Glossary
The list below contains definitions for terms used in the AQASR. We also have a list of other terms related to assessing quality of systematic reviews that are not used in the AQASR.
| Term | Description/definition |
|---|---|
| Abstract | An abstract is a concise and brief summary that provides an overview of a research paper, thesis, article, or academic work. It serves as a condensed representation of the essential elements of the study, allowing readers to quickly understand the purpose, methods, key findings, and significance of the research without having to read the entire document. |
| Abstract review | In systematic reviewing, the review of abstracts is an intermediary step leading from all references produced by a literature search using bibliographic databases to a final set of reports to be included in the full text review. |
| Abstract reviewers | Research professionals who review the abstracts of articles and documents generated by literature searches to determine whether they qualify for further review in the full text review stage. |
| Adverse (health) outcomes | Negative conditions attributed to an intervention or to other clinical actions examined in the research reviewed. Also called adverse effects. |
| Agency for Healthcare Research and Quality (AHRQ) | The Agency for Healthcare Research and Quality (AHRQ) is the lead Federal agency focused on improving healthcare in the areas of quality, safety, efficiency, and effectiveness. AHRQ promotes evidence-based decision making and supports research that improves the quality of healthcare.. The agency is active in supporting evidence-based practice and evidence development methodologies. Website: https://www.ahrq.gov |
| AGREE II | The AGREE II tool is used to assess the quality of practice guidelines. The AGREE Enterprise is an international network of researchers and policy makers that work to improve clinical practice guidelines through the development of a reporting and assessment framework to assess quality and effectiveness. Website: https://www.agreetrust.org |
| American Academy of Neurology (AAN) | AAN is an international professional association consisting of neuroscience professionals and clinicians that are dedicated to promoting quality patient-centered neurologic care. The AAN has a clinical practice guideline development process that has often been used by others, including rehabilitation systematic reviewers. Website: https://www.aan.com |
| Ancestor search | Ancestor search refers to analyzing the reference lists of articles identified from an electronic search, or of other (systematic) reviews in the area of interest, to identify earlier potential primary studies. Some bibliographic databases (e.g., CINAHL) include the references of the journal articles they index and allow for electronic searches of these ancestors. |
| Applicability | The degree to which something is relevant, suitable, or can be appropriately used or applied to a particular context, problem, or set of circumstances. |
| Attrition | The loss of participants over time in a longitudinal study, reducing the statistical power and possibly introducing bias, because attrition is likely to be selective. |
| Attrition bias | Bias resulting from the fact that dropout of subjects in a long-term study is almost always selective. The disappearance of certain subgroups more than others (e.g., males more than females; healthy patients more than unhealthy) may confound the study findings. Intent-to-treat analysis may be an appropriate counter to attrition bias. |
| Australian New Zealand Clinical Trials Registry (ANZCTR) | The Australian New Zealand Clinical Trials Registry (ANZCTR) is an online register of clinical trials being conducted in Australia and New Zealand. Website: https://www.anzctr.org.au |
| Benefits | The expectation of receiving a gain from the treatment or intervention studied. Benefits can occur in the mental, physical, economic, and/or social arenas. |
| Best-studies analysis | A variation of sensitivity testing in which the pooled effect size calculation is repeated, but using only those studies that exceed a cutoff level for study quality. |
| Bias | A systematic error or deviation in results or inferences. In systematic reviewing, the concern is both with bias in individual studies (e.g., selection bias, performance bias, attrition bias, detection bias, etc.), and with biases created by selective reporting of studies (publication bias) and of findings (publication bias in situ, selective outcome reporting).Conflicts of interest and pre-existing preferences for certain interventions, diagnostic tests, etc., may result in biases that correspond to the conventional use of the word in which bias refers to a partisan point of view. See also methodological quality. |
| Bibliographic and other databases | Searchable electronic resources, available for free (e.g., PubMed) or for a fee (e.g., CINAHL), that contain abstracts and other key bibliographic information indexed using a predetermined set of criteria such as subject matter, keywords, or other descriptive terms representing the content of the record of publications in a particular area of science or practice, or a subset selected based on journal quality or other criteria. Materials include records of published studies including books, articles, and abstracts, conference presentations, research reports, educational materials, and advocacy resources and more. Bibliographic databases usually store collections of bibliographic records in a structured way and have various search options, including author name, keyword, thesaurus term. Major bibliographic databases relevant to disability and rehabilitation researchers include PubMed (MEDLINE), PsycINFO, CINAHL and Embase. |
| Black and Downs | The Black and Downs “Checklist for Measuring Quality” is a tool to assess the quality of original or primary source research articles and to synthesize evidence from quantitative studies for public health practitioners, policymakers, and decision makers. |
| Blinding | Keeping secret group assignment (e.g., to treatment or control, or being positive or negative on the reference (gold standard) diagnostic test) from the study participants (“single blind”) or investigators (“double blind”). Blinding is used to protect against the possibility that knowledge of assignment may affect patient response to treatment, provider behaviors (performance bias) or outcome assessment (detection bias) by outcome assessors (“triple blind”). |
| Boolean operators | A set of logical operators, such as a symbol or word, used to indicate relationships between thesaurus terms or keywords. The operators AND, OR, and NOT are used to formulate search commands in electronic databases, as well as to either broaden or narrow the retrieved results of a search. |
| Campbell Collaboration | The Campbell Collaboration was established in 2000 as an international research network that “promotes positive social and economic change through the production and use of systematic reviews and other evidence synthesis for evidence-based policy and practice.” There are 11 Coordinating Groups that provide editorial services and support to review authors and are responsible for the relevance and scientific merit of the systematic reviews produced under their guidance. There are two National and Regional Centres that are geographically based. |
| Ceiling effect | The phenomenon that a measurement cannot take on a value higher than some limit or “ceiling,” which is imposed not by the phenomenon being measured, but rather by the finite nature of the measuring instrument. Opposite of floor effect. |
| CINAHL | CINAHL®, the Cumulative Index to Nursing and Allied Health Literature, provides indexing for almost 3,800 English-language journals covering the fields of nursing and allied health disciplines, such as health sciences, biomedicine, and consumer health. CINAHL Ultimate includes over 900 full-text journals. Website: https://www.ebsco.com/products/research-databases/cinahl-database Ultimate: https://www.ebsco.com/products/research-databases/cinahl-ultimate |
| Citation records | Documentation of published and unpublished information that includes author, title, source, and publication date (and sometimes abstract) needed to locate or identify referenced notations |
| Classical test theory | A set of theoretical notions on the proper ways of developing psychometric measures and assessing their key metrological characteristics, such as reliability and validity. Classical test theory may be regarded as roughly synonymous with true score theory. The term “classical” refers to these theories and methods having been developed prior to more recent psychometric theories, generally referred to collectively as item response theory, which sometimes are called “modern” as in “modern latent trait theory.” |
| Clinical practice guideline | Systematically developed statements to assist practitioners and patient decisions about appropriate health care for specific circumstances. |
