Introduction

A review is only as useful as its report. A reader who cannot see which databases were searched cannot judge whether the search was complete, and a reader who cannot see how studies were appraised cannot judge how far to trust the conclusions. Chalmers and Glasziou (2009) and Glasziou and colleagues (2014) counted incomplete and unusable reporting among the main sources of avoidable waste in health research, and studies of published systematic reviews have repeatedly found incomplete reporting of search strategies and appraisal methods (for example, Page et al., 2016).

The response has been the reporting guideline: a checklist of the minimum information a report should contain, developed by consensus among methodologists, editors and users, and usually published with an explanation and elaboration paper that gives the reasoning and examples for each item. The EQUATOR Network (Enhancing the QUAlity and Transparency Of health Research) maintains a searchable library of reporting guidelines for every kind of health research, including the review guidelines in this section.

1.1 What a Reporting Guideline Does

A reporting guideline describes what a report should say so that readers can understand, appraise and, if they wish, repeat the work. The methods for conducting a review come from methods guidance such as the Cochrane Handbook and the JBI Manual for Evidence Synthesis, which earlier lessons used. A well-reported review can still be a weak review. Its strength is that the weakness is visible, so readers can discount its conclusions accordingly.

Editors and peer reviewers use reporting guidelines to check submissions, and many journals require a completed checklist. Decision-makers' analysts use the reported methods to judge whether a review is trustworthy, and later teams use the reported strategies and forms to update it.

1.2 The Structure of PRISMA 2020

The PRISMA 2020 statement (Preferred Reporting Items for Systematic reviews and Meta-Analyses) replaced the original 2009 statement (Moher et al., 2009). It contains 27 items, several with sub-items, grouped into seven parts of a report (Page et al., 2021). It was written for systematic reviews of the effects of interventions, with or without meta-analysis, but most of its items apply to other kinds of review. A companion paper explains each item and gives examples of good reporting, and a separate 12-item checklist, PRISMA 2020 for Abstracts, sets out what a structured abstract should contain. Lesson 7 introduced the PRISMA 2020 flow diagram, which answers item 16a.

The methods and results mirror each other, so a reader who learns how risk of bias was assessed (item 11) can find the judgements (item 18). The discussion separates the limitations of the evidence (item 23b) from those of the review process (item 23c), a distinction Section 1.4 returns to.

1.3 Extensions and Related Guidelines

PRISMA 2020 is the parent of a family of extensions, each adapted to a kind of review or to one part of a report, and other groups have written guidelines for syntheses that PRISMA does not cover. A team chooses one main checklist that matches its review type and adds others for particular parts of the report. Open each card for a short description.

The Cedar Valley report combines several of these. Its main checklist is PRISMA-ScR, because the review is a scoping review. Its search is reported with PRISMA-S, and its synthesis of effects on loneliness follows SWiM. PRISMA-ScR has no item on the certainty of evidence, because scoping reviews seldom rate it, so the team reports its GRADE ratings with PRISMA 2020 items 15 and 22 and explains in the methods why a scoping review rated certainty for one outcome. The environmental scan has no single agreed reporting guideline, and the team reports the elements set out in Lesson 11. The figure shows the resulting plan.

1.4 Writing Each Part of the Report

A checklist says what to include. The guidance below concerns how to write each part.

The distinction between the two kinds of limitation helps readers decide what to do next. Limitations of the evidence concern the included studies, and they would remain if a perfect review were done. Limitations of the review process concern decisions the team made, and another team could avoid them. The table shows examples from the Cedar Valley draft.

1.5 Amendments, Funding, Competing Interests, AI Use and Data

The final items of PRISMA 2020 are easy to treat as formalities, yet each tells readers something they need in order to judge the review.

Amendments to the protocol

Item 24c asks authors to describe and explain any change from the registered protocol. Changes are normal, especially in a rapid review, and reporting them lets readers judge whether a change could have shaped the results. The Cedar Valley team recorded six amendments as dated notes on its OSF registration, and the report lists them in a table.

Funding and competing interests

Items 25 and 26 ask who paid for the review, what role the funder played, and whether any author has interests that a reader might see as relevant. The Cedar Valley review was funded by the health authority that will make the decision, and the evidence officer is employed by that authority. The report therefore states that the planning team set the question and commented on the draft brief, that the evidence team made all decisions about methods and wording, and that the evidence officer's employer intends to launch the program under review.

Use of artificial intelligence tools

Lesson 6 described how to record and verify any use of AI tools. The report describes each use in the methods: the tool and version, the date, the task, and how the output was checked. In week 5, after the database searches, the Cedar Valley team used a conversational assistant and an AI research assistant to suggest sources for citation chasing and to check the sensitivity of the search (Lesson 6), verified every citation they suggested against the original source, and placed its AI-assisted search log in an appendix. Emerging guidance such as the RAISE recommendations (Responsible AI in Evidence Synthesis) expects this disclosure.

