From ‘attribution’ to ‘triangulation’ – we know that evaluation and evidence terminology can be confusing. Our A-Z glossary of terms explains what the commonly used terms mean.


A priori

A priori comparisons are planned in advance of any data analysis. They are more reliable than post-hoc comparisons.

Absolute risk reduction (ARR)

The difference in the event rate between control group (CER) and treated group (EER): ARR = CER - EER.

For reference, see understanding statistics: risk (BMJ best practice).

Allied and Complementary Medicine Database (AMED)

A unique bibliographic database produced by the Health Care Information Service of the British Library. It covers a selection of journals in complementary medicine, palliative care, and several professions allied to medicine. Covers the years from 1985 to present and is updated monthly.

Allocation concealment

Used when randomising participants to treatment groups in an RCT. Allocation is concealed when neither the participants nor the researchers know or can predict to which group in a study (control or treatment) the patient is assigned. Allocation concealment takes place before the study begins as patients are being assigned. It is a sign of a robust study.


To examine associations between variables—these may be hypothesised causal or therapeutic relations.

ARC West

ARC stands for Applied Research Collaboration. ARCs are funded by National Institute of Health and Care Research (NIHR), and there are 15 of these pioneering research teams across the country, which focus on applied research. ARC West covers the West of England. University Hospitals Bristol and Weston NHS Foundation Trust hosts ARC West.


A statistical dependence between two or more events, characteristics or other variables. An association may be fortuitous or may be produced by various other circumstances; the presence of an association does not necessarily imply a causal relationship

Average - mean, median, mode

Mean - The average value, calculated by adding all the observations and dividing by the number of observations.

Median - Middle value of a list. If you have numbers 2, 3, 4, 5, 6, 7, and 8, the median is 5. Medians are often used when data are skewed, meaning that the distribution is uneven. In that case, a few very high numbers could, for instance, change the mean, but they would not change the median. Other definitions include the smallest number such that at least half the numbers in the list are no greater than it. If the list has an odd number of entries, the median is the middle entry in the list after sorting the list into increasing order. If the list has an even number of entries, the median is equal to the sum of the two middle (after sorting) numbers divided by two. The median can be estimated from a histogram by finding the smallest number such that the area under the histogram to the left of that number is 50%. But all mean the same thing in the end.

Mode - For lists, the mode is the most common (frequent) value.



Any tendency to influence the results of a trial (or its interpretation) other than the experimental intervention.


A technique used in research to eliminate bias by hiding the intervention from the patient, clinician, and/or other researchers who are interpreting results.

Bristol Health Partners

Bristol Health Partners is a collaboration between the city's three NHS trusts, three integrated care boards, two universities and its local authority. They aim to maximise Bristol's health research, and transform the understanding, prevention and treatment of key health problems in Bristol.

British National Formulary (BNF)

The BNF (British National Formulary) provide UK healthcare professionals with authoritative and practical information on the selection and clinical use of medicines (includes BNF for Children).

British Nursing Index (BNI)

The BNI (British Nursing Index) is a bibliographic database that indexes articles from the most popular English language nursing journals published primarily in the UK. BNI is a comprehensive index covering all aspects of nursing, midwifery and community healthcare. Covers the years from 1992 to the present and is updated monthly


Case-control study

The observational epidemiologic study of persons with the disease (or other outcome variable) of interest, and a suitable control (comparison, reference) group of persons without the disease. The relationship of an attribute to the disease is examined by comparing the diseased and non-diseased with regard to how frequently the attribute is present or, if quantitative, the levels of the attribute, in each of the groups.


A group or series of case reports involving patients who were given similar treatment. Reports of case series usually contain detailed information about the individual patients. This includes demographic information (for example, age, gender, ethnic origin) and information on diagnosis, treatment, response to treatment, and follow-up after treatment.


The relating of causes to the effects they produce. Most of epidemiology concerns causality and several types of causes can be distinguished. It must be emphasized, however, that epidemiological evidence by itself is insufficient to establish causality, although it can provide powerful circumstantial evidence.


The Cinahl® database is a comprehensive and authoritative source of information for nurses, allied health professionals, and others interested in health care.

Clinical vs statistical significance

It is important to ask whether statistically significant differences are also clinically significant.