| Clinical trials | Clinical trials are studies designed to assess the efficacy or effectiveness of an intervention under controlled or laboratory conditions (as opposed to wide scale application of the intervention under study to the population as a general practice). |
| Clinical trials register | Publicly available database of interventional (clinical) trials. Clinical trial registers describe intervention studies that are completed or in progress and allow one to identify studies that have not been published, possibly because of negative results. |
| Clinical utility | Clinical utility refers to the practical usefulness and relevance of a medical test, diagnostic procedure, treatment, or intervention in a clinical setting. It assesses the extent to which a medical intervention or information obtained from a test or procedure benefits patient care and decision-making. |
| ClinicalTrials.gov | A website and database that provide researchers, families, patients, and health care professionals access to information from clinical research studies on a wide range of health topics. Website: https://clinicaltrials.gov |
| Cochran’s Q and I-squared | Cochran's Q test and I-squared (I²) statistic are statistical methods used to assess and quantify the degree of heterogeneity among the results of individual studies included in a meta-analysis. These methods help researchers determine whether it is appropriate to pool the results of multiple studies and calculate an overall effect size or summary statistic. |
| Cochrane (formerly Cochrane Collaboration) | Cochrane (established in 1993 as The Cochrane Collaboration) is an independent, international organization consisting of a network of professionals, researchers, caregivers, and patients focused on producing trusted synthesized evidence, ensuring that it is accessible to all. Website: https://www.cochrane.org |
| Cochrane Central Register of Controlled Trials | The Cochrane Central Register of Controlled Trials (CENTRAL) provides a comprehensive source of reports of randomized and quasi-randomized controlled trials. About: https://www.cochranelibrary.com/central/about-central |
| Cochrane Library | The Cochrane Library, owned by Cochrane, is a collection of databases that contain a variety of high-quality, independent evidence to inform health care decision making. Website: https://www.cochranelibrary.com |
| Cointerventions | In a randomized controlled trial, the application of additional therapeutic procedures to members of either or both the experimental and the control groups. The cointerventions may either be part of the study or searched out by subjects outside the research. |
| Comparator | A drug or another intervention element used instead of the traditional placebo control mechanism to assess the effectiveness of treatment in clinical trials. A comparator drug or other intervention is required to prove superiority of the intervention of interest to existing treatments. In systematic reviews of interventions, the intervention with which the treatment of interest is being compared. The comparator may be “nothing,” waiting list, sham, usual care, the traditional treatment, a specific alternative treatment, etc. In systematic reviews of diagnostic tests or assessment instruments, the comparator may be an alternative (reference, gold standard) test or assessment. |
| Concealment (allocation concealment) | The process used to prevent foreknowledge of group assignment in a randomized-controlled trial, until the subject has fully consented and has been determined to be qualified to participate based on inclusion and exclusion criteria. This prevention sometimes is extended (in research with a placebo or sham) until treatment and all follow-ups for outcome assessment have been completed. Concealment is the means to achieve subject blinding. |
| Confidence intervals | The range within which the “true” value (e.g., size of effect of an intervention) is expected to lie with a given degree of certainty (e.g., 95% or 99%). The confidence interval is expressed in the same units as the estimate. Wider intervals indicate lower precision; narrow intervals indicate greater precision. Just like confidence intervals can be calculated for primary studies, they can be calculated for the “average” effect size calculated in a meta-analysis. Note that confidence intervals represent the probability of random errors, but not of systematic errors (bias). |
| Conflicts of interest | In systematic reviewing, conflict of interest refers to a systematic reviewer (or the organization that sponsors the review) having a financial or other interest in a treatment or diagnostic tool being evaluated. Even though the protocol-specified rules for conducting the systematic review are designed to preclude such interests from affecting the findings, there almost always are opportunities for such interests to result in biases. |
| Confounder | A confounder (also confounding factor, a confound, or confounding variable) is an extraneous variable in a statistical model that correlates (positively or negatively) with both the dependent variable and the independent variable. Studies therefore need to control for these factors to avoid a false positive (Type I) error; an erroneous conclusion that the dependent variables are in a causal relationship with the independent variable. Such a relation between two observed variables is termed a spurious relationship. |
| Confounding | A situation in which a measure of the effect of an intervention or exposure is distorted because of the association of exposure with other factor(s) that influence the outcome under study. |
| CONSORT | CONSORT (Consolidated Standards of Reporting Trials) involves a variety of initiatives developed by the CONSORT Group to improve the problems occurring from inadequate reporting of randomized controlled trials (RCTs). The CONSORT Statement is evidence-based and comprised of a 25-item checklist including a minimum set of recommendations for reporting. CONSORT can be accessed through the EQUATOR website: https://www.equator-network.org/reporting-guidelines/consort/ |
| Construct validity | Whether a scale measures or correlates with the theorized psychological scientific construct that it purports to measure. In other words, it is the extent to which what was to be measured was actually measured. |
| Consumer price index | A measure of price inflation, determined by calculating the price of a market basket of goods and services at a specified time point relative to the price in a base year. |
| Contacting experts and/or prominent authors | Many articles and documents that could be relevant to a particular systematic review are not readily found, because they are not indexed in an electronic database, or are misclassified. Grey literature documents (e.g., conference presentations, monographs) may be even more difficult to find. Contacting known experts and authors is a means used to locate and acquire these more difficult to find documents. |
| Controlled vocabulary terms | A collection of terms that provides a way to organize knowledge for subsequent retrieval. Used in subject indexing schemes, subject headings, and thesauri. Each concept from the domain of discourse is described using only one term and each term describes only one concept. A selection of the terms is made when cataloging, abstracting and indexing; or when searching books, journal articles or other documents. The control is intended to avoid the scattering of related subjects under different headings. The list may be altered or extended only by the publisher or issuing agency In bibliographic databases, the controlled vocabulary terms may be called Medical Subject Headings (MeSH terms, in PubMed) or thesaurus terms (in CINAHL). |
| Convergent validity | The degree to which a measure provides data similar to (converges on) other measures that it theoretically should also be similar to. High correlations between the scores of two measures of the same characteristic would be evidence of a convergent validity. It is ideal that scales rate high in discriminant validity as well, which unlike convergent validity is designed to measure the extent to which a given measure differs from other scales designed to measure a different concept. Discriminant validity and convergent validity are the two good ways to measure construct validity. |
| Cost-benefit, Cost-benefit analysis | A technique for measuring net gain or loss to society of a new program or project. It considers allocative efficiency. Values of benefits are usually given in monetary terms. |
| Cost-effectiveness, Cost-effectiveness analysis | A technique for comparing alternative approaches to care, using metrics such as cost per life-year gained. Originally derived to assess the technical efficiency. |
| Cost-utility, Cost–utility analysis | A technique for comparing the costs and the utility of health gained for different alternatives, such as cost per quality-adjusted life-year gained. |
