The glossary below provides definitions of some common terms encountered in clinical research. We welcome your suggestions for further terms you think it would be useful to include, or feedback on the definitions provided. Please add any comments in the box at the bottom of the screen.
An active comparator is an existing known medication or intervention for a condition or process that is used for one arm of a clinical study or trial. An active comparator contrasts with other types of comparators such as a placebo, for example, which is inactive but simply a means of comparison for the intervention.
Adverse Event (AE)
Any type of medical event (for example a symptom/side effect or laboratory finding) which is temporally linked to the investigational intervention and may (or may not) be caused by it. The protocol should specify which grades of events will be recorded, as many trials will not require recording of minor events. The links below show some examples of gradings of adverse events:
Adverse Reaction (AR)
Any Adverse Event (AE) that is, in the opinion of a clinician, causally related to an Investigational Medicinal Product (IMP)
In a clinical trial, an ‘arm’ is an assigned group, for example a trial may include a placebo arm and an investigational intervention arm. Some types of trial design include more than two arms. Participants are usually randomly assigned to these groups.
Respecting a participant’s autonomy in clinical research involves respecting their capacity to make decisions about research participation, either on an individual basis, or following discussion with others, such as family members.
The point in a study or clinical trial prior to the respondent taking the investigational intervention (or control). Baseline measurements may be considered the ‘known value’ against which other measurements in the trial can be compared.
Beneficence (see also non-maleficence)
The principle of doing good and promoting the interests/wellbeing of research participants.
Blinding(or masking) – see also double blind
A blind (or mask) is a design in which one of the parties in a trial (e.g. the investigator, the patient) is unaware of the investigational group to which they’ve been assigned – for example, there may be two arms in a trial, the control arm (could be a placebo) and an arm involving the investigational intervention. In a single-blinded trial, the patient is not aware of which group they are in.
"A clinical trial is any research study that prospectively assigns human participants or groups of humans to one or more health-related interventions to evaluate the effects on health outcomes. Interventions include but are not restricted to drugs, cells and other biological products, surgical procedures, radiological procedures, devices, behavioural treatments, process-of-care changes, preventive care, etc" (from WHO/ICMJE – see refs)
A clinical trial is one type of clinical research (see below).
Empirical research happens whenever there is planned systematic collection of data with the intention to analyse them, make the findings generalizable, and publish them. Clinical research, whether it involves an intervention (drug delivery, surgery, etc.) or not (sample taking and observation) is concerned with human health, and therefore requires a protocol defining the question and an ethical approval prior to starting data collection, because non-standard information and/or samples are being collected.
Clinical research includes clinical trials as one category of research, as well as other types of research such as observational studies, social science, etc.
A perceived pressire which could force or unduly encourage someone to take part in research, whether or not they want to. This could range from a threat, such as suggesting that they will not have adequate healthcare if they do not take part, as well as other types of coercion, such as offering inflated monetary incentives for them to take part.
A group of individuals with a common characteristic (epidemiological and statistical term). For example, the group may all have been born in a certain year, share a disease and another factor (e.g. age and disease onset time), lifestyle characteristics such as being a smoker, and so on. Cohort studies often therefore involve observing a group with a common characteristic over a period of time, so that the epidemiological effects of the characteristic can be observed.
The process whereby potential participants in a clinical trial are informed about the potential risks and benefits of participating in the trial. They should also be informed of all procedures to be carried out and what would be required of them, as well as their rights to decline or withdraw from the study at any time. The completion of the process is frequently documented on a signed consent form.
Special ethical concerns exist for those not able to consent for themselves (e.g. minors or unconscious adults). In studies for emergency conditions it may be appropriate to have an initial basic consent process with further information later. In very rare cases it may be appropriate for professional representatives (e.g. doctors or judges) to consent on behalf of an individual.