Statistical significance is a mathematical technique to measure whether the results of a study are likely to be true, or whether they could have occurred by chance. Statistical significance is calculated as the probability that an effect observed in a research study is occurring because of chance. Statistical significance is usually expressed as a P-value. The smaller the P-value, the less likely it is that the results are due to chance (and more likely that the results are true). Researchers generally believe the results are probably true if the statistical significance is a P-value less than 0.05 (p<.05).
In a large study, a small difference may be statistically significant. For example, does a 1- or 2-point difference on a 100-point dementia scale matter to your patients?

Statistically significant does not necessarily mean important. Size of effect determines clinical importance not the presence of statistical significance. Conversely, if a study finds no difference, it is important to ask whether it was large enough to detect a clinically important difference and if a difference actually existed. A study with too few patients is said to lack the power to detect a difference.

Cochrane Collaboration

The Cochrane Collaboration is a worldwide association of groups who create and maintain systematic reviews of research literature for specific topic areas.

Cohort study

The analytic method of epidemiologic study in which subsets of a defined population can be identified who are, have been, or in the future may be exposed/not exposed/ exposed in different degrees, to a factor or factors hypothesized to influence the probability of occurrence of a given disease or other outcome. The main feature of cohort study is observation of large numbers over a long period (commonly years) with comparison of incidence rates in groups that differ in exposure levels.

Confidence interval (CI)

The range around a study's result within which we would expect the true value to lie. CIs account for the sampling error between the study population and the wider population the study is supposed to represent.

Conflict of interest

A conflict of interest occurs when those who are involved with the conduct or reporting of research also have financial or other interests, or where they can benefit in some other way, depending on the results of the research. The obvious example is where a pharmaceutical company reports results of a trial of its product. Conflict of interest statements often accompany published papers. It is a statement by a contributor to a report or review of personal financial or other interests that could have influenced the findings or their interpretation. Conflicts of interest are the norm, and not the exception.

Confounding variable

A variable which is not the one you are interested in, but which may affect the results of trial.

Control event rate (CER)

Control event rate (see event rate)

Critical appraisal

The assessment of evidence by systematically reviewing its relevance, validity and results to specific situations. (Chambers, R. (1998). Clinical effectiveness made easy: first thoughts on clinical governance. Oxford, Radcliffe Medical Press.) It is a balanced assessment of benefits and strengths of evidence against its flaws and weaknesses.

Critically appraised topic (CAT)

A short summary of an article from the literature, created to answer a specific clinical question.


Decision analysis

The application of explicit, quantitative methods to analyse decisions under conditions of uncertainty.


The process of determining health status and the factors responsible for producing it, and may be applied to an individual, family, group or community. The term applied both to the process of determination and to its findings.

Diagnostic test

Any medical test performed to confirm, or determine the presence of disease in an individual suspected of having the disease, usually following the report of symptoms, or based on the results of other medical tests. Some examples of diagnostic tests include performing a chest x-ray to diagnose pneumonia, and taking skin biopsy to detect cancerous cells. (Harvard Guide to Diagnostic test)

Directory of Open Access Journals (DOAJ)

DOAJ is the Directory of Open Access Journals.

Disease oriented outcomes (DOOs)

DOOs refers to the outcomes of studies that measure physiologic or surrogate markers of health. This would include things such as blood pressure, serum creatinine, glycohaemoglobin, sensitivity and specificity, or peak flow. Improvements in these outcomes do not always lead to improvements in patient-oriented outcomes (POOs), such as symptoms, morbidity, quality of life, or mortality, which are the type of outcomes patients care more about. Any POO evidence that would change practice is a POEM (patient-oriented evidence that matters).



Experimental event rate (see event rate).


A measure of the benefit resulting from an intervention for a given health problem under usual conditions of clinical care, for a particular group.


A measure of the benefit resulting from an intervention for a given health problem under the ideal conditions of an investigation.


A biomedical and pharmacological database with good Western European source coverage, and therefore complementary to Medline. The emphasis is on the pharmacological effects of drugs and chemicals. Covers the years 1980 to present and is updated daily.

EMIS reference aka Mentor

Accessible via EMIS.  This is an on-line system for searching and displaying the full contents of EMIS medical reference publications.