| Data extractors | In systematic reviewing, individuals (generally with training in research methods and a particular clinical field) who (after training) systematically review journal papers and other reports of primary studies and extract information needed for the review. See data extraction. |
| Data extraction | In systematic reviewing, the process of selecting from the reports of primary studies information on the nature of the studies and on their findings, and entering this information on extracting forms, directly into a custom database, or directly into an evidence table. |
| Data synthesis | A designated methodology for combining the results of a set of studies. Data synthesis can be either qualitative or quantitative (meta-analysis). |
| Database bias | Database bias occurs when research papers and other information indexed for a particular database varies systematically from the non-indexed studies. |
| Deduplication | The process of identifying and removing duplicate records or entries from the pool of potentially relevant studies or articles gathered during the initial literature search. Duplicates typically arise when searching multiple databases, search engines, or journals. |
| Descendant search | A search for later papers that cite primary studies or reviews that have been identified as relevant to a systematic review. The only feasible way of doing such a search is using the Web of Science, |
| Deviations (from the protocol) | In systematic reviewing, departures from the pre-established protocol, whether acknowledged or not. Deviations may be fully justifiable and improve the study’s results, but they should be described. |
| Diagnostic (test) study | Research that aims to determine the diagnostic accuracy of a diagnostic test. |
| Diagnostic accuracy | The ability of a diagnostic test (as used by a clinician with a certain skill level) to classify patients correctly into diseased vs. non-diseased. Most commonly, accuracy is determined by comparing the results of an index test with a reference standard, which may be another test, or a patient outcome that can be reliably tied to the disease the index test aims to establish. |
| Diagnostic accuracy studies | Studies performed to assess the ability of a diagnostic instrument to differentiate between patients who are positive (have a condition of interest) and those who are negative. |
| Diagnostic test or instrument | Any (laboratory) test, interview, etc. designed to establish that a person has a disorder. |
| Diagnostic test | A method to assess a patient, using a combination of human (e.g., components of a physical examination) and/or machine (whether processed automatically or “read” by a human, as in X-rays) evaluation, that results (most typically) in a binary judgment of diseased (case) vs. not diseased (not a case) |
| Disability-adjusted life year (DALY) | The number of healthy years of life lost due to disability. Originally developed by the World Health Organization, this measure of disability burden is becoming increasingly common in the field of public health and health impact assessment. See Quality-adjusted life year. |
| Discount rate | A technique for estimating the present value of costs and benefits occurring in different time periods. |
| Disposition | Refers to the decisions made at time of abstract review and full-paper review. This denotes whether a document will be included/excluded from additional review (after the abstract reviewing stage) or for inclusion in the systematic review analysis and report. |
| Divergent validity | The degree to which the operationalization of a construct is not similar to (diverges from) other operationalizations that it theoretically should not be similar to. The opposite of convergent validity. |
| Economic | Having to do with measures of cost of production, delivery, or benefit from actions taken. In research it typically reflects the cost to change an outcome/behavior. |
| Effect size | A dimensionless quantitative measure of the strength of the relationship between two variables, whether intervention and outcome, prognostic factor and outcome, etc. Pearson correlation, Cohen’s d and Glass’s delta are all effect size measures, among many available. |
| Electronic databases | Bibliographic databases that contain references to published literature that are organized in some systematic way so that a search for desired documents can be done. Information that can be retrieved includes a reference to where documents can be found. Frequently article abstracts are also provided, and in some instances, full text documents can be obtained directly from the database. Such databases include MEDLINE (PubMed), PsycINFO and RehabData. (See bibliographic and other databases) |
| Eligibility | The criteria or standards used to determine which studies or research articles are suitable for inclusion in the review. Eligibility can have a variety of limiting factors. For instance a researcher may want to limit the review to a particular outcome, timeframe, study design, or population. |
| EMBASE | Excerpta Medica Database (EMBASE) is a bibliographic database with citation records indexing pharmacological and biomedical publications and information dating from 1947. EMBASE covers much of the European medical literature not indexed by MEDLINE. Website: https://www.embase.com |
| EU Clinical Trials Register | A register of interventional clinical trials on medicines in the European Union (EU) and European Economic Area (EEA). It was established in 2004 and has been a primary registry in the WHO’s Registry Network since 2011. |
| Evidence | In evidence-based practice (EBP), the generic term for all research-based and experiential published or unpublished information that informs (or might be used to inform) decisions by researchers, clinicians or other practitioners. |
| Evidence grading | The classification of evidence into a hierarchy from weakest (e.g., expert opinion, case studies) to the strongest (e.g., large randomized controlled trials with adequate concealment and blinding). The hierarchy is different for diverse clinical questions (treatment, diagnosis, etc.) because of the study designs that are possible and optimal for these questions, and various organizations have developed variations of the schemes proposed when EBP first developed. |
| Evidence table | Tabular presentation of the relevant points from a set of primary studies included in a systematic review. The tables could summarize the sample size, description of the sample population, outcome measures, major results, and limitations. |
| Evidence-based practice | Evidence-based practice is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients or clients. Evidence-based practice means integrating individual clinical expertise and patient/client values with the best available external clinical evidence from systematic research. |
| Extracting | In performing a systematic review, extracting is selecting key information from a primary study and entering it into an evidence table or database for further (statistical) processing. |
| Extracting form | A form customized for a particular systematic review on which data extractors are to enter specific data elements gleaned from the reports of primary studies. See data extraction. |
| Fail-safe N | The number of studies with a negative finding (“no correlation”) that should exist in file drawers to wash out the combined effect of the studies with positive findings that were found in published research. The concept and calculation were developed by Rosenthal (1979). His method calculates the number of additional studies, NR, with mean null result necessary to reduce the combined significance to a desired alpha level (usually 0.05). |
| False positive | In diagnostic accuracy studies, a case that is designated positive by the index test but negative by the reference standard. |
| Fixed effects model | A fixed effects research model assumes that the patients selected for a specific treatment have the same true quantitative effect of the treatment and that the differences observed are residual error. If, however, there is reason to believe that certain patients respond differently from others, then the spread in the data is caused not only by the residual error but also by between-patient differences. |
| Floor effect | Floor effect is the result of a large proportion of subjects scoring the minimum, such that it is not possible to discriminate among those subjects scoring at the lower end of the measure. |
| Flow diagram | A flow diagram shows from beginning to end the steps involved in finding studies to be included in a systematic review, and the number of abstracts or full papers that were found (by source) and included/excluded in next stages, by reason for exclusion. |
| Forest plots | A graphical plot typically consisting of two columns that display the strength of treatment effects from a set of comparable studies of a specific problem or research question. The left column typically contains a list of the relevant studies in chronological order and the right column plots the effect size with the 95% confidence interval for each of the studies. A vertical line representing “no effect” also is commonly shown. |