Availability of data and materials

Item 27 asks which materials are publicly available and where. The Cedar Valley team posts its search strategies, screening decisions, charting form and completed charting table on OSF, which lets others check the review and makes an update easier. The survey responses and interview notes from the 11 responding programs and 7 key informants were collected on the understanding that they would be reported only in summary, so the team shares the scan template and a summary table and keeps individual responses confidential (Lesson 6).

1.6 Using a Checklist While Writing

A checklist completed after the report is finished tends to be filled in generously. Teams that use the checklist as a writing tool draft an outline with every item placed under a heading, write each section, then ask someone who did not write it to check each item against the text and to record the location. In the Cedar Valley team, the intern completed the PRISMA-ScR checklist, the librarian checked the PRISMA-S items, and the evidence officer checked the rest.

Introduction

Many readers of a review go first to its tables. A manager wants to see in one place what was studied and what was found, and an analyst wants to see whether the included studies resemble the local population. Tables and figures carry most of a review's results, and their design decides whether readers find what they need quickly and interpret it correctly. Rosenbaum and colleagues (2010) found in a randomized comparison that adding a summary of findings table to Cochrane reviews improved readers' understanding of the results and the speed with which they found key information.

This section covers the characteristics table, the summary of findings table and the evidence map. All the Cedar Valley examples are fictional and illustrative, and identifiers such as CV-007 refer to invented studies.

2.1 Principles for Review Tables

A few principles apply to every table in a review. Each table should answer one question that its title states, naming the population, the interventions and, where relevant, the outcome. Rows should be ordered in a way that means something to the reader, such as by intervention category and then by design. Terms should match those in the text and the charting form, numbers of participants should accompany any percentage, and abbreviations and longer explanations belong in footnotes. The table should distinguish "not reported" from "not applicable", using the conventions of the data dictionary from Lesson 8. A reader should be able to understand the table without reading the text around it.

2.2 The Characteristics of Included Studies Table

PRISMA 2020 item 17 and PRISMA-ScR item 15 ask authors to describe each included study or source of evidence. The usual vehicle is a characteristics table, with one row per study, drawn directly from the charting form described in Lesson 8. Its columns follow the question: the study and country, the design, the participants, the intervention and comparator, and the outcomes with their measures and time points. In a long review the full table goes in an appendix and the report carries an excerpt and a summary. The excerpt below shows four rows of the Cedar Valley table.

A summary table groups the same information so that readers can see the shape of the evidence at a glance. The Cedar Valley team grouped its 42 studies by intervention category and design. The table shows at once that all 14 randomized trials evaluated group-based or one-to-one programs, while the evidence on community connector programs comes from non-randomized, before-and-after, qualitative and mixed-methods studies.

The text beside the table adds what the reader cannot see in it. For Cedar Valley, that includes the countries (6 studies in Canada, 11 in the United Kingdom, 9 in the United States, 9 in other European countries, 5 in Australia and 2 in Japan) and the fact that most studies recruited through community organizations, with only the community connector studies recruiting mainly through primary care. The 26 grey-literature documents, most of which are short program evaluation reports, are summarized in a separate table, because the team charted them with a shorter form (Lesson 7).

2.3 From Evidence Profile to Summary of Findings Table

Lesson 8 built a GRADE evidence profile for the Cedar Valley review, which records the judgement for each certainty domain (risk of bias, inconsistency, indirectness, imprecision and publication bias) and the resulting rating. The profile is a working document for the review team and for guideline panels, who need to see each judgement. A summary of findings table presents the same evidence for readers who need the results and the certainty without the domain-by-domain detail, with footnotes that explain each rating (Guyatt et al., 2011). Both are produced from the same judgements, and software such as GRADEpro GDT, available in 2026, generates either from one set of entries.

When a review pools results in a meta-analysis, the standard summary of findings layout has columns for the outcome, the anticipated absolute effects (the risk or mean with the comparator and with the intervention), the relative effect with its confidence interval, the number of participants and studies, the certainty and comments. HSCI 230 Lesson 2 explains how those numbers are computed. GRADE suggests that a summary of findings table present no more than seven outcomes, chosen for their importance to decision-makers, and that it include important outcomes even when no study measured them, so that the absence of evidence is visible.

The Cedar Valley review synthesized results without meta-analysis, using the approach that Lesson 9 teaches, so it has no pooled effect to report. Following Murad and colleagues (2017), the team replaced the effect columns with a description of the direction, size and consistency of the results, and it worded each conclusion to match its certainty (Santesso et al., 2020).

Loneliness was measured with validated self-report scales at the end of each program. ^a All 8 trials had some concerns (6) or high risk of bias (2), mainly because participants knew their group and reported their own loneliness. ^b Results varied in size, and program length and setting did not explain the variation. ^c Three trials were at high risk of bias and three had some concerns. ^d The total number of participants was modest and most confidence intervals were wide. ^e Non-randomized evidence starts at low certainty; three of the 5 studies were at serious risk of confounding, because clinicians referred people they judged likely to benefit. The team considered rating up for the gradient with connector contacts and did not, because people who improved early may have chosen to keep meeting their connector.