In a controlled trial, one intervention is measured against another– e.g. a new drug might be tested against a placebo or an existing medication for the condition, or a new vaccine tested against an existing vaccine (the existing vaccine being the control). In this trial design, subjects are usually randomly assigned to either the control arm of the trial or the investigational arm(s). Commonly these trials are single or double blinded (see blinding, double blinded).
CRF- Case Report Form (sometimes also known as Case Record Form)
A form which records full trial information as required by the trial protocol for an individual subject – this form can be electronic or on paper – electronic CRFs may be known as eCRFs. A CRF will include data which will contribute to determining all measurements as required for the trial, for example this may include tumour size, responses to questionnaires on symptom relief, responses to Quality of Life questionnaires and so on.
Declaration of Helsinki
Originally adopted in 1964, the Declaration of Helsinki set out fundamental ethical rights for human subjects of clinical research. The Declaration has been revised many times since then (most recently in 2008) but has no inherent legal force. The EU currently recognises the fourth revision (1996). The FDA no longer refers to the Declaration of Helsinki at all in its requirements, instead using ICH GCP.
Double blinded (or masked)
A trial in which two parties (the patient and the investigator, for example) are unaware of which arm of a trial (control or investigational intervention) they have been assigned to until the end of the trial. It may also be the case that the trial monitors and data analysts are unaware of the assigned groups.
This study design should lead to more objective results than an unblinded trial, because the expectations of the doctor or patient about the investigational intervention have no influence.
DSMB (data safety monitoring board)
The DSMB (also known as a Data Monitoring Board (DMB), Independent Data Monitor Committee (IDMC), Data Safety and Monitoring Committee (DSMC)) in a trial is a group of individuals who are independent (not affiliated to the trial or its sponsors), and are responsible for monitoring the trial and the data as it arises during the trial, to ensure that subjects are not being put at undue risk. DSMBs are typically expected to include a statistician and a clinician. Not every clinical trial has to have a DSMB.
An endpoint is an aspect that the trial is designed to measure. For example, a trial or study relating to testing a new intervention for treating malaria may have an endpoint of death – the measurements could relate to whether the new intervention increased survival rates. Multiple endpoints are likely in one trial; for example, a malaria trial may have death as its primary (main) objective endpoint, but have a secondary endpoint related to symptom relief. In this case the trial could mainly be concerned with the survival rate of the patients, but would also relate to the symptoms that the patient is suffering and whether or not they are experiencing symptom relief from the intervention.
ERC (Ethics Review Committee)
An ERC (Ethics Review Committee) (also known as an EC (Ethics Committee), ERB (Ethics Review Board), or IEC (independent ethics committee)) is a group of people who are independent of) the trial or its institution/researchers/sponsors or other related parties. The ERC is responsible for reviewing the research to protect the rights of the participants and to ensure that it is ethical and scientifically robust. The committee should consist of a diverse and multidisciplinary group of individuals who between them have the necessary medical and ethical expertise to evaluate the proposed research. The panel should include at least one representative of the community (this person doesn’t necessarily have to have a scientific background), often referred to as a lay member, as well as diverse scientific experts, physicians, statisticians and researchers.
Every institution conducting or supporting research (whether biomedical or behavioural) must have approval of an ERC before the research goes ahead.
Like an ERC, an IRB (Institutional Review Board) is also responsible for ensuring the rights and protection of the participants in a study. However, an IRB typically conducts a detailed review of the science of the research, protocol and development of any intervention as well as the ethics. Institutional review boards are usually affiliated with an institution, for example one individual hospital or a society, while ethics committees may review the research of a number of different research centres. Institutional review is therefore performed prior to ethics review in some places, but this will depend on the individual location.
In some settings, the terms ERC and IRB are used interchangeably.
Epidemiological research is based on disease incidence or prevalence and distributionacross a population or community.
Exploitation (see also justice)
Offering a research participant less than they should be entitled to simply because they live in a resource-poor environment. It could involve not remunerating them fairly given their participation in the trial, and/or not giving them fair access to the benefits of the trial.
