INTRODUCTION

Literature reviews have been increasingly used by health professionals to assimilate the results of studies in the context of health care.¹

Among the various reviews, the systematic literature review (SLR) is defined as a systematic, explicit and reproducible method that allows the identification, evaluation and synthesis of studies carried out by researchers, academics and health professionals.² This methodology starts from question that is clearly formulated and uses systematic and explicit methods to identify, select and critically appraise studies; in addition, it allows for the collection and analysis of data from the studies that were included in the review.¹

Although this method of literature synthesis has had greater expression in recent years, it is not a recent idea. James Lind, in 1753, conducted the first randomized clinical trial, recognized the value of systematic methods to identify, to extract and to evaluate information from studies in order to avoid biased interpretations of the investigation.^3-4

There are many important historical events on the SLR, for example: in 1904, when Pearson publishes a historical review on the effects of vaccines against typhoid fever; in 1976, Glass coined the term "meta-analysis"; in 1984, Light and Pillemer report summarizing the results; in 1987, Mulrow publishes a medical review article on the state of science; in 1989, Enkin and colleagues publish Effective Pregnancy and Childbirth Care; in 1992, Antman and colleagues illustrated the value of accumulating results; in 1993, the launch of Cochrane Collaboration; in 1994, the creation of the UK NHS Center for Reviews and Dissemination; and in 2000, the creation of the Campbell Collaboration Foundation.

Literature review^3,10 can assume different expressions related to the degree of systematization and function for which they are intended. However, the SLR is based on an explicit, clear and standardized method so that it can be reproduced, which describes a priori in a rigorous way how it should be planned. ^3-5

The use of SLR makes it possible to "take stock" and have an overview of the knowledge produced so far, to identify opportunities that have not been explored yet and carrying out an innovative research project, in short, makes it possible to know the "state of the art". On the other hand, it allows verifying a specific hypothesis, in order to select tools, instruments or scales that are useful to conduct research and also to know gaps in studies, indicate unexplored topics or help to formulate research questions.³

The key features of an SLR are: clear definition of objectives based on pre-defined eligibility criteria for studies; explicit and reproducible methodology; systematic search that tries to identify all studies that meet the eligibility criteria; assessing the validity of the results of the included studies, for example, by assessing the risk of bias; and systematic presentation and synthesis of the characteristics and findings of the included studies.⁶ The fact that it is reproducible is highlighted, which emphasizes the need for clarity in each of the steps.

This essay aims to present the different types of literature review and describe the main stages of an SLR.

MATERIAL AND MÉTHODS

A narrative review of the literature⁷ was carried out in order to obtain a synthesis of the various types of systematic review, as well as their characteristics and functions in the context of nursing.

The essays included were obtained through the following platforms: Academic Google, Scientific Electronic Library Online (SciELO), EBSCO Host and Virtual Health Library (VHL), where it was possible to access the following databases: IBECS; CINAHL Complete; Library, Information Science & Technology Abstracts and MEDLINE Complete.

The subject titles and free terms were: 1. (Meta-Analysis/Meta-analysis) AND (Evidence-Based Medicine/Evidence-Based Medicine) AND 2. (Review/review) OR (Review Literature as Topic/review literature as subject) AND (Methodology /Methodology).

The inclusion criteria were: Language (Portuguese, English and Spanish); Availability (full text), all types of articles and books. The references of these articles or books were also considered.

RESULTS AND DISCUSSION

Literature reviews can be named as: Critical review; Integrative review; Literature revision; Systematic mapping/map review; Meta-analysis; Review of mixed studies/mixed methods review; Overview; Qualitative systematic review/synthesis of qualitative evidence; Quick review; Scoping review; Review of the state of the art; Systematic review; Systematized review; Systematic research and review; and Umbrella Review.^3,8

The critical review of the literature aims to demonstrate that an extensive literature search was carried out and its quality was critically assessed.⁸ In order to help with this assessment, guidelines are available to critically assess the quality of studies with a qualitative design.⁹

The integrative literature review allows for the combination of primary and secondary research, after evaluating the methodological quality and consists of six distinct phases: 1) Identification of the theme and selection of the hypothesis or research question for the elaboration of the integrative review; 2) Establishment of inclusion and exclusion criteria for studies/sampling or literature search; 3) Definition of information to be extracted from selected studies/categorization of studies; 4) Evaluation of the studies included in the integrative review; 5) Interpretation of results and, 6) Presentation of the review/synthesis of knowledge.^10,11

The steps of a narrative literature review or also called traditional are: selection of a review topic; literature search; selection/collection, reading and analysis of literature; review writing; and references.¹² Review essays can cover a variety of subjects and may include research findings. As it presents a very broad description, generalization is not possible.⁸

