EBDM Study Designs 101 Flashcards
Biostatistics
The use of numerical techniques to extract information from data. The discipline focuses on the variation in sets of data, that occurs either because of some intervention or from the effect of other variables linked to the data. Biostatistical tests frequently look at the comparison between an observed effect, or a difference, and the anticipated results of random variation. Some have stated that biostatistics is the study of variability or uncertainty.
Population
The group of people who meet the criteria for entry into a study (whether they actually participated in the study or not). The group of people to whom the study results can be generalized.
Sample
That part of the population selected to be studied. The group specifically included in the actual study.
Outcome (dependent variable)
The target variable of interest. The variable that is hypothesized to depend on or be caused by another variable, the independent variable
Clinical Epidemiology
- A discipline that describes quantifies and postulates causal mechanisms for health phenomena in populations.
- frequently used to investigate possible causality through non-experimental data
Causality
The relationship between cause and effect. A fundamental concept in epidemiology. A given cause may be necessary, sufficient, neither or both. A “necessary” cause must always precede an effect. A “sufficient” cause inevitably initiates an effect but is not the only necessary cause for a given effect.
Case Series
A study design that reports data on a series of consecutive patients with the same diagnosis. It is crucial that a rigorous case definition be used to insure all actually are members of the same group. There is NO control group.
Prospective Observational Study
Any study done forwards in time. Important in studies on therapy, prognosis, or harm, where retrospective studies make hidden biases more likely. When used to study potential causes of a disorder, it is a prospective investigation in which a cohort of individuals who do not have evidence of an outcome of interest but who are exposed to the putative cause are compared with a concurrent cohort who are also free of the outcome but not exposed to the putative cause. Both cohorts are then followed forward in time to compare the incidence of the outcome of interest. When used to study the effectiveness of an intervention, it is a prospective investigation in which a cohort of individuals who receive the intervention are compared with a concurrent cohort who do not receive the intervention. Both cohorts are then followed forward in time to compare the incidence of the outcome of interest.
Retrospective Observational Study
Any study in which the outcomes have already occurred before the study and collection of data has begun.
Randomized Clinical Trial
An interventional study in which the patients are randomly selected or assigned either to a group which gets the intervention or to a control group. Essentially an experiment in which individuals are randomly allocated to receive or not receive an experimental preventive, therapeutic, or diagnostic procedure and then followed to determine the effect of the intervention
Systematic Review
A formal review of a focused clinical question based on a comprehensive search strategy and structured critical appraisal of all relevant studies.
Incidence
- The rate at which an event occurs in a defined population over time. The number of new cases (or other events of interest) divided by the total population at risk.
- number of new cases of disease occurring during a specified period of time; expressed as a percentage of the number of people at risk
- this data should only come from prospective Cohort studies or RCTs
Prevalence
- The proportion of people affected with a particular disease at a specific point in time
- an accurate prevalence rate can inform a clinician’s pre-test probability estimate
Probability
Chance or frequency of a random event occurring (Frequentist). Also, the quantitative estimate of the likelihood of a condition existing (as in diagnosis) or of subsequent events (such as in an intervention study) given some assumed situation or patient scenario (Conditional/Bayesean).
Odds Ratio
A ratio of the odds of an event in an exposed group to the odds of the same event in a group that is not exposed. An odds ratio of 1 indicates that the condition or event under study is equally likely in both groups. An odds ratio greater than 1 indicates that the condition or event is more likely in the first group. And an odds ratio less than 1 indicates that the condition or event is less likely in the first group. For example, suppose that in a sample of 100 men, 90 have drunk beer in the previous week, while in a sample of 100 women only 20 have drunk beer in the same period. The odds of a man drinking beer are 90 to 10, or 9:1, while the odds of a woman drinking beer are only 20 to 80, or 1:4 = 0.25:1. Now, 9/0.25 = 36, so the odds ratio is 36, showing that men are much more likely to drink beer than women.
What is important to remember about Disease Causality?
statistical association between a risk factor and disease does NOT equate to causality
Strength of association
magnitude of the measure of association
consistency upon repitition
different studies find consistent results
specificity
specific risk factor associated with specific diseases
time sequence
shorter the duration of time the stronger the evidence
dose response
increase in the risk factor dose is associated with an increase in rate of outcome
biological plausability
basic science offers a sound explanation
coherence of explanation
explanation isn’t contradictory to what is already known about the disease or risk factor
experimental data
animal experiments or “natural” experiments are consistent
e.g., outcomes already known in populations with risk factor exposures
What are five types of study designs?
- descriptive
- observational - retrospective
- observational - prospective
- randomized controlled/clinical trial (RCT)
- systematic review (SR)
Descriptive Study Design:
case report
describes a single individual case
Descriptive Study Design:
case series
- a series (2+) individuals with the same or similar disease
- no control group
- most are descriptions of newly discovered or rare diseases or treatments
Descriptive Study Design:
cross-sectional study
- at one point in time
- individuals with same or similar disease form a certain population are studied (such as their potential risk exposures, treatment side-effects, etc.)
- capable of estimating prevalence if done properly
- not capable of calculating incidence
Observational - Retrospective Study Design:
case-control
- used to calculate the ODDS of exposure to a risk factor in individuals with and without a disease
- subjects chosen because they have (cases) or do not have (controls) a certain disease
- relatively inexpensive and quick to perform
- subject to selection bias, recall bias, and reporting bias
- cannot, and should not, be used to estimate prevalence or calculate incidence
Case-control studies are subject to what bias (3)?
