Home
Subjects
Explanations
Create
Study sets, textbooks, questions
Log in
Sign up
Upgrade to remove ads
Only $2.99/month
Terms
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (92)
A priori:
Defined before the study starts.
Alpha Error (type 1 error):
Falsely determining that a statistically significant difference between two study variables exists when, in reality, no difference exists (i.e. false positive).
Absolute Risk Difference (ARD):
The absolute difference between the rates of outcomes between two groups.
Absolute Risk Reduction (ARR):
The absolute difference between the rates of outcomes between two groups when the difference is a risk reduction. Also called Risk Difference. It is the inverse of Number Needed to Treat.
Absolute Risk Increase (ARI):
The absolute difference between the rates of outcomes between two groups when the difference is a risk increase.
Baseline Characteristics:
Factors that describe study participants at the beginning of the study (e.g., Age, sex, disease severity). In comparison studies, it is important that these characteristics be initially similar between groups; if not balanced or if the imbalance is not statistically adjusted, these characteristics can bias the study results.
Beta Error (type II error):
Falsely determining that NO statistically significant difference between two study variables exists when, in reality, a difference truly does exist (i.e., false negative).
Bias:
A systematic error that provides results different from the true results and leads to a potentially wrong interpretation of the overall study results.
Blinding:
A study design method utilized to neutralize the potential for biased study results based on inaccurate reports from patients, health care providers, and study investigators. Also referred to as 'masking'.
Case-Control study:
An observational study design method commonly used when studying rare disease states or outcomes, such as a rare side effect. A case is identified with the outcome that the investigator is studying. Control patients (those lacking the outcome being studied) are then matched to the case as closely as possible, usually using age, gender, race, etc. Then, the investigator retrospectively evaluates the case and control histories to identify a certain exposure or characteristic associated with the case group and not associated with the control group.
Categorical Data:
Data that is in mutually exclusive categories. Includes both nominal and ordinal data.
Censoring
Term used in survival analysis studies to indicate when study subjects are either: 1) Lost-to-follow-up 2) Withdrawn from the study 3) At study conclusion before the death or event can take place
Cohort:
A group of study subjects who share a common exposure or characteristic. A cohort study commonly compares two large groups of individuals- those who have received (but not assigned) a specific intervention and those who have not. The two groups are observed prospectively over time to evaluate the long-term outcomes.
Cohort Study:
An observational study in which a defined group of people (the cohort) is followed over time and outcomes are compared in subsets of the cohort who were exposed, or not exposed at different levels, to an intervention or other factor of interest. Cohort studies may be done prospectively (collecting data forward in time) or retrospectively (using data already previously collected).
Cointervention:
A treatment or treatments that are NOT the primary study intervention (concomitant treatments) that may actually have an effect on the outcome. It will bias the study results since some of the effect is due to the cointervention instead of the intervention being studied.
Comparative Effectiveness Research
Studies that seek to compare the effects of active comparator interventions to each other. This distinguishes it from placebo-controlled trials which compare active drug to a non-active placebo drug.
Concealment (Allocation concealment):
Randomization is concealed if the person who is making the decision about enrolling a patient is unaware of whether the next patient enrolled will be entered in the intervention or control group.
Confidence Interval:
The range of values within which we are confident that the "true" value of a parameter (e.g., a mean, a relative risk) lies. Most times, 95% CI are used. The CI is related to the alpha level set for statistical significance.
Confounding Variables:
A third factor (e.g., patient characteristic, environmental cause, changes in medical practices over time) that is related to BOTH exposure & outcome AND that accounts for some/all of the observed relationship between the exposure and outcome. These variables may be called covariates in some studies.
Contamination:
Occurs when participants in either the experimental or control group receive the intervention intended for the other arm of the study.
Continuous Data:
Data that could take on any value in a range. (e.g. height, weight, blood pressure)
Control Group:
A group of patients that do not receive the study intervention. They may receive a placebo or other active treatment that is often times the "gold standard".
Correlation:
The magnitude of the relationship or association between two variables.
Cross-over Trial:
A type of study design in which study participants receive both treatments during the study, switching at a pre-defined time point in the protocol. By "crossing over" to the comparison treatment arm, each participant serves as its own control group.
Cross-sectional Study:
The observation of a defined population at a single point in time or during a specific interval. Exposure and outcome are determined simultaneously. Often, surveys are cross-sectional.
Dependent Variable:
The outcome variable in a study. This variable is affected by changes in the independent variable.
Double Blinding:
The subject and study investigators are unaware of what treatment the subject has been assigned.
Double Dummy:
A form of double-blind study called a "double-dummy" design allows additional insurance against bias or placebo effect. All patients are given both placebo and active medication during the study. For example, in a study where an oral medication is being compared with a subcutaneous medication, Group 1 receives active oral medication plus placebo subcutaneous medication and Group 2 is given placebo oral medication plus active subcutaneous medication.