GCP (Good Clinical Practice)
A standard as laid out by the International Conference for Harmonisation of Good Clinical Practice (ICH-GCP) which applies to all aspects of clinical research, and aims to ensure that the results yielded are obtained in a scientific, credible, accurate and professional manner, whilst enduring the rights and confidentiality of participants.
The ICH GCP guidelines were developed in 1996 in response to the need for international harmonisation of the guidelines on human research.
HRQoL (Health Related Quality of Life) and QoL (Quality of Life) are measurements of the patients’ well-being, comfort and standard of living in their day to day functioning and everyday life. HRQoL may be a primary endpoint in some chronic illness trials, but may also be a secondary or other endpoint for many other conditions. QoL is usually measured by self-report questionnaires.
IEC (see ERC)
Clinical trials should have some form of indemnity (insurance, protection) in place to compensate participants for any unforeseen adverse outcomes arising from the interventions or medical negligence (deviation from the protocol). This does not include compensation for outcomes related to the condition being studied nor (usually) to risks clearly outlined in the informed consent process.
Informed Consent (see consent)
Investigational Medicinal Product (IMP)
The drug or drugs being evaluated in a clinical trial of a drug (as opposed to non-drug interventions), such a clinical trial may be referred to as a CTIMP (a clinical trial of an investigational medicinal product).
Justice (see also exploitation)
Justice requires that participants are treated fairly, i.e. there is a fair distribution of risks and benefits in research (one group is not unreasonably asked to assume all the risks of research while another receives all the benefits). If research participants are not to be treated equally, there must be appropriate grounds for the inequity (such as participants having differing needs).
Non-maleficence (see also beneficence)
The principle of minimising harm to research participants.
Observational research is research in which no investigational intervention is applied to the participants; they are simply observed without interference. Therefore observational research does not include studying an intervention or sample-taking; instead it could include for example epidemiological research to observe a group of respondents to determine the common characteristics which might relate to a disease, research about patients’ experienced with standard care, and so on.
Operational research uses scientific or mathematical modelling to try to rationally structure, understand and resolve complex problems. It has its roots in military planning in the Second World War, but is very useful in the medical sector, for example for disease modelling, emergency planning, logistics and so on. The use of modelling to help resolve a problem can help researchers to gain more insight and information into the system that they are using and the best ways of overcoming problems through strategic utilisation of their resources. Studies or trials can be conducted to evaluate operational strategies for healthcare delivery; these may not be drug trials (CTIMPs).
An outcome is the measurable event for trial endpoints, but is usually more general than an endpoint; for example, an endpoint may look at the efficacy of an intervention, while outcomes will look at other aspects such as whether the intervention improves quality of life, works more quickly than existingmedications, is more practical, and so on.
For example where improving survival is the primary objective then the outcome is mortality with the trial endpoint being death. Or in a pain study the patient reported measurements are the outcome and the endpoint is pain.
The outcomes of the trial should directly relate to the planned endpoints as laid out in the trial protocol and the objectives. Outcomes can be patient-reported (i.e. the patient tells how they’re feeling, for example filling in questionnaires on their pain level of quality of life), clinician-reported (reported by a doctor), or caregiver-reported (reported by the person who looks after the patient – this may be in the case that the patient is unable to give their own views on their feelings, for example in cases of extreme disability).
There are different types of outcomes such as;
- Clinical Outcomes – directly relevant to the patients, such as length of hospital stay, or a clear event such as stroke or death. Single clinical outcomes often need very large sample sizes.
- Surrogate Outcomes - a laboratory measurement or a physical sign used as a substitute for a clinically meaningful outcome that measures directly how a patient feels, functions or survives (for example CD4 cell counts for AIDS related morbidity).
- Composite Outcomes – several clinical events are used as this can shorted the length of the study, lower the number of subjected needed and give a clearer answer to the primary objective.
- It is important for protocols to state what type of outcome they are using and for surrogate outcomes to state whether they are validated.