- selection bias
- recall bias
- reporting bias
Well done case-control studies have what three characteristics?
- controls are representative of the target population
- cases are representative of all cases of the disease
- teh disease is rare (allows OR to estimate RR)
Observational - Retrospective Study Design:
Outcome and effectiveness registries
existing databases are retrospectively studied searching for associations between specific outcomes and risk exposures or therapies
ex: data on individuals entered for non-research reasons then ‘mined’ for the secondary research purposes
Observational - Prospective Study Design:
cohort
- patients chosen for study if they are documented to NOT HAVE the disease outcome in question
- subjects then carefully screened on the basis of exposure to a risk factor under investigation or no exposure
- the disease outcome (onset of disease) is then determined over time
- incidence of disease (and other outcomes evaluated) can be accurately calculated
- Relative Risk can also be calculated
- this is the gold-standard study for estimating prognosis and measuring the association between risk factors and disease outcomes
- very expensive, very time-consuming
Randomized Controlled/Clinical Trial (RCT)
- gold-standard study design for assessing positive effects of therapy interventions
- a true experiment that is best when it includes a hypothesis, inclusion and exclusion criteria for subjects, randomization of subject to intervention or control groups, blinding to decrease chance of bias, clear methods of administering the intervention and “intention-to-treat” analysis of results
- informed consent is a crucial ethical step; must be part of every RCT
- conflicts of interest by investigators is also a concern
Systematic Review (SR)
- strongest study design for answering a clinical question is an SR of multiple RCTs
- SR is highest quality when a strict methodology for searching and selecting RCTs is followed to insure that only well-done RCTs are included
- complex data pooling and analysis are usually done in a process called “meta-analysis”
- advantages: increased sample size compared to single RCT, analysis of possible heterogeneity of single study results, careful review of individual studies and a summary conclusion that frequently answers the clinical question at this point in time
What are the advantages of an SR?
- increased sample size compared to single RCT
- analysis of possible heterogeneity of single study results
- careful review of individual studies
- a summary conclusion that frequently answers the clinical question at this point in time
What are three common epidemiological estimates?
- incidence
- prevalence
- mortality rate
mortality rate
- the incidence or probability of death
- the number of people who die (during a specific period of time) divided by the number of people at risk of death
Risk
- risk = probability = quantitative estimate of an occurrence
- Frequentist theory: chance or frequency of a random event occurring
- Bayesean: conditional probability that depends on certain data; a pre-test and post-test probability exist that depend on whether the data or test result is known
Frequentist theory (risk)
chance or frequency of a random event occurring
Bayesean (risk)
conditional probability that depends on certain data; a pre-test and post-test probability exist that depend on whether the data or test result is known
absolute risk
probability that an individual with a risk factor exposure has the outcome
a/(a+b)
Relative Risk
- ratio of risk of an outcome among individuals exposed to the risk factor being considered to the risk of the outcome in people NOT exposed
- a measure of the strength of the associatin between risk factor and outcome
- data should come from prospective Cohort studies or RCTs
[a/(a+b)] / [c/c+d)]
Odds Ratio
- ratio of the ODDS of an outcome among individuals exposed to the ODDS of the outcome in people NOT exposed to the risk factor
- used as an estimate of RR in case-control studies
- this data usually comes from Case Control studies. Incidence CANNOT therefore be calculated (the subjects are selected because they already have the disease).
(a/b)/(c/d) = ad/bc
RR vs OR
- relative risk is a BETTER estimate of risk
- the odds ratio is an ESTIMATE of the relative risk and does this reasonably well when a disease is very rare. In those situations, RR is approximately equal to OR.
Attributable Risk
estimate of the amount of risk of an outcome in individuals exposed to the risk factor is actually due to, or attributable to, the risk factor; Two major ways to express this (AAR and ARI)
Absolute Attributable Risk (AAR)
- simple difference between absolute risks with and without the risk factor
- also known as the Absolute Risk Reduction because this is the estimated amount of disease burden that could be reduced if we could just control for this risk factor
- Number needed to Harm (NNH) is the number of individuals that need to be exposed to the risk factor to result in ONE disease outcome.
AAR = a/(a+b) - c/(c+d)
NNH = 1/AAR
Relative Attributable Risk or Attributable Risk Percent (ARP)
the attributable risk relative to the risk in non-exposed individuals.
ARP = [a/(a+b) - c/(c+d)] / [c/(c + d)]
PICO
- Patients
- Intervention
- Control
- Outcome
- structured clinical question
PICO: Patients
- refers to the population group to which you want to apply the information
- cannot be too specific or you will have trouble finding evidence in the literature.
- Good choices: elderly women, diabetic children.
- Poor choices: middle-aged Hispanic male with chronic hypertension, young-adult Asian female with pulmonary fibrosis
PICO: Intervention
therapy, harmful exposure or diagnostic test you want to find evidence for
PICO: Control
- the alternative to the therapy, harmful exposure or diagnostic test that you are interested in finding evidence for
- commonly this would be placebo or the standard approach to therapy or diagnostic testing
PICO: Outcome
- the meaningful endpoint of interest to you or your patient
- beware of “surrogate markers” or potentially non-important pseudo-makers of disease
The Two-by-Two Table
- a common and functional way to compare two dichotomous variables - such as the status (presence or absence) of a risk factor and the status of a disease outcome