Effect Modification
Also referred to as 'Interaction'. Situation in which the true effect of an exposure differs depending on a subject characteristic. For example, the effect of antioxidants for cardiovascular benefits is proposed to be different for men and women. In this case, gender is an effect modifier or simply 'a modifier' of antioxidant benefit.
Effect Size:
Difference in outcomes between the study group and the control group divided by a measurement of variance such as the standard deviation.
Effectiveness:
A term used to describe a treatment's impact on a particular disease state in a "real world" setting.
Efficacy:
The measurement of a treatment's "potential" impact on a particular disease state in a "perfect" setting, such as a clinical trial.
Endpoint:
Health event(s) or outcome(s) that signify completion or termination from a trial such as death or loss to follow-up.
Equivalence Study:
A study designed to show equivalence between treatments in safety, efficacy, or other parameters. Also called non-superiority studies.
Event Rate:
Proportion of patients in the experimental group (experimental event rate) and/or control group (control event rate) that have a predefined event observed.
Exclusion Criteria:
The characteristics that render potential participants ineligible to participate in a study or that render studies ineligible for inclusion in a systematic review.
External Validity:
A term used to describe the degree to which study results may be applied or generalized to a patient or population of patients other than those studied. It is also called generalizability.
Factorial Study:
A type of study design used in RCTs where two or more treatment factors are being compared simultaneously. Subjects are allocated to different treatment groups featuring different combinations of the treatments.
Follow-up:
A process to determine the outcome in every patient who participated in a clinical trial.
Gold Standard:
A well accepted therapeutic approach considered the standard of practice for a particular disease state.
Hazard Ratio:
The weighted relative risk of a pre-defined event over the time of the entire study.
Incidence:
Number of new cases of a specific disease over a set time.
Inclusion Criteria:
The characteristics that define the population eligible for a study or that define the studies that will be eligible for inclusion in a study.
Independent Variable:
The treatment variable that is assumed to have some effect on the outcome or dependent variable.
Intention-to-Treat (ITT) Analysis:
A method of statistical analysis that maintains randomization by incorporating all patients in their originally assigned groups during data analysis, regardless of whether the patient finished the study. This also includes patients lost to follow-up. This type of analysis maintains prognostic balance of randomization since no patients are excluded from analysis.
Interaction
See 'Effect Modification'
Internal Validity:
A term used to describe study methodology and analysis that is free from systematic bias. It describes the extent to which results from a study can be said to reflect the "true" results when study design and methodology are taken into consideration.
Lost-to-follow up:
The inability to determine the outcome of a patient who participated in a clinical trial.
Matching:
A deliberate process in which individual case subjects are matched to 1 or more control subjects to make the intervention group and comparison group more similar with respect to factors that are extraneous to the purpose of the investigation but that might interfere with the interpretation of the study's findings. Commonly done in case-control studies.
Mean:
A measurement of central tendency or, in other words, the arithmetic average.
Median:
The middle value of ordered data observations. It is generally used as a descriptive measure with skewed continuous data since it is not prone to influence from outlier data points as occurs with calculation of the mean value.
Meta-Analysis:
A statistical method of compiling results from various studies to obtain a pooled result representing "new data". The goal of a meta-analysis is to increase statistical power through an increase in the pooled sample size.
Negative Study:
A study with results that show no statistical differences exist between the two comparison groups.
Nominal Data:
Data that is classified into exclusive categories without a specific ranking order to the categories.
Non-inferiority Study:
This type of trial is helpful when we want to see whether a cheaper, safer, simpler intervention is not "that much" worse in terms of efficacy as what is done currently. "That much" needs to be defined a priori.
Non-parametric Tests:
Statistical tests that do not assume a normal distribution of data. These statistical tests are used when the data is skewed, but can also be used for non-skewed data.
Null Hypothesis:
For a 2-tailed test, the null hypothesis is no difference between groups in terms of the measured outcome. For a typical superiority study, the statistical null hypothesis is that an outcome variable has no association with an independent variable or set of variables. The null hypothesis is what we must reject based on study results in order to demonstrate efficacy or group difference.
Number Needed to Harm (NNH):
The number of patients who, if they received the experimental intervention, would lead to 1 additional patient being harmed during a specific period.
Number Needed to Treat (NNT):
The number of patients who need to be treated during a specific period to achieve 1 additional good outcome or prevent 1 additional bad outcome. It is the inverse of the absolute risk reduction.
Observational Study:
A study in which nature is allowed to take its course. Changes or differences in one characteristic are studied in relation to changes or differences in others, without the intervention of the investigator.