Phase I is the first phase in a trial of an investigational drug or vaccine in humans. Prior to phase one, researchers will have considered the dosing requirements through testing with animals or on tissues. Phase I usually involves giving different doses to different participants to look for any short-term side-effects or toxicity and evaluate the metabolism and pharmacologic action of the drug or vaccine with humans specifically. Typically phase I trials involve a small number of healthy volunteers compared with the following phases.
Phase II trials involve a controlled (comparative) study of the intervention to determine its effectiveness with patients. This phase also aims at determining the safety, dosage, side effects and risks of the medication.Typically phase II studies involve around 100-300 patients.
Phase III trials are intended to gain further insight into the safety and efficacy of an intervention in the population for which its use is intended, and therefore this phase should create analysable data about the risk-benefit ratio of the drug. The data from this phase is then used as the basis for regulatory approval of the drug and subsequent labelling claims.
A Phase IV study may also be known as a surveillance or post-marketing study, because it occurs once the drug has been approved by the regulatory authorities and is being marketed. This phase is useful for identifying any effects of the drug which occur in a minority of users, and also provides additional information on efficacy and safety.
A comparator to an investigational intervention that is expected to have no clinical activity (as opposed to an active comparator (see above)), used to blind participants (and trial staff) to which arm of the trial a participant is in. It is often considered unethical to use a placebo as the sole comparator if a known active treatment exists.
The protocol for clinical research is the plan on which the study is based. It manifests as a document particular to that piece of research, which details the study design and organisation, methodology (including exclusion/inclusion criteria for patients, dosing requirements, length of study etc), objectives and endpoints. It may also detail the rationale and background of the trial.
The protocol should be designed so as to safeguard the rights of participants and to answer the study questions.
QoL (see HRQoL)
Serious Adverse Event (SAE)
An SAE is any unexpected outcome in a participant which results in death, is life-threatening, requires hospitalisation (or prolongs existing hospitalisation), results in persistent/significant disability or results in a birth defect (congenital abnormality). It may or may not be related to the intervention
Serious Adverse Reaction (SAR) - See also SUSAR
A serious adverse event where a causal relationship between a medicinal product and the SAE is at least a reasonable possibility, i.e., the relationship cannot be ruled out. A serious AR (in comparison with a more general AR) is one that results in death, is life-threatening, requires hospitalisation (or prolongs existing hospitalisation), results in persistent/significant disability or results in a birth defect (congenital abnormality).
“an individual, company, institution or organisation which takes responsibility for the initiation, the management and for setting up the financing of the clinical trial” [ Directive 2001/20.EC and ICH-GCP definition ]. The role of the sponsor includes, but is not restricted to, legal responsibilities, oversight of indemnity and financing, and ensuring that the individuals involved in the research project are appropriately insured, if necessary (especially where the sponsor is also the PI).
Suspected Unexpected Serious Adverse Reaction (SUSAR)
A serious reaction that is likely to be related to the product or intervention (relationship cannot be ruled out) and also not expected with this product or intervention.
References and further reading:
C-Disc Clinical Research Glossary: (http://www.cdisc.org/system/files/all/article/application/pdf/cdisc_2009_glossary.pdf)
Clinical trials.gov (http://clinicaltrials.gov/ct2/info/glossary)
Earl-Slater, A (2002), The handbook of clinical trials and other research, Radcliffe Medical Press: Oxford
EMEA ICH-GCP (2006) (http://ichgcp.net/1-glossary)
Fitzpatrick, S (2006), Clinical Trial Design, Institute of Clinical Research: Buckinghamshire
Hackshaw, A (2009), A Concise Guide to Clinical Trials, Wiley-Blackwell: UK
Medical Research Council Clinical Trials Toolkit (http://www.ct-toolkit.ac.uk/glossary.cfm)
WHO: Clinical Trials (http://www.who.int/topics/clinical_trials/en/)