Adjective to describe activity that assesses relevance, effectiveness, and consequences.

Event rate

The proportion of patients in a group in whom an event is observed.


Adjective describing something that makes observations intelligible and understandable.


Forrest plot

A diagrammatic representation of the results of individual trials in a meta-analysis.

Funnel plot

A method of graphing the results of trials in a meta-analysis to show if the results have been affected by publication bias.


Google Scholar

Provides a search of scholarly literature from one place across many disciplines and sources, including theses, books, abstracts and articles.


Health Business Elite

Database of journal content detailing all aspects of healthcare administration, and other non-clinical aspects of health care institution management including hospital management, hospital administration, marketing, human resources, computer technology, facilities management and insurance from 1922 to present (updates not stated). Access via NICE Journals and Databases

Health Foundation

The Health Foundation is an independent charity working to continuously improve  the quality of healthcare in the UK.  It conducts research and evaluation, puts ideas into practice through a range of improvement programmes, supports and develops leaders, and shares evidence to drive wider change.  Its work is centred on two priority areas – patient safety and person-centred care.

Health Innovation West of England

Health Innovation West of England is a diverse network of NHS care providers in universities, industry and commissioning concerned with innovative, evidence-based practice to improve health and care quality.

Health Integration Team (HITs)

Local cross-organisational and interdisciplinary groups tackling major health priorities by working together in an integrated way.


In systematic reviews, the amount of incompatibility between trials included in the review, whether clinical (i.e. the studies are clinically different) or statistical (i.e. the results are different from one another).

Historically controlled study

A control study recruiting control subject(s) for whom data was collected before the time data was gathered on the group being studied.


The Health Management Information Consortium (HMIC) database is a compilation of data from two sources, the Department of Health's Library and Information Services and King's Fund Information and Library Service. The majority of records are from 1983 onwards, although coverage of departmental materials dates back to 1919. Available via NHS Evidence Services.


Inception cohort study

A group of individuals identified for subsequent study at an early, uniform point in the course of the specified health condition, or before the condition develops.


The number of new cases of illness commencing, or of persons falling ill, during a specified time-period in a given population.


Characteristic of a study where patients are analysed in the groups to which they were originally assigned, even though they may have switched treatment arms during the study for clinical reasons.


Description of activity that identifies and explains meaning, usually from a particular perspective


This employs subjectivity and is characterised by multiple realities; meanings that individuals attach to experiences and views of the world and utilises qualitative methodologies. Its language is that of words.


Joint strategic needs assessment (JSNA)

Joint Strategic Needs Assessment is a key source of evidence that describes current and future health and care needs of the local population to inform and guide the planning and commissioning of health and social care services within a local authority area.


King’s Fund

The King’s Fund is an independent charitable organisation that works to improve healthcare in the UK by providing research and health policy analysis. Offers library services and an information database.

Knowledge management

A conscious strategy for moving the right knowledge to the right people at the right time which will be translated into action to improve the organisational performance.


Likelihood ratio

The likelihood that a given test-result would be expected in a patient with the target disorder, compared to the likelihood that the same result would be expected in a patient without that disorder. For a positive test result = LR+ = sensitivity/(1-specificity) For a negative test result = LR- = (1-sensitivity)/specificity.

Literature searches

A literature search is a well-thought-out and organised search for all of the literature published on a topic. A well-structured literature search is the most effective and efficient way to locate sound evidence on the subject you are researching.

Local and current random census

Local: Of, belonging to, or characteristic of a particular locality or neighbourhood.

Current: Occurring in or belonging to the present time

Random sample: A sample that is arrived at by selecting sample units such that each possible unit has a fixed and determinate probability of selection

Census: An enumeration of a population, originally intended for purposes of taxation and military service. Census enumeration of a population of a population usually records identities of all persons in every place of residence, with age, or birth date, sex, occupation, national origin, language, marital status, income, and relationship to head of household - in addition to information on the dwelling place.

Local and current random sample survey

Local: Of, belonging to, or characteristic of a particular locality or neighbourhood.

Current: Occurring in or belonging to the present time.

Random sample: A sample that is arrived at by selecting sample units such that each possible unit has a fixed and determinate probability of selection.

Survey: An investigation in which information is systematically collected but in which the experimental method is not used.