The steps of the mapping or systematic map review are: 1) Establishment of a review and stakeholder engagement team; definition of scope and issue; definition of inclusion criteria for studies; scope of studies; development; and publication of protocols; 2) Search for evidence; 3) Selection of evidence; 4) Encoding; production of a systematic map database; 5) Critical evaluation (optional); 6) Description and visualization of results; production report; and supporting information.¹³ This review allows you to map and categorize existing literature on a specific subject, by identifying gaps in the literature and justifying further reviews and/or primary studies.^3,8

Meta-analysis is a technique that statistically combines the results of primary studies in order to find a more accurate effect of the results, decreasing the bias and increasing objectivity, robustness and correlations of the results.^8,14A Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) for Protocols is a 17-item checklist designed to facilitate the preparation and reporting of a robust protocol for systematic review.¹⁴ The latest update of PRISMA's recommendations has 27 checklists, which allows to improve the quality of the report as well as the methodological quality.¹⁵

Mixed methods review can refer to any combination of methods in which at least one of the components is a literature review (usually systematic). For example, it may include a systematic review accompanied by interviews or a stakeholder consultation.^3,8

An overview review is a generic term describing a review of the medical literature. As such, it can be used for many different types of literature review, with different degrees of systematicity.^8,16

Qualitative reviews, according to the Cochrane Collaboration's handbook and the Center for Reviews and Dissemination methodologies, are gradually gaining more weight.^3,8 They are also part of primary studies, applied in a insightful and uniform way, but not statistically combined. They find their genesis in the deepening of human interaction and individual experiences. Used in research work on attitudes, beliefs, preferences and life experiences.¹⁷

Rapid review methods were considered by some authors as an undesirable need for evidence-based decisions. This type of review allows for an assessment of what is already known about an issue of policy or practice, using systematic review methods to research and critically assess existing literature.^8,18

The scoping review provides a preliminary assessment of the potential size and extent of available research literature. It is intended to identify the nature and scope of the evidence.^8,19

The state-of-the-art review focuses on more current issues. This review may offer new perspectives on an issue or highlight an area that needs further investigation.^3,8

Systematic review is the best known type of review. The systematic search for, evaluating and synthesizing evidence from studies, often adhering to guidelines on conducting a review provided by the Cochran Collaboration.^3,8

Systematic review and research combines the strengths of a critical review with a comprehensive research process. Typically, this type of review addresses broad issues and the result is a synthesis of better evidence.^3,8

Systematized reviews attempt to include one or more elements of the systematic review process and are not considered a true systematic review. It is usually performed by a graduate student.⁸

The umbrella review (also called coverage review) uses only units of analysis taken from systematic reviews and meta-analysis. It must comply with a peer review protocol and the tools available to assess the research synthesis, which must be explicit, clear and objective.^8,20

In general, the methods used in a review comprise conducting the review in four stages: Research (search and selection of studies), evaluation, synthesis and analysis.^3,8,21 The main types of literature review and will be presented and analyzed based on these four steps (Table 1).

A “systematic approach” refers to the elements/attributes that a literature review, whether done individually or collectively, has to present so that its methods are considered explicit and reproducible.³ In this sense, conducting a systematic review involves the work of at least two researchers, who will independently assess the methodological quality of each selected article, based on a research protocol⁹ and who will then compare the results obtained, which, if there is no agreement, should proceed to the next step, so that can be re-screened.

Systematic approaches are evidenced both in terms of conduct and in presentation of the literature review, and are summarized in the description of the method. Specifically, these approaches include:

Table 2 presents the main types of review that exist and the most used in the health field, where their advantages and disadvantages are explained.

Considering the 14 review types and methodologies associated with systematic review labels, there are frequent inconsistencies or overlaps between descriptions of nominally different review types. Currently, there is no international consensus on the types of reviews that are serious, coherent and mutually exclusive. The most pragmatic way to identify which of these various types of review is the most appropriate is through the application of the four main processes associated with the development of that review.^3,8

The Cochrane Collaboration22 recommends that a systematic review be carried out in 8 (eight) steps: Definition of a review question and criteria for including studies; Search for studies; Selection of studies and data collection; Assessment of the risk of bias in the included studies; Data analysis and conducting meta-analyses; Placement of biases in the report; Present results and tables with "results summary", and lastly; Interpretation of results and conclusions.