| Formal tests of agreement | Statistical tests that are used to determine how well raters agree, and sometimes referred to as showing the reliability of raters. The statistics that result sometimes are percentages, indicating how often exact agreement between or among raters occurred. Frequently, 90 percent agreement is expected. The statistics can also be correlation coefficients. Minimum correlations expected are typically around .70. Kappa and weighted kappa as well as the intraclass correlation coefficient are also used to quantify agreement. The agreement in question can be on the inclusion-exclusion of an abstract, the inclusion/exclusion of a full paper, or the presence or absence of particular features of the studies described, e.g., blinding of patients. The various tests are typically used to assess the agreement between two raters but can be used with more than two. |
| Free text term | A word or group of words used by authors in their abstract or full text that can be used to search for particular studies. Also called keywords, free text terms are to be distinguished from thesaurus terms, or controlled vocabulary terms. |
| Full paper | The complete, full text document describing a study, as opposed to the abstract of the study, which may be all that is included in a bibliographic database. Web-based supplemental digital information may be considered part of the full papers used by systematic reviewers. |
| Funding bias | Funding bias occurs when the conclusions of a study get biased towards the outcome the agency funding the research wants. Funding bias can occur in systematic reviews as well as in primary studies. |
| Funnel plot | A graph plotting the effect size against the sample size for all studies relevant to a particular clinical question. In the absence of publication bias, the plot is symmetrical around the average effect size. If there is publication bias, there is a “hole” in the upside-down funnel (or “Christmas tree”) where small studies with negative results should have been. |
| Generalizability | Generalizability is the application or extension of the results and conclusions from a sample of participants to the population represented in that sample. In a systematic review, generalizability refers to the degree to which the recommendations/results can be applied to different populations, different demographic groups, different interventions, different outcome measures than the ones included in the primary studies that were reviewed. The applicability of the findings of a systematic review needs to be restricted to populations with similar characteristics to the ones studied in the review. |
| Google Scholar | A Google program that allows one to search for articles, theses, books, abstracts and court opinions from academic publishers, professional societies, online repositories, universities and other web sites. In addition, later scholarly product(s) that cited each index paper or document are identified. Website: https://scholar.google.com |
| GRADE | The Grading of Recommendations Assessment, Development and Evaluation is a comprehensive approach to grading the quality of evidence and the strength of recommendations that is recommended by Cochrane. The approach specifies four levels of quality of the evidence from research studies: high, moderate, low, and very low. Website: https://www.gradeworkinggroup.org |
| Grey literature | Grey literature refers to papers, reports, technical notes, white papers, or other documents produced and published by governmental agencies, academic and other research institutions and other groups that are not distributed or indexed by commercial publishers. Many of these documents are difficult to locate and obtain. Founded in 1992, the Grey Literature Network Service (GreyNet International) facilitates dialog among persons and organizations in the field of grey literature. Website: https://www.greynet.org |
| Hand searches | In systematic reviewing, the practice of manually going page by page through hardcopy versions of the journals that are of key relevance to the clinical question, in order to find articles that may have been missed or misclassified by the indexers used by bibliographic databases. |
| Harm | Adverse effects resulting directly from or associated with the administration of the treatment or intervention studied. Harms can occur in the mental, physical, economic, and/or social arenas. |
| Heterogeneity | In systematic reviewing, a degree of variation in the effect sizes of all the studies addressing a particular question that cannot be explained as the result of the random sampling used in the individual studies. Formal statistical tests of heterogeneity are available (such as Cochran’s Q test and the I-squared statistic; if the tests are positive, meta-analysis will need to use the random effects model, or a more qualitative synthesis is the only step possible. The opposite of heterogeneity is homogeneity. |
| Heterogeneity (sample) | The degree to which cases in a sample differ significantly on one or more key variables. |
| Heterogeneous | Consisting of dissimilar elements or parts; for example, different age groups within a diagnostic group. In systematic reviewing, a set of studies addressing the same question may be called heterogeneous if differences in their methods or outcomes make it impossible to statistically combine them in a meta-analysis. |
| Homogeneity (sample) | The degree to which cases in a sample are very similar to one another on one or more key variables. |
| Homogeneous | Consisting of similar elements or parts; for example, two separate studies that examine an intervention in individuals with mild traumatic brain injury who have similar demographics. In systematic reviewing, a set of studies addressing the same research question using the same methods, which come up with very similar findings. The opposite of homogeneous is heterogeneous. |
| Inclusion and exclusion criteria | When referring to a primary study, criteria that are set prior to selection of research participants to guide who will actually be recruited to take part in the research. The criteria typically consist of demographic variables, such as age, and medical condition. When referring to a systematic review, criteria that are set prior to selecting articles and documents for the review to ensure the right studies are included. These criteria can refer to the content of the article, such as the intervention studied, the time frame during which the study was done, and the population studied, as well as aspects such as language and peer review status. |
| Index test | A test where accuracy is being evaluated in a diagnostic test accuracy study, most commonly by comparison with the reference standard. |
| Indexed | Organization of the content of a document by keywords. Also, the feature of the search engine that allows optimizing speed and performance to find documents relevant to a search query. |
| Indexer | A person working for a bibliographic database who characterizes study reports and other published articles in terms of their method, population, health problem addressed and other topic issues. |
| Intent-to-treat (ITT) analysis | ITT analyses are based on the initial treatment intent, not on the treatment administered. ITT analysis is designed to avoid misleading artifacts that arise in intervention research. All subjects who begin the treatment are considered to be part of the study, whether they finish it or not, and whether they received the correct treatment (see treatment integrity) or any treatment or none. ITT can be contrasted with per-protocol analysis. |
| International Clinical Trials Registry Platform (ICTRP) | The WHO International Clinical Trials Registry Platform ensures that a complete view of research is accessible in order to improve research transparency and strengthen the validity and value of the scientific evidence base. About: https://www.who.int/clinical-trials-registry-platform Search: https://trialsearch.who.int |
| Intervention | The treatment procedure, approach or technique that is under study. It is typically compared to no intervention (control group) or an existing intervention under controlled research conditions. In a systematic review, the intervention is the focus of the review. |
| ISRCTN Registry | ISRCTN is a worldwide registry and curated database. Originally 'International Standard Randomised Controlled Trial Number,' over the years the scope of the registry has widened to include any study designed to assess the efficacy of health interventions in a human population. Website: https://www.isrctn.com |
| Item Response Theory | A paradigm for the design, analysis, and scoring of tests, questionnaires, and similar instruments measuring abilities, attitudes, or other variables. Also known as latent trait theory, strong true score theory, or modern mental test theory. |
| Jadad | Jadad and colleagues (1996) developed a scale that is used to assess the risk of bias, an important to the methodological quality of clinical trials. The items scored include randomization, blinding, and characteristics of study dropouts. |
| Keywords | Informative words or terms that pertain to the main search goal, topics or ideas of a systematic review and are used to perform bibliographic database or hand searching. Sometimes called “free text terms.” The quality of a search query depends on the precision of keywords used. |
| Knowledge base | Research reported to date on the subject being addressed in the systematic review, including a specification of area(s) in which there are gaps |
| Knowledge Translation | Knowledge Translation (KT) is the process of taking research findings, scientific knowledge, or expertise and making it accessible, understandable, and applicable to specific audiences beyond the academic or research community. |
| L’Abbé plot | L’Abbé plots show variations in observed results by plotting the event rate in the treatment group on the vertical axis and in the control group on the horizontal axis. Useful for assessing potential sources of heterogeneity in meta-analysis. |
| Language bias | Language bias refers to the systematic selection or rejection of research or information published in a particular language (e.g., including only studies published in English when appropriate research for a topic is available in a non-English language). This may be problematic because there is evidence that the quality of research and the outcomes of research published in English as opposed to in other languages may not be comparable. |
| Level I | “Level I” is the traditional designation of the highest level of study quality in an evidence grading hierarchy. (Also known as class I.) |
| Level of agreement | Most formal tests of agreement have an algorithm that results in the level of agreement being expressed on a scale that ranges from 0.0 (no agreement at all) to 1.0 (perfect agreement). |
| LILACS | A database of Latin American and Caribbean Health Sciences Literature Website: https://lilacs.bvsalud.org/en/ |
| Limitations | For primary studies, study limitations are provided by authors to inform the reader of how generalizable the results should be. These are characteristics of design or methodology that impacted or influenced the interpretation of the findings from the research. Systematic reviewers, through their careful scrutiny of multiple studies in the same area, may identify additional limitations in the primary studies, which likely inform the conclusions of the systematic review. For systematic reviews, study limitations can be the result of the limitations in the primary studies. Other limitations can result from explicit choices the reviewers make, such as exclusions of primary studies based on language, publication in a peer-reviewed journal, etc. |
| Literature search | In systematic reviewing, the protocol-steered process of systematically identifying published and unpublished research of relevance to a clinical question, using searches of bibliographic databases, ancestor searches, communication with experts, etc. |
| Manualized | Experimental behavioral and similar interventions that are delivered based on an extensive set of instructions that are documented carefully are referred to as “manualized,” because they are described in a manual used for training therapists and for checking treatment integrity. |
| Measurement instrument | Measurement instruments, and formal test methods which define the instrument's use, are the means by which the physical characteristics of objects and events can be recorded and compared. |
| Medical consumer price index | An index of consumer prices that includes only “medical care commodities” and “medical care services” |
| MEDLINE | MEDLINE is the National Library of Medicine’s journal citation database. Started in the 1960s, it now provides more than 29 million references to biomedical and life sciences journal articles dating back to 1946. MEDLINE includes citations from more than 5,200 scholarly journals published around the world. MEDLINE is searched via PubMed: https://pubmed.ncbi.nlm.nih.gov |
| MeSH | MeSH (Medical Subject Headings); a set of subject headings the National Library of Medicine uses to designate the subject matter of articles in the database. |
| Meta-analysis | A quantitative statistical procedure that combines the results of several studies that address the same question. This is normally done by identification of a common measure of effect size and other parameters that are more precise and less likely to be in error (due to sampling) than the individual studies being reviewed. |
| Meta-regression | A regression analysis in which the unit of analysis is the study (or subgroup in a study) rather than the individual, as in primary studies. The predictor variables can be characteristics of the study as a whole (e.g., number of hours of treatment specified in the study protocol) or attributes of the groups studied (e.g., percent female in each sample). |
| Methodological quality | In systematic reviewing, the term used for the overall quality of a research project, based on design and in some cases, implementation of the investigation. In most evidence grading schemes, four to ten levels of studies are distinguished, based primarily on strength of the research design. |
| Methodologist | A researcher with special expertise in one or more areas of research methodology. |
| Metric | A system or standard of measurement |
| Metrologic | Metrologic refers to the science of measurement. Metrology includes all theoretical and practical aspects of measurement. |
| Minimal clinically important difference | The smallest change in their status which patients perceive as beneficial |
| Minimal detectable change | The minimal amount of change outside of error that reflects true change by a subject between two time points (rather than a variation in measurement). |
| Missing data, missing values | In systematic reviewing, a parameter describing a study that is not reported in the primary study’s paper/other report and cannot be calculated – e.g., the standard deviation corresponding to the mean of the outcome for the treatment and control group. Sometimes it is justifiable to estimate the missing data point based on other similar studies. |
| Natural language | A common set of terms used for communication across a particular discipline; a human written or spoken language used by a community; as opposed to a computer language or a lexicon of controlled terms, such as the MeSH terms. |
| Nesting | The way terms are grouped within the search query to clarify their relationships. A nesting strategy is most often applied to synonymous terms when the search statement also contains the default AND Boolean Operator. Parentheses can be used to specify the way in which terms in a Boolean expression should be grouped or nested. |
| NeuroBITE | NeuroBITE (previously PsycBITE, 2004-2019) is a database that catalogues studies of cognitive, behavioral and other treatments for psychological problems and issues occurring as a consequence of acquired brain impairment (ABI). Website: https://neurorehab-evidence.com |
| Odds ratio (OR) | The ratio of the odds of an event in the experimental (intervention) group to the odds of the same event in the control group. Odds are the ratio of the number of people in a group with an event to the number without an event. Thus, if a group of 100 people had an event rate of 0.20, 20 people had the event and 80 did not, and the odds would be 20/80 or 0.25. An odds ratio of one indicates no difference between comparison groups. For undesirable outcomes an OR that is less than one indicates that the intervention was effective in reducing the risk of that outcome. When the event rate is small, odds ratios are very similar to relative risks. |
| Operationalizing | The process of specifying the measurement operations that need to be taken to quantify a construct or characteristic. An operational definition defines something (e.g., a variable, term, or object) in terms of the specific process or set of validation tests used to determine its presence and quantity. That is, one defines something in terms of the operations that count as measuring it. |
| OTSeeker | A database of occupational therapy systematic evaluation of evidence. Prior to July 2013, most of the randomized controlled trials in OTseeker were critically appraised using the PEDro scale. Website: http://www.otseeker.com |
| Outcome assessors | The researchers (commonly, research assistants, but sometimes clinicians) who are designated and trained to collect trial outcome information are called outcome assessors. |
| Outcome reporting bias | Research reporting in which authors of primary studies present only the significant results of multiple outcomes considered and none of the non-significant outcomes. |
| Outcome, patient outcome | For a review of treatment(s) or of the economic costs of treatments: the conditions that are influenced by the intervention examined in the research reviewed. For a review of prognostic studies: the patient statuses that are predicted. For a review of diagnostic or assessment instruments: the condition or characteristic the test/measure aims to determine |
| PEDro | The Physiotherapy Evidence Database (PEDro) is a no-cost repository containing reviews, guidelines, and trials all of which evaluate physiotherapy interventions. Founded in 1999, PEDro is currently situated within the Institute for Musculoskeletal Health, which is housed at the University of Sydney and Sydney Local Health District. Website: https://pedro.org.au |
| Per-protocol (PP) analysis | In contrast to intent-to-treat analysis, per-protocol analysis is an approach in which only subjects who complete the trial are included in the final results. Per protocol analysis excludes all cases that drop out, but also all who received an incomplete or erroneous treatment. |
| Perspective | The point of view from which an economic analysis is conducted. An economic evaluation from one perspective (for example, the patient's) may consider the impact of different sets of costs and outcomes than one conducted from another perspective (for example, the insurance company's). Most experts recommend that analyses be conducted from the societal perspective because it considers the broadest range of costs and benefits. |
| PICO | PICO (Patient/Population, Intervention, Comparator, and Outcome) is a method used for structuring clinical questions that allows clinicians to search MEDLINE/PubMed using handheld devices. This format can also be used for structuring literature searches and may be helpful to practitioners and researchers interested in evidence-based medicine. A PICO feature is available on the main screen of PubMed for Handhelds (https://pubmedhh.nlm.nih.gov) and uses a fill-in-the-blank and menu format. Another format that evolved from PICO is askMEDLINE (https://pubmedhh.nlm.nih.gov/ask/) that guides a clinician through the search process by thinking along PICO elements. |
| PICOT/PICOS | PICOT (Population or Patients, Intervention, Comparator, Outcome and Timeframe) and PICOS (Population or Patients, Intervention, Comparator, Outcome and Study type) are formats of a search query. Like the PICO format, PICOT/PICOS provides key words for a literature search of pre-appraised evidence and original research studies that address the clinical scenario, but with additional parameters Time and Study type respectively. |
| Pooling | A term used to represent the combining of raw data from a set of studies (meta-analysis) or the results from a set of studies to generate answers to the posed problem or research question. |
| Power | The probability (generally calculated before the start of a study) that a study will detect a statistically significant effect, if there is one, The prespecified study sample size is often chosen to give the trial the desired power as determined in a power analysis. |
| Power analysis | Formal calculation of the sample size needed in a study to achieve a desired level of power. The calculation involves an estimate of the effect size, as well as specification of the type I and type II errors the researcher is willing to run. |
| Precision | Precision is defined as the number of relevant reports identified divided by the total number of reports identified. A precise search would capture relevant literature, and fewer instances of irrelevant results. It relates to the degree to which the results of the systematic review are accurate, specific, and closely aligned with the research question or objectives. |
| Primary research | In systematic reviewing, the individual studies that the systematic reviewers scrutinize for inclusion in their review. These are studies that directly address the subject/question of a systematic review, and which have collected and analyzed data in a controlled context. Also called primary studies. Performing a systematic review might be termed secondary research. |
| Primary studies | Research presented as an original scientific work based on data collected on humans, animals, plants or other entities, as opposed to secondary studies (such as meta-analyses and other systematic reviews), which are based on the findings of primary studies. |
| PRISMA | PRISMA stands for Preferred Reporting Items for Systematic Reviews and Meta-Analyses. It is an evidence-based minimum set of items for reporting in systematic reviews and meta-analyses. The PRISMA Statement consists of a 27-item checklist and a four-phase flow diagram. The PRISMA Statement is an update and expansion of the now outdated QUOROM Statement. Website: https://www.prisma-statement.org |
| Prognostic study | A prognostic study is designed to identify, assess, and interpret particular participant, study, or intervention characteristics (variables) that would serve as risk factors in predicting a particular outcome of treatment or result from exposure to positive and/or negative factors. |
| Protocol | In systematic reviewing, a written document created from scratch or based on an existing template that sets forth all steps in the systematic review process, including searching for literature, selecting abstracts and then full papers, extracting data from the primary studies, and synthesizing this information qualitatively or quantitatively. |
| Proxies | Individuals completing a measure on behalf of the index person – the person being measured. |
| Psychometric | Relating to the science of measuring, specifically the development of scales (measures, instruments, tools) to quantify psychological/mental traits, processes and abilities. By extension, issues involved in the measurement of properties of all intangible objects and states. |
| PsycINFO | PsycINFO is an electronic bibliographic database cataloging literature in the field of psychology, social, and behavioral sciences. PsycINFO is produced and maintained by the American Psychological Association (APA) and distributed through APA’s PsycNET and other third-party vendors. Updated weekly, this database contains more than 5 million records (as of 2023) including citations and abstracts for journal articles, book chapters, books, dissertations etc. |
| Publication bias | The phenomenon that the published literature contains mostly studies with positive results (i.e., supporting a hypothesis), because potential authors, peer reviewers and journal editors all prefer such positive results (the drug works; the test has sensitivity and specificity over 0.90, etc.), even though studies with “negative” results may have sufficient statistical power to make reliable claims of ineffectiveness. The absence of negative reports may result in unjustified support for an intervention, assessment instrument, etc., because only those investigators who by chance found positive findings get into print. Funnel plots can be used to assess how likely publication bias is with respect to a clinical question. The fail-safe number can be calculated to determine how strong publication bias needs to be to counter positive findings resulting from a systematic review. |
| PubMed | PubMed serves as a cost-free repository that facilitates the exploration and retrieval of literature in the subject matter areas of biomedical and life sciences. The PubMed database has over 35 million records (as of 2023) that include abstracts, citations, summaries etc. pertaining to biomedical literature. While complete journal articles are not available by PubMed freely, links to full-text items are frequently provided when accessible from alternate origins, such as the publisher's site or PubMed Central (PMC). PMC, an open-access full-text repository, can be accessed without charge Website: https://pubmed.ncbi.nlm.nih.gov |
| Qualitative research | A type of research focusing on quantifying the collection and analysis of data to test theories and hypothesis related to phenomena in the natural, applied, formal, and social sciences. |
| Qualitative synthesis | In systematic reviewing, using descriptive methods to combine the results of a set of primary studies addressing a specific problem or research question. |
| Quality assessment | In systematic reviewing, quality assessment is the assessment (using a checklist or similar instrument) or measurement (using a scale) of the methodological quality of the primary studies. In systematic reviews, quality assessment summaries can be reported in tabular and narrative form. Readers should be able to identify key quality aspects of studies quickly and to understand the reviewers’ rationale for rating a study good vs. poor. The review should also state how the evaluations of quality were used (delete poor quality research, weight studies by quality in a meta-analysis) etc., and why this use was appropriate. |
| Quality checklist | A list of criteria and categories relevant to research design and implementation that is used to systematically determine the methodological quality of individual studies. If the entries of the checklist are combined in some way to create a single “quality score”, it is a quality rating scale. |