Local non-random sample

Local: Of, belonging to, or characteristic of a particular locality or neighbourhood.

Non-random sample: A sample selected by a non-random method, and as a result, some elements of the population have no chance of selection. For example, a scheme whereby units are selected purposively would yield a non-random sample. Again, a sample obtained by taking members at fixed intervals on a list is a non-random sample, unless the list was arranged in a random order.


Mechanism-based reasoning

Involves an inference from mechanisms to claims that an intervention produces a patient-relevant outcome. Such reasoning will involve an inferential chain, linking the intervention (such as antiarrhythmic drugs) with a clinical outcome (such as mortality). (Howick)


MEDLINE™ is the United States National Library of Medicine's (NLM™) premier bibliographic database providing information from the following fields: Medicine, Veterinary medicine, Nursing, Allied Health, Dentistry and Pre-clinical sciences. Covers the years 1950 to present and is updated weekly.


Medical Subject Headings: a thesaurus of medical terms used by many databases and libraries to index and classify medical information.

Meta analysis

A method often used in systematic reviews. Meta-analyses use specific statistical techniques to combine data from different studies to arrive at one summary point estimate of effect. This should be done only if the studies were of good quality and were reasonably homogeneous (i.e. most had generally similar characteristics).


The overall, guiding approach taken by a research project or research design e.g. a randomised controlled trial, a cohort study, a case-control study, a cross-sectional study, a case series, a grounded theory study, a phenomenological study.


The approaches taken to identify a study sample, recruit participants, collect data or analyse data

Monitoring test

Any medical test performed to confirm, or determine the presence of disease in an individual suspected of having the disease, usually following the report of symptoms, or based on the results of other medical tests. Some examples of diagnostic tests include performing a chest x-ray to diagnose pneumonia, and taking skin biopsy to detect cancerous cells. (Harvard Guide to Diagnostic test).


n-of-1 trial

A variation of a randomised controlled trial in which a sequence of alternative treatment regimens is randomly allocated to a patient. The outcomes of regimens are compared, with the aim of deciding on the optimum regimen for the patient.

Negative predictive value (-PV)

The proportion of people with a negative test who are free of disease.

Nested case-control study

A case control study in which cases and controls are drawn from the population in a cohort study. As some data is already available about both cases and controls, the effects of some potential confounding variables are reduced or eliminated. In this type of case-control study, a set of controls is selected from subjects i.e. non-cases at risk at the time of occurrence of each case that arises in a cohort, thus allowing for the confounding effect of time in the analysis.

NICE Guidelines

NICE Guidelines National Institute for Health and Care Excellence publishes a table of clinical guidelines.  Guidance sets the standards for high quality healthcare and encourages healthy living.  NICE guidance can be used by the NHS, Local Authorities, employers, voluntary groups and anyone else involved in delivering care or promoting wellbeing.  This is available as a free app for mobile devices

Nuffield Trust

Nuffield Trust is an authoritative and independent source of evidence-based research and policy analysis for improving health care in the UK

Number needed to treat (NNT)

The number of patients who need to be treated to prevent one bad outcome. It is the inverse of the ARR: NNT=1/ARR. Numbers needed to harm (NNH) is the number of patients who, if they received the experimental treatment, would lead to one additional person being harmed compared with patients who receive the control treatment. It is calculated as 1/ARI.


Observational study

A family of studies in which investigators compare people who take an intervention, with those who do not. The investigators neither allocate patients to receive the intervention nor administer the intervention. Instead, they compare records of patients who took an intervention and have been treated in routine practice, with similar patients who had not taken the intervention. The most common observational designs are case-studies, case-series, case-control studies, cohort studies, and historically-controlled studies. (Howick)


A ratio of events to non-events. If the event rate for a disease is 0.2 (20%), its non-event rate is 0.8 and therefore its odds are 2/8.


p value

The probability that a particular result would have happened by chance.

Participatory evaluation

“Improve program implementation and outcomes by actively engaging all stakeholders in the evaluation process”


A placebo is a fake or inactive intervention, received by the participants allocated to the control group in a clinical trial. It is indistinguishable from the active intervention received by patients in the experimental group. One definition is that use of a placebo describes what happens when you do nothing, so that in the context of a clinical trial, for instance, a placebo group could describe the natural history of a disorder without the intervention under test. It is the effect we see when we do nothing.

