a goal of understanding the relationship between the two
case series
describes two or more patients who have the same condition/have undergone the same procedures. identifies new symptoms, refines diagnosis, develops hypotheses for future research etc.
atomistic fallacy
Wrongly assuming from observations causing the disease in individuals applies to all populations. prone to the opposite of the ecological fallacy
cross sectional studies
an observational study where exposure and disease are determined simultaneously in a given population. A temporal relationship between exposure and disease can not be determined
case controlled studies
used when the disease is uncommon and the general population is unlikely to yield enough cases. compares two preexisting groups to identify possible etiologies
case control analysis
used to find the proportion of individuals exposed in the cases (a/a+c) and controls (b/b+d)
odds ratio
ratio of odds of exposure in cases to that of controls
OR = 1
odds of exposure are the same for cases and controls
OR > 1
cases have higher odds of exposure than controls, exposure is risky to health
OR < 1
cases have lower odds of exposure than controls, exposure is protective to health
OR statistically significant
95% CI is greater than 1, and the exposure is deemed risky 95% CI is less than one, and the exposure is deemed protective
OR not statistically significant
95% confidence interval overlaps OR = 1 exposure and disease have no association
Matching
the process of selecting controls in a case-control study so that controls are similar to cases
no matching
assumes similar inclusion and exclusion criteria will result in case and control populations having similar distributions
frequency matching
control population is recruited similar to the case population, individual cases are not tied to individual controls
matched pairs
each case is personally linked to an individual control for all characteristics that are not disease specific
Limiting misclassification bias
all participants meet inclusion and exclusion criteria.
adherence to strict definitions of cases and controls
recall bias
information on past exposures depends on the memory of events from all people (oftentimes inadequate and limited)
occurs when the memory is better among cases than controls due to the disease
cohort
a group of individuals with similar characteristics followed along a timeline
cohort study
follows participants along a timeline to calculate the rate where a new disease occurs, identifies disease risk factors
baseline
exposure status of the participants
follow-up assessments
determine the number of participants who have developed a new disease since the initial examination
rate ratio
the incidence of disease for exposed (a/a+b) and control (c/c+d)
RR = 1
incidence rate was the same in exposed and not exposed groups, exposure is not associated with disease
RR > 1
the incidence rate was higher in the exposed group than non-exposed, exposure is risky
RR < 1
the incidence rate was lower in the exposed than non-exposed, exposure is protective
CI overlaps RR = 1
RR is not significant, no association
rate ratio CI > 1
RR is statistically significant, exposure is a risk factor
rate ratio CI < 1
RR is statistically significant, exposure is protective
Prospective (longitudinal) cohort studies
This study follows a population over time to track the development of the disease. Begins with the identification of the population and exposure status. It takes a long time to complete
retrospective cohort studies
existing data from the past is used to identify the population and exposure status. Determines the present status of the disease, no future follow-up is needed. Relatively quick
Combined prospective and retrospective cohort study
Past data is used to identify the population and exposure status. Quick to assemble using the study populations past data and follows the population into the future for development of disease.
Framingham Heart Study
one of the longest research studies. Began in 1948 using participants from 30-62 years old. For each possible exposure they made a hypothesis and identified the exposed and not exposed subjects were followed every two years to see the development of disease
Selection bias
selecting participants into exposed and non-exposed based on characteristics that could affect the outcome
information bias
collecting different quality and quantity of information for the exposed and non exposed
misclassification bias
misclassify exposure status and diseases status
observational study
investigators use the data observed in the population to make inferences on the relationship between the variables
experimental study
investigators intervene in the natural history by actively altering one variable and making inferences on the relationship between the variables based on the outcome
comparison groups in an experimental study
therapy/no therapy, therapy/placebo, therapy A/therapy B
randomization
the process where allocation of patients into treatment groups is done by chance and without the ability to predict who is in what group
clinical trial
a controlled experiment that has a clinical event as well as an outcome measure. Done in a clinical setting with subjects with a specific disease or health condition
randomized clinical trial
a clinical trial where patients are randomly assigned to specific groups comparing different treatments
stratified randomization
the random assignment within group defined by the participants' characteristics like age or disease severity. Ensures a good balance of factors across intervention groups
primary purpose of randomization
prevents bias within allocating participants to treatment groups
secondary purpose of randomization
achieve compatibility between groups
single blind study
subjects are unaware of their exposure status
double blind study
subjects and researchers are unaware of who is in the research and control groups
Hawthorne bias
subjects may change their behaviour because they know what is being observed. Can interfere with the accurate measurement of the impact of interventions
Parallel treatment
participants are randomized into A and B groups. Group A is the variable and B is control, the outcomes are then followed
Planned crossover treatment
participants are randomized into A and B groups. Group A is the variable, and B is the control. Halfway through the study, both groups "washout" and switch, A becomes the control, and B becomes the variable.
Unplanned crossover treatment
participants are randomized into groups A and B, some participants in A may refuse treatment and switch into B. Some from B require different care and are switched into A
treatment received approach
limits analysis to participants who were fully compliant with their assigned intervention (ideal world scenario)
treatment assigned approach
includes all participants even if they were not fully compliant with their assigned intervention (real world)
efficiency
The proportion of subjects in the control group who experienced an unfavourable outcome that could have had favourable outcome if they were in the treatment group instead. High efficacy is an indicator that an intervention is successful.
number needed to treat
the expected number of subjects needed to treat to achieve a favourable outcome in one person. Smaller number indicates a more effective intervention.
efficiency = (rc - ri)/rc
efficiency equation
NNT = 1/(rc - ri)
Number needed to treat equation
equipoise
experimental research should be conducted. only when there is uncertainty about which treatment will work better
distributive justice
infers the source population must be appropriate and non-exploitative
beneficence (do good) and non-maleficence (do no harm)
research must balance benefits and risks of the study
respect for participants
subjects must volunteer without being coerced or unduly influenced by their compensation
cross sectional analysis: prevalence of exposure
a/(a+c) vs. b/(b+d)
cross sectional analysis: prevalence of disease
a/(a+b) vs c/(c+d)
exposure
a personal characteristic, behaviour, environmental encounter, or intervention that may change the likelihood of developing a health condition
risk factor
an exposure increasing an individual's likelihood experiencing a particular disease or outcome
protective factor
an exposure reducing an individuals likelihood of experiencing a disease or outcome
non modifiable risk factor
a risk factor for a disease that cannot be changed through health interventions
modifiable risk factor
a risk factor for a disease that can be avoided or mitigated
feasibility study
an evaluation of the likelihood that a task can be completed with resources available
external funding
funding from private foundations, corporations, government agencies, etc.
internal funding
funding through universities or employers
research proposal
usually a request for funding or supervisory support
principal investigator
the primary investigator accepting responsibility for the study
research protocol
a detailed handbook describing all actions that will be taken during the study implementation
Note
Flashcard