Odds Ratio (OR):
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. In a case-control study, it refers to the odds of exposure vs non-exposure in cases to the odds of exposure vs non-exposure in the control group.
Open Label:
When there is no attempt at blinding the intervention in a clinical trial.
Ordinal Data:
Data that are organized into more than two categories where there is a natural order to the categories.
Outcome:
The results associated with a medical intervention on a subject or patient.
Parametric Tests:
Statistical tests that assume the data is normally distributed
Per Protocol Analysis:
An analysis restricted to patients who adhered to their assigned treatment in a randomized trial (omitting patients who dropped out of the study or for other reasons did not actually receive the planned intervention).
PICO (Patient, Intervention, Comparison, Outcome):
A method for answering clinical questions.
Placebo:
An inactive substance or procedure administered to patients, usually to compare its effects with those of a real drug or other intervention.
Positive Study:
A study with results that show a difference that investigators interpret as beyond the play of chance. (A term referring to a study with results indicating a beneficial effect of the intervention being studied.)
Power:
The ability of a study to reject a null hypothesis when it is false (and should be rejected). Sometimes stated as the ability to find a difference when in reality there is a difference. Power is linked to number of events that occur in a study and the adequacy of the sample size: if a sample size is too small, the study will have insufficient power to detect differences between groups. Mathematically, it equals 1 - β. See Beta error.
Precision:
A measure of the likelihood of random errors in the results of a study, meta-analysis, or measurement. Confidence intervals around the estimate of effect from each study are a measure of precision.
Prevalence:
The total number of existing cases of a disease in a defined population over a stated period.
Prospective Study Design:
In evaluations of the effect of health care interventions, a study in which people are divided into groups that are exposed or not exposed to the intervention(s) of interest before the outcomes have occurred.
P-Value:
The probability that results as extreme as or more extreme than those observed would occur if the null hypothesis were true and the experiment were repeated over and over. A smaller p-value indicates a smaller probability that the differences between groups are due to chance alone.
Randomization:
Allocation of participants to groups by chance, usually done with the aid of a table or random numbers generated by a computer program.
Randomized Controlled Trial (RCT):
Experiment in which individuals are randomly allocated to receive or not receive one or more interventions that are being compared.
Relative Risk (RR):
The ratio of the risk of an event among an exposed population to the risk among the unexposed. Also called Risk Ratio or Rate Ratio.
Relative Risk Reduction (RRR):
The percent reduction in events in the treated group event rate compared with the control group event rate.
Reliability:
The consistency of test results on repeated measures. Reliability values range from 0 (no agreement) to 1 (complete agreement). Examples of reliability tests include, Kappa statistic, intraclass correlation coefficient (ICC) and Cronbach's alpha.
Responsiveness:
The ability of an instrument to detect small but important clinical changes. It is directly related to the magnitude of change in subject scores, which constitutes a clinically important change. Responsiveness includes sensitivity or detecting a change when change actually occurs and specificity - not detecting change when it doesn't occur.
Retrospective Study Design:
A study in which the outcomes have occurred to the participants before the study commenced.
Sample:
Part of the population that is selected to participate in the study.
Selection Bias:
Systematic differences between intervention and comparison groups attributable to the manner in which subjects were allocated to experimental and control groups. Also called treatment selection bias.
Sensitivity Analysis:
An analysis used to determine how sensitive the results of a study or systematic review are to the changes in how it was done.
Single Blinding:
Either the subject or the study investigator, but only one of them, is unaware of what treatment arm the subject is assigned.
Statistical Significance:
A term indicating that the results obtained in an analysis of study data are unlikely to have occurred by chance and the null hypothesis was rejected (most often P<0.05).
Stratified Randomization:
Used to ensure that equal numbers of participants with a characteristic thought to affect prognosis or response to the intervention will be allocated to each comparison group.
Subgroup Analysis:
The separate analysis of data for subgroups of patients, such as those at different stages of illness, those with different comorbid conditions, or those of different ages. Subgroup results different than the overall estimate of effect are considered hypothesis generating and should be tested in a future prospective study.
Systematic Review:
A review of a clearly formulated question that uses systematic and explicit methods to identify, select, and critically appraise the relevant research, and to collect and analyze data from the studies that are included in the review.
Type I Error:
See Alpha Error
Type II Error:
See Beta Error
Validity:
The degree to which a result of a measurement or study is likely to be true and free of bias.
Sets with similar terms
EBM
50 terms
Research Methods Definitions
28 terms
ILE 3: Biostatistics
81 terms
Sci Lit Quiz 1
95 terms
Other sets by this creator
ID test 2 bug drug
49 terms
Exam 1 bug drug chart
34 terms
ID exam 2 bug drug chart
76 terms
Bugs and Drugs (352)
35 terms