Positive predictive value (+PV)

The proportion of people with a positive test who have disease.


This employs objectivity and is characterised by single truths; observation; measurable features and utilises quantitative methodologies. Its language is that of numbers.

Post-marketing surveillance

A procedure implemented after a drug has been licensed for public use, designed to provide information on the actual use of the drug for a given indication, and on the occurrence of side effects and adverse reactions. It’s a method for epidemiologic study of adverse drug reactions.

Post-test probability

The probability that a patient has the disorder of interest after the test result is known.

Pre-test probability

The probability that a patient has the disorder of interest prior to administering a test.


The baseline risk of a disorder in the population of interest.


Prevention refers to measures taken by an individual or a society to prevent disease happening or its consequences. In general, prevention includes a wide range of interventions, aimed at reducing risks to health. These are grouped into three categories:

Primary prevention: refers to strategies used to prevent a disease happening in the first place. An example may be salt reduction to prevent an individual becoming hypertensive. Medication can be used in primary prevention such as the use of blood lowering or cholesterol lowering drugs to lower the risk of a stroke or heart attack.

Secondary prevention: refers to strategies used in those with an existing disease which prevent recurrence, or significant morbidity. For example, in someone who has a heart attack cholesterol lowering drugs are used to lower the risk of subsequent heart attack and death.

Tertiary prevention: refers to the prevention of long term chronic disease progression, physical deterioration and attendant suffering. For example, removing allergens which may aggravate asthmatic patients; screening for eye, renal, eye, and foot problems among diabetics to reduce the risks of complications.


The prospect of survival and recovery from a disease as anticipated from the usual course of that disease, or indicated by special features of the case.

Prognostic cohort study

This was in the CEBM glossary without anything against it

PsycINFO (Psychology Abstracts)

The PsycINFO (Psychology Abstracts) database provides extensive international coverage of literature on psychology and allied fields. PsycINFO covers psychological practice and research as well as the related clinical, social and biological disciplines. This includes information on drug and behavioural therapy, treatment of disease, drug addiction, developmental psychology, and educational psychology, as well as the psychological aspects of such areas as linguistics, social processes, pharmacology, physiology, nursing, education, anthropology, business and law. Covers the years 1806 to present and is updated weekly.

Publication bias

A bias in a systematic review caused by incompleteness of the search, such as omitting non-English language sources, or unpublished trials (inconclusive trials are less likely to be published than conclusive ones, but are not necessarily less valid).



Qualitative research is used to explore and understand people's beliefs, experiences, attitudes, behaviour and interactions. It generates non- numerical data, e.g. a patient's description of their pain rather than a measure of pain. In health care, qualitative techniques have been commonly used in research documenting the experience of chronic illness and in studies about the functioning of organisations. Qualitative research techniques such as focus groups and in-depth interviews have been used in one-off projects commissioned by guideline development groups to find out more about the views and experiences of patients and carers.

Qualitative approach

Qualitative research or methodology does not try to quantify anything or use statistical methods - it's not about counting, but about words and their meaning.

It provides information about the “human” side of an issue –  behaviours, beliefs, opinions, emotions, and relationships of individuals. The most common method used to generate data in qualitative research is an interview which may be structured, semi-structured or unstructured. Focus groups and case studies are other examples.


Quantitative research generates numerical data or data that can be converted into numbers, for example clinical trials or the National Census, which counts people and households.

Quantitative approach

A quantitative approach is more logical and data-led and provides a measure of what people think from a statistical and numerical point of view. It is used to quantify the problem by way of generating numerical data or data that can be transformed into useable statistics.

Quasi-experimental evaluation

A quasi-experiment is an observational study in which the subjects to be observed are not randomly assigned to different groups in order to measure outcomes, as in a randomized experiment, but grouped according to a characteristic that they already possess


Randomized controlled trial (RCT)

An epidemiological experiment in which subjects in a population are randomly allocated into groups, usually called study and control groups, to receive or not receive an experimental preventive or therapeutic procedure, manoeuvre or intervention. The results are assessed by rigorous comparison of rates of disease, death, recovery or other appropriate outcomes in the study and control groups.

