Case-control study: An observational study that compares individuals with a specific disease or outcome of interest (cases) to individuals from the same population without that disease or outcome (controls) and seeks to find associations between the outcome and prior exposure to particular risk factors.6 This design is particularly useful where the outcome is rare and past exposure can be reliably measured. Case-control studies are usually retrospective, but not always.

Causality: An association between two characteristics that can be demonstrated to be due to cause and effect (i.e., a change in one causes change in the other).6 Experimental studies such as randomized controlled trials can be used to support causality.6 However, observational studies usually cannot determine causality.6 See the Bradford-Hill Criteria for assessing evidence of causation.8, 9 Sometimes called causation or causal effect.

Centers for Education and Research on Therapeutics (CERTs): A national initiative to increase awareness of the benefits and harms of new, existing, or combined uses of therapeutics (drugs, medical devices, and biological products) through education and research.10 The CERTs program is a network of research centers, each focusing on a broad therapeutic theme. The program is funded and run as a cooperative agreement by the Agency for Healthcare Research and Quality, in consultation with the U.S. Food and Drug Administration (FDA). Website: http://certs.hhs.gov/.

Clinical outcomes: Medical events occurring as a result of disease or treatment (e.g., stroke, disability, hospitalization). Also called clinical endpoint.1

Clinical practice guideline: User-friendly, evidence-based, systematically developed statements to assist primary health providers and patients in making appropriate health care decisions.1

Clinical trial: A prospective experimental study that tests the safety, efficacy, and/or effectiveness of a health care intervention intended to prevent, diagnosis, or treat a specific disease or condition in humans.1 An umbrella term for a variety of designs of health care trials.6

Cluster randomized trial: A randomized controlled trial in which participants are randomly assigned to the intervention in groups (clusters) defined by a common feature, such as the same physician or health plan.11

Cochran’s Q test: The classical test used in meta-analysis to assess whether a set of individual studies is heterogenous.12 An indication of the presence or absence of heterogeneity, but not the extent of heterogeneity. Also called Cochran’s Q statistic. See also I2 statistic.

Cohort: A group of participants who remain together in the same study over time.4

Cohort study: An observational study with a defined group of participants (the cohort) that is followed over time.6 Outcomes are compared between subsets of this cohort who were exposed or not exposed (or exposed at different levels) to a particular intervention or other factors of interest. A prospective cohort study identifies participants and follows them into the future. A retrospective (or historical) cohort study identifies participants from past records and follows them from a previous time point to the present.

Comparative effectiveness research (CER): The generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat, and monitor a clinical condition or to improve the delivery of care.13 The purpose of CER is to assist consumers, clinicians, purchasers, and policymakers to make informed decisions that will improve health care at both the individual and population levels.

Comparative effectiveness reviews: Systematic reviews that evaluate evidence on alternative interventions to help clinicians, policymakers, and patients make informed treatment decisions.14 Many comparative effectiveness reviews funded by the Agency for Healthcare Research and Quality’s Effective Health Care (EHC) Program are developed by Evidence-based Practice Centers (EPCs).15 The other type of research review produced by the EHC Program is called a technical brief.15

Comparison group: See control group.

Compliance: See adherence.

Complications: A term often used to describe adverse events following surgery or other invasive interventions.2

Composite endpoint: Endpoints that capture the number of patients who experience one or more of several events of interest in clinical trials.16 Aggregates of individual endpoints may be used to increase the event rate and thus the statistical power of the study and to capture the overall impact of interventions.16 Study results with composite endpoints may be misleading if the individual endpoints are of varying clinical importance, the number of events in the more important components is small, or the magnitude of effect differs markedly across components.17 Also called composite outcome.

Concealment of allocation: The process used to ensure that the investigator enrolling a participant into a randomized controlled trial does not know the group to which the participant is assigned.6 This process is aimed at preventing selection bias and is distinct from blinding. Some attempts at concealing allocation are more prone to manipulation than others, and the method of allocation concealment is used as an assessment of the quality of a trial.

Confidence interval (CI): A measure of the uncertainty around the main finding of a statistical analysis.4, 6, 7 If the study were repeated multiple times, it is the range of values within which the mean for each trial would occur 95% of the time. Estimates of unknown quantities (e.g., odds ratio) are usually presented as a point estimate and a 95% confidence interval. Alternatives to 95%, such as 90% and 99%, are sometimes used. Wider intervals indicate lower precision and narrow intervals indicate greater precision.

Confidence limits: The upper and lower boundaries of a confidence interval.6

Confounder: See confounding variable.

Confounding variable: A variable (or characteristic) more likely to be present in one group of participants than another that is related to the outcome of interest and may potentially confuse (confound) the results.4 For example, if individuals in the experimental group of a controlled trial are younger than those in the control group, it will be difficult to determine whether a lower risk of death in one group is due to the intervention or the difference in ages (age is the confounding variable).6 Randomization is used to minimize imbalances in confounding variables between experimental and control groups.6 Confounding is a major concern in non-randomized studies.6 Also called confounder.

Consistency: The extent to which the effects from studies in a systematic review appear to have the same direction and magnitude.18 See also effect or effect size. Sometimes also refers to the reliability of a measure or study to reproduce the same results.

CONSORT: Acronym for Consolidated Standards of Reporting Trials.19 Encompasses various initiatives developed by the CONSORT Group to alleviate the problems arising from inadequate reporting of randomized controlled trials. Extensions of the CONSORT Statement have been developed for other types of study designs, interventions, and data.

Construct validity: The degree to which the items on a test or measurement scale actually represent the characteristic being measured (usually not observable).1, 4 See also validity.

Content validity: The degree to which a test or measurement scale actually measures what it is designed to measure as determined by expert opinion.1 For example, a depression scale that assesses only one symptom of depression (e.g., cries a lot) will have a low content validity. A depression scale that assesses all major symptoms of depression will have higher content validity. See also validity.

Control: 1. In a controlled trial, a participant in the group receiving placebo, no treatment, an active comparator, or standard of care) that serves as a comparator for the experimental intervention.4, 6 Also called control participants. 2. In a case-control study, an individual in the group without the disease or outcome of interest.6 3. In statistics, to adjust for, or take into account, extraneous influences or observations.6

Control group: Participants in the control arm of a study. See also control, controlled trial, experimental group, and treatment group.

Controlled trial: A clinical trial that has a control group.6, 11 More specifically, an experimental study that compares the outcomes observed in one study group (or arm) receiving the intervention of interest (experimental group) to one or more comparison (control) group(s) receiving placebo, no treatment, an active comparator, or standard of care. Such trials are not necessarily randomized. Also called controlled clinical trial.

Conventional treatment: See standard of care.

Correlation: See association.

Criterion validity: An indication of how well a test or scale predicts another related characteristic or outcome.1, 4 May be tested when the results obtained by one instrument can be verified through an independent observation or another instrument that has already been validated, ideally a “gold standard” if one exists. See also validity.

Critical appraisal: The process of assessing and interpreting evidence by systematically considering its validity, results, and relevance.6

Cross-sectional study: An observational study that examines a characteristic (or set of characteristics) in a set of participants at a specific time or time period.4, 11

Cumulative meta-analysis: A meta-analysis that adds studies one at a time in a specified order (e.g. according to date of publication or quality) and the results are summarized as each new study is added.6 In a graph of a cumulative meta-analysis, each horizontal line represents the summary of the results as each study is added, rather than the results of a single study.