The certainty column carries both a symbol and a sentence, so a reader who skips the symbols still receives the message. The footnotes explain every rating, and the last two rows report what the review did not rate, so that no one mistakes silence for an absence of effect.

2.4 Evidence Maps

An evidence map displays how much evidence exists for each combination of two dimensions, usually interventions and outcomes, so that readers can see where evidence is concentrated and where gaps lie. Evidence and gap maps, developed for policy use by Snilstveit and colleagues (2016) and published by the Campbell Collaboration, are one well-known form; Lesson 10 teaches how they are built. In a review report, a simpler map often serves as the main figure of a scoping review, because mapping is the purpose of the review. The Cedar Valley map counts the studies in each intervention category that reported each outcome domain, with a final column for qualitative and mixed-methods studies of participants' experiences and of implementation.

Read across the rows and down the columns, the map shows several things. Loneliness was the most measured outcome (33 studies), and group-based and community connector programs account for most of the evidence. Health service use, which matters to a health authority deciding where to place a program, was reported in only 7 studies, five of them evaluations of community connector programs. Intergenerational and technology-based programs have few studies in any column, and no technology-based study examined participants' experiences.

A map shows the volume of evidence and says nothing by itself about effects or certainty. A large circle means that many studies measured an outcome; it does not mean the program works or that the evidence is trustworthy. A map should also be read with the review's eligibility criteria in mind: an empty cell shows a gap in the evidence the review looked for, which may differ from a gap in all research.

2.5 Other Figures and Accessible Design

Reviews use several other figures, each answering a particular question. Earlier lessons introduced most of them, and the tabs show what each is for.

Each figure needs a caption that states its message and alternative text for screen-reader users. Colour should never be the only way of carrying meaning, because about one man in twelve of Northern European ancestry has a red-green colour-vision deficiency; labels, shapes or patterns should carry the same information, which is why the Cedar Valley map prints a number in every circle and a zero in every empty cell. Text should be readable in print and on a phone, with labels placed beside the data where possible.

Introduction

The Cedar Valley planning team asked for a review because it has to decide how to launch a community connector program, and it needs to know what the evidence supports. Turning a review into something that decision-makers can use is part of knowledge translation, which the Canadian Institutes of Health Research describe as a dynamic and iterative process that includes the synthesis, dissemination, exchange and ethically sound application of knowledge. In the knowledge-to-action framework of Graham and colleagues (2006), products tailored to particular users, such as summaries and briefs, are a stage of knowledge creation that follows synthesis.

Lavis and colleagues (2003) proposed five questions for planning how research knowledge is shared with decision-makers: what should be transferred, to whom, by whom, how, and with what effect. For Cedar Valley, the message is the findings with their certainty and the scan results; the audiences are the planning team, the executive and the older adults and organizations affected; the messenger is the evidence team; the products are a brief, a plain-language summary and a meeting; and the hoped-for effect is a launch designed to answer the questions the evidence leaves open.

3.1 Different Readers, Different Products

A single review can produce several products, each complete for its reader. The table compares the four that the Cedar Valley team prepared.

All four products must say the same thing. A summary that sounds more confident than the review misrepresents the evidence even if each sentence is accurate, so the Cedar Valley team built every product from its summary of findings table.

3.2 Writing a Plain-Language Summary

A plain-language summary tells readers without research training what a review asked, what it found and how much confidence they can place in the findings. Cochrane publishes one with every review, using headings phrased as questions, such as What did we find?, What are the limitations of the evidence? and How up to date is this evidence? The last two matter because readers otherwise assume that published evidence is certain and current.

The main answer comes first. Sentences are short and active ("we searched" instead of "searches were conducted"), everyday words replace technical ones, and the word "significant" is avoided because readers take it to mean "important". Certainty uses the verbs of the summary of findings table: "may" for low, "probably" for moderate and "we are very uncertain" for very low certainty (Santesso et al., 2020). Readability formulas such as the Flesch-Kincaid grade level can flag long sentences, but only testing with readers shows whether a summary is clear.

The summary has 297 words including its headings, and its 18 sentences average about 14 words each. It explains the one technical term it needs, "community connector", and states its limitations and search date. Members of the health authority's patient and family advisory council read a draft and asked for the word "intervention" to be replaced.

3.3 Evidence Briefs for Decision-Makers

An evidence brief is a short document that presents the findings of a review, or of several reviews, to people who must make a decision, together with what the findings mean in their setting. Lavis and colleagues (2009) described how policy briefs can set out a problem, the options for addressing it and the considerations for putting an option into practice, drawing on systematic reviews and local evidence. The McMaster Health Forum in Hamilton, Ontario, prepares briefs of this kind.

Decision-makers seldom read a full report, and the Canadian Health Services Research Foundation (later the Canadian Foundation for Healthcare Improvement, and now part of Healthcare Excellence Canada) promoted a 1:3:25 format for reports written for them: one page of main messages, a three-page executive summary and a report of no more than 25 pages. Lavis and colleagues (2005) argued that reviews would serve managers and policymakers better in this kind of graded-entry format, with information on applicability, costs and implementation.

A brief has recognizable parts. Open each card for what