Relative risk (RR or risk ratio)

The ratio of the risk of an event in the experimental group compared to that of the control group (RR=EER / CER). Not to be confused with relative risk reduction.

Relative risk reduction (RRR)

The percentage reduction in events in the treated group event rate (EER) compared to the control group event rate (CER): RRR = (CER-EER) / CER.


The extent to which the method used to collect data about a variable, produces consistent results from the same subjects, reproducible at different times when there is no evidence of change in the subjects


The systematic collection, analysis and interpretation of data to answer a particular question in order to generate new knowledge by establishing a truth.

Research evidence

Studies can be categorised according to the questions they seek to answer, and the evidence for evidence-based commissioning could include the following study types:
Descriptive: to identify the qualities and distributions of variables
Taxonomic: to compare and classify variables into related groups or categories.



Participants of a study recruited from part of the study's target population. If they are recruited in an unbiased way, the results from the sample can be generalised to the target population as a whole.


The way participants are selected for inclusion in a study.

Sampling frame

A list or register of names used to recruit participants to a study.


The proportion of people with disease who have a positive test.

Social Care Institute for Excellence (SCIE)

Social Care Institute for Excellence is an independent charity working with adults, families and social work services across the UK, including health care and housing. They gather and analyse knowledge about what works and translates it into practical resources, learning materials and services including training and consultancy.

Social Science Research Network (SSRN)

Social Science Research Network (SSRN) is devoted to the rapid worldwide dissemination of social science research and is composed of a number of specialized research networks in each of the social sciences.


The proportion of people free of a disease who have a negative test.

Standard deviation

A measure used to summarise numerical data and describe how 'spread-out' a set of measures (or values) are from the average. For example, the average height of a group of schoolchildren can be calculated using the total of all their heights added together and then divided by the number of schoolchildren in the group. Standard deviation measures the 'spread' of those heights. So, in the example it tells you whether all those in the group were about the same height, or whether some were very tall and some were short.

Statistical power

The ability of a study to demonstrate an association or causal relationship between two variables (if an association exists) means that the study is statistically significant. The statistical power of a study is primarily related to the number of people included. If too few people are included, any differences in the outcomes will not be statistically significant.

Structure measure

“measures of organizational characteristics (such as staffing ratios, number of hospital beds)” Or “Denotes attributes of the setting in which care occurs”

Donabedian (1997)

Systematic review

The application of strategies that limit bias in the assembly, critical appraisal and synthesis of all relevant studies on a specific topic. Systematic reviews focus on peer-reviewed publications about a specific health problem and use rigorous, standardised methods for selecting and assessing articles. A systematic review differs from a meta-analysis in not including a quantitative summary of the results.


Treatment benefits

Positive patient-relevant outcome associated with an intervention, quantifiable by epidemiological measures such as absolute risk reduction (ARR) and number needed to treat (NNT).

TRIP Database

The TRIP Database is a medical search engine with emphasis on evidence-based medicine (EBM) and clinical guidelines and queries, including content from Cochrane and Bandolier. Designed to find and use high-quality clinical research evidence from medical literature for clinical practice. Sign-up to Trip to save searches, get tailored results sent to you, and to see what articles are available in NBT libraries or with an NBT online access.



The extent to which a variable or intervention measures what it is supposed to measure, or accomplishes what it is supposed to accomplish. The internal validity of a study refers to the integrity of the experimental design. The external validity of a study refers to the appropriateness by which its results can be applied to non-study patients or populations.


Observable, measurable (or categorisable) characteristics that vary among cases in a study e.g. height, weight, blood pressure, gender, eye colour and length of hospital stay.

In an experiment, the independent variable is the variable that is varied or manipulated by the researcher, and the dependent variable is the response that is measured. An independent variable is the presumed cause, whereas the dependent variable is the presumed effect.


York Centre for Reviews and Dissemination (CRD)

The York University Centre for Reviews and Dissemination databases provide access to systematic reviews, economic evaluations, summaries of completed and on-going health technology assessments (HTAs) and summaries of all Cochrane reviews.

The Evaluation and Evidence toolkits go hand in hand. Using and generating evidence to inform decision making is vital to improving services and people’s lives.

About the toolkits