| Quality rating scale | An instrument to quantify the methodological quality of primary studies, based on a list of items considered relevant to the dependability and generalizability of findings, overall or in light of a particular systematic review’s purpose (answering a question relevant to diagnosis, prognosis, etc.) |
| Quality-adjusted life year (QALY) | A measure of disease burden, including both the quality and the quantity of life lived. It is used in assessing the value for money of a medical intervention. See Disability-adjusted life year. |
| Quantitative research | A type of research that gathers non-numerical, descriptive data to understand individuals’ perceptions of social phenomenon. Methods used in qualitative research include interviews, focus groups, and observation. |
| Question | In systematic reviewing, the clinical question on the proper approach to treatment, assessment, prognosis, etc. that leads a practitioner to a review, or leads practitioners together with methodologists to create a systematic review. The main subject of the inquiry addressed in a review. Also called clinical question. |
| Random effects model | A random effects model is a statistical framework used in the analysis of data from various sources that exhibit inherent variability beyond the observed variability. In this model, it is assumed that the study samples being analyzed are drawn from a larger population where both observed and unobserved (latent) factors contribute to the variability in the data. These unobserved factors are referred to as "random effects." The random effects model takes into account both within-group variability (often referred to as "error" or "residual" variability) and between-group variability (which accounts for differences between the various groups or samples being studied). |
| Randomization | A method that uses chance to assign participants to comparison groups in a trial, e.g., by using a random numbers table or a computer-generated random sequence. Random allocation implies that each individual or unit being entered into a trial has the same probability of receiving each of the possible interventions (also called Random allocation or Random assignment). |
| Randomized controlled trials | Trials of interventions that use randomization to create a treatment and control group, whose outcomes are compared to determine whether the treatment being studied had an effect. Abbreviation: RCTs. |
| Rasch analysis | A variety of Item Response Theory. In the Rasch model, the probability of a specified response (e.g., right/wrong answer) is modeled as a function of person and item parameters. Specifically, in the simple Rasch model, the probability of a correct response is modeled as a logistic function of the difference between the person and item parameter. |
| Raters | Research professionals who are reviewing literature, either abstracts or complete, full text documents and using the rating forms to determine what literature will be included in the review and the qualities and overall quality of the primary research described in each document. |
| Rating form | An instrument used in systematic reviews by raters on which they place values relating to features of the studies being reviewed that will be used to make selection decisions about what literature to include in the review. A rating form typically permits a rater to provide a quantitative measure of a qualitative feature of a study, or the study’s description in a journal paper or other document. |
| RCT | See randomized controlled trial |
| Receiver operating characteristic (ROC) curve | A plot of the true positive rate (sensitivity) against the false positive rate (1-specificity) for all the different possible cut-points of a diagnostic test. |
| Reference standard (test) | A well-accepted measurement instrument with good reliability and validity that is used as a basis of comparison in the development of new measures of the same construct. Commonly known as the gold standard. |
| Registry | As relevant to systematic reviews, a Registry (trials registry) is an electronic database in which clinical trials (and other types of studies) are registered before data collection begins. In some countries and for some types of studies, registration is mandatory. Systematic reviewers can consult a registry to find research that has not or not yet been published. |
| Relative risk (RR) | The ratio of risk in the intervention group to the risk in the control group. The risk (proportion, probability or rate) is the ratio of people with an event in a group to the total in the group. A relative risk of one indicates no difference between comparison groups. For undesirable outcomes an RR that is less than one indicates that the intervention was effective in reducing the risk of that outcome. |
| Reliability | The consistency of a set of measurements or of a measuring instrument, often used to describe a test. Test-retest reliability, internal consistency reliability and other aspects of reliability are distinguished. |
| RePORTER | The NIH Research Portfolio Online Reporting Tool Expenditures and Reports (RePORTER) is a freely accessible public database, that was previously known as CRISP (Computer Retrieval of Information on Scientific Projects). RePORTER serves as a searchable repository of biomedical research projects that receive federal funding. It offers supplementary search categories to highlight publications and patents that have acknowledged the support of each respective project. The database allows users to conduct searches based on various criteria, such as Principal Investigator (PI), Institution, Government Agency, and State, among others. Additionally, RePORTER offers direct links to resources like PubMed Central, PubMed, and the US Patent and Trademark Office Patent Full Text and Image Database. Website: https://reporter.nih.gov |
| Reproducibility | The ability of a test (measurement, operationalization) to be accurately reproduced or replicated by someone else working independently. |
| Research design | An approach to the collection, analysis and interpretation of data in order to address a scientific question or test a hypothesis. |
| Reviews | Published materials that provide an examination of recent or current literature. Review articles can cover a wide range of subject matter at various levels of completeness and comprehensiveness based on analyses of literature that may include research findings. The review may reflect the state of the art. See also systematic review |
| Risk difference (RD) | The absolute difference in the event rate between two comparison groups. A risk difference of zero indicates no difference between comparison groups. A RD that is less than zero indicates that the intervention was effective in reducing the risk of that outcome. (Also called absolute risk reduction) |
| Scopus | Scopus is Elsevier's database funded in 2004 that houses abstracts and citations. As of 2021 Scopus has over 40 thousand indexed titles covering disciplines such as physical, health, social, and life sciences. Website: https://www.scopus.com |
| Search strategy | A structured and comprehensive plan or methodology for identifying and retrieving relevant research studies and literature from various sources. A search strategy typically contains the following components: Research question, search terms, databases, Boolean operators, filters and limitations, search execution, record management, and screening and selection. |
| Search term | A keyword or a phrase relevant to the search goal (e.g., “traumatic brain injury rehabilitation”). Search terms form a query, or user-defined request to the database or an online source. Terms in a query can be linked together through Boolean operators to increase the effectiveness (sensitivity [most if not all of the records that are desired are found] and specificity [few or none of the records that are not desired are found]) of the search outcome. |
| Selective outcome reporting | The tendency of researchers who investigated multiple outcomes (in an intervention, prognosis, etc. study) to only report on those outcomes for which statistically significant results were found. A similar preference for what appears most publishable (see publication bias) may extend to one of multiple interventions trialed, one of multiple time points at which outcomes were assessed, etc. Also called publication bias in situ, or within-study publication bias. |
| Selective publication | The phenomenon that studies that find support for the hypothesis are more likely to be published, because authors, peer reviewers and editors prefer positive results. Especially small studies with insufficient statistical power are likely to be missing from the published literature. See also selective outcome reporting. |
| Sensitivity | The sensitivity of a diagnostic (or screening) test is the proportion of people who truly have a designated disorder who are so identified by the test. The term sensitivity has various other meanings, as in the (closely related) sensitivity of a psychometric measure to detect change in a patient characteristic, and sensitivity analysis. |
| Sensitivity analysis | An analysis used to determine how sensitive the results of an analysis are to changes in assumptions made, and/or in how it was done. This may include determining whether the combined effect size from a meta-analysis changes to a clinically significant degree if the assumptions and the protocol for combining the data from the primary studies are varied. In systematic reviewing, sensitivity analyses are used to assess how robust the results are to certain decisions or assumptions about the data and the methods that were used, such as including vs. excluding weaker evidence. In an economic evaluation systematic review, in a one-way sensitivity analysis, only one variable is changed at a time; in multiway analysis, many variables are adjusted at the same time. The method can be used to consider thresholds of patient risk, effectiveness, or cost at which a health intervention might be judged a “good buy.” |
| Source selection bias | The systematic selection of data/information from a particular source while excluding other sources of data/information that cover the same or similar data/information. |
| Specificity | The specificity of a diagnostic or screening test is the proportion of people who are truly free of a designated disorder who are so identified by the test. The test may consist of or include clinical observations. |
| Spectrum of disease | Diseases typically involve a spectrum of pathologic changes, some of which are considered disease states and some pre-disease states. This range of related, sequential states a patient may go through as the disease progresses should be considered in systematic reviews of diagnostic tests. For instance, a test that is very useful in detecting individuals with a pre-disease state could be useless to diagnose patients with full-blown disease because all of them will test positive. |
| speechBITE | Founded in 2008 at the University of Sydney, speechBITE is a database of intervention studies covering subject matters related to speech pathology. Website: https://speechbite.com |
| Standardized mean difference | The difference between two means divided by an estimate of the within-group standard deviation. |
| Stop words | Common words (i.e., articles, prepositions) that are frequent and have little meaning (e.g., THE, AN, A, OF). Stop words should be avoided when a search query is constructed, unless they have a special meaning. If the latter is case it is recommended to use symbol + to emphasize the importance of a particular preposition (e.g., +to +become fertile). |
| Subgroup analyses | In systematic reviewing, subgroup analysis may be used to address specific questions (based, ideally, on characteristics selected prior to study start.) when data for subgroups of subjects are available in the set of comparable studies, Data may come from completely different studies (investigators L and M studied the association between A and B in women, and investigators N and O in men), or may come from a single study which reported separately for each sex (e.g., investigator P reported on the association between A and B separately for the men and women in her study). When conducted in a post-hoc fashion, results should be interpreted carefully. |
| Subject headings | Terms or labels used to identify primary topics or subject matter, specifically in a bibliographic database; in systematic reviews, the MeSH (Medical Subject Headings) subject headings of the National Library of Medicine are often used to identify potential studies in PubMed; in PsycINFO and CINAHL, subject headings are called thesaurus terms |
| Supplemental information | Materials that supplement a published paper but are too big for the printed version, and are published either on the journal publisher’s website, or (less commonly) on the website of the authors and their university/institution. |
| Syllabus | A document that provides step-by-step guidance or instruction on what is to be done. |
| Synthesis | In systematic reviewing, combining the data that have been collected in evidence tables qualitatively or quantitatively (meta-analysis) |
| Systematic review | A systematic review synthesizes research evidence focused on a particular clinical question and follows an a priori protocol to systematically find primary studies, assess them for quality, extract relevant information and synthesize it, qualitatively or quantitatively (meta-analysis). Systematic reviews reduce bias in the review process and improve the dependability of the answer to the question, through electronic and manual literature search and critical appraisal of individual studies. See: https://www.campbellcollaboration.org/what-is-a-systematic-review.html |
| Target groups | The subjects that are being studied in each of the studies included in the review; generally, the specification of the patient group(s) (by age, sex, condition, co-morbidities, etc.) for which prognoses, treatment outcomes or diagnostic/assessment test qualities are evaluated |
| Template protocol | Organizations that sponsor or organize many systematic reviews may develop protocol templates that their reviewers are invited or required to follow. These templates may specify all aspects of the systematic review from beginning to end. Cochrane, the Campbell Collaboration and the American Academy of Neurology are among the organizations. |
| Test administrator / test reader | In diagnostic accuracy studies, a clinician (e.g., a radiologist) or technician (e.g., laboratory technician) who reviews the result of a machine-produced image or other reflection of a disease process and classifies the result as positive (disease) or negative (normal). |
| Thesaurus | A collection of words (e.g., synonyms or antonyms) for a particular construct or concept that provides a cross referencing system of related terms. |
| Thesaurus terms | The name used in some bibliographical databases for controlled vocabulary terms, such as keywords or descriptors combined by their semantic relationships and chosen to describe a particular subject area. Thesaurus terms allow a search engine to map relevant words to related concepts or to show the relevant pages even if the vocabulary of the text did not match. |
| Treatment integrity | Treatment integrity (fidelity) typically refers to the correct delivery of the independent variable in all aspects: timing, quantity and quality of treatments, etc. Fidelity of treatment in outcome research is a confirmation that the manipulation of the independent variable occurred as planned. Verification of fidelity is needed to ensure that fair, powerful, and valid comparisons of replicable treatments can be made. |
| Truncation | An electronic database search strategy in which only the first part of a word (keyword) is used to find any word in a database that starts with those letters. After typing in the first part of the word, a truncation symbol is then typed in to represent any number of letters to follow (e.g., rehab). |
| Truncation symbols | A symbol put at the end of a word in order to catch all variant endings or spellings of that word when searching a database. The truncation symbol in PubMed is “*.” |
| UMIN Clinical Trials Registry | University Hospital Medical Information Network of Clinical Trials Registry (UMIN-CTR) is an online registry of clinical trials performed in Japan. Website: https://center9.umin.ac.jp/FAQ/en/UMIN-CTR_e/ UMIN-CTR is part of the wider Japan Primary Registries Network (JPRN) that includes three other national registries. Search: https://rctportal.niph.go.jp/en |
| Unpublished materials | Research or information not readily available via traditional bibliographic databases of (peer reviewed) published papers, or in the grey literature. |
| Usual care | Services and supports received by people who do not receive the intervention being studied in a systematic review |
| Utility | A term used by economists to sum up the satisfaction gained from a good or service. In health care evaluations, utility is often used in measures such as the quality-adjusted life-year or healthy-year equivalent, which take into account effect on quality of life as well as life-years gained. |
| Validity | The extent to which measurement instruments (scales, tests) measure what they purport to measure. Validity refers to the degree to which evidence and theory support the interpretations of test scores entailed by proposed uses of tests. |
| Variety of the evidence | In systematic reviewing, the diversity of the samples, treaters, outcome measures, treatment variations, etc. in the evidence base. When all these diverse studies come up with the same finding, one will have more confidence in the conclusions and recommendations of the systematic review. On the other hand, diversity may lead to heterogeneity, which may make drawing conclusions difficult. |
| Web of Science | A bibliographic database part of the ISI (Institute for Scientific Information) Web of Knowledge databases by Thomson Reuters |