Other authors^9-10 refer to seven steps:

1. Construction of the research protocol so that the review follows the same accuracy as a primary research. The components of this protocol are: the review question, inclusion criteria, and strategies for seeking research, how research will be critically evaluated, data collection and synthesis. The review planning is carefully prepared and it is recommended that the protocol be evaluated by a competent professional, prior to the start of the review.^9-10 It is recommended to register the protocol on the PROSTERO platform to avoid redundancies (https://www.crd .york.ac.uk/PROSPERO/).

2. Formulation of the question using the acronym PICO,⁹ where P corresponds to the person or population (population), I is the intervention (intervention), C is the comparison or control (comparison/control) and O is the outcome or result (outcomes). In some specific cases, acromium derived from PICO should be used.²³

3. Search for studies with the definition of descriptors, search strategies in each of the various electronic databases (MEDLINE, CINAHL, EMBASE, LILACS, Cochrane Controlled Trials Database, SciSearch, among others).^9-10

4. Selection and review of studies applying predetermined inclusion and exclusion criteria.^9-10

5. Critical evaluation of each article; for this purpose the grids must be used. Joanna Briggs Institute (JBI) criteria may be used according to the study design: JBI-QARI for the qualitative assessment and review of the instrument and is designed to facilitate critical assessment, data extraction and meta-synthesis of results of qualitative studies; JBI-MAStARI is specific for quantitative studies and is used to perform meta-analysis; JBI-NOTARI allows to evaluate text narratives, opinions and evaluations, facilitating critical evaluation, data extraction and the synthesis of expert opinions in texts and reports; and JBI-ACTUARI which uses in cost analysis, technology and the use of assessment and instrument review, facilitating critical assessment, data extraction and synthesis of economic data.²⁴ However, as an alternative, one can use the following grids and guidelines to assess the methodological quality of the studies, based on their design: Randomized clinical trials - CONSORT; observational studies - STROBE; Systematic reviews - PRISMA and AMSTAR; case studies - CARE; Qualitative research - SRQR & COREQ; Diagnostic/prognostic studies - STARD&TRIPOD; Quality improvement studies - SQUIRE; Economic evaluations - CHEERS; Clinical guidelines / guidelines - AGREE II. Guidelines for each type of study are accessible on the website - http://www.equator-network.org/.²⁵

6. Data collection using instruments that analyze in pairs (two researchers independently) the methodological validity. At this stage, the level of evidence, quality²⁶ and degree of recommendation^27-28,the applicability of the results, the cost and current practice is determined, in addition, the limits between the benefits and risks of a given intervention are clearly determined.^{9- 10}

7. Summary of results/data, where the studies should be grouped based on the homogeneity of the studies. The presentation and synthesis of data must be pre-established in the protocol, as well as the graphical and numerical presentation mode, to facilitate the reader's understanding of the reader.^9-10

The main criticism that has been made to literature reviews is related to the non-use of clear, formal, explicit and systematic methods, which has harmed their status and usefulness as research.²⁹

Regardless of the choice of the type of study to be applied when there is a need to investigate a particular subject or theme, it is essential that these studies are credible. This must reflect accuracy and quality in its conduct. Well-conducted reviews increase the possibility of unbiased results, and of making valid and robust interpretations. This type of writing remains a challenge, but its importance is crucial, enabling all this information produced to have an impact on the provision of nursing care and also on the knowledge that is produced.^30-31

The characteristics of a high-quality systematic literature review in contrast to low-quality ones (main errors and pitfalls) are presented below.

Managing a literature review is similar to managing any research project. In this sense, it is necessary to identify the skills, mastery of tools and methodologies, as well as the necessary resources (human, database, time, financial, among others).^3,36

In order for rehabilitation nursing to expand the production of its knowledge and demonstrate health gains sensitive to its care, it is necessary to expand both the strength of the evidence and the degree of recommendation²⁸ and the quality of the evidence being produced.

Therefore, the source of the scientific evidence must be identified, which preferably should be primary studies, but may also be secondary, but an assessment of the quality must be carried out, in terms of its robustness (validity and reliability) and its relevance to the context location (applicability).⁸

FINALTHOUGHTS

There are several types of systematic review of the scientific literature, all of them with advantages and disadvantages. In this narrative review, 14 types of review were found, all of which can be important to synthesize the knowledge produced.

To help in making the decision for the type of review, it is necessary to balance and consider the investment of resources and energy in new researches if there are others done previously.

A systematic review of the literature has common principles and similar processes, but it can vary like primary studies, both in terms of length, scope and depth, as well as in the types of questions, data and methods used.

These secondary studies, like any other primary study, need proper quality assurance processes to assess them so that the result is representative of reality.

Researchers in general, and rehabilitation nurses in particular, should be aware of the many practical, methodological and policy challenges involved in this type of study and its broader role in the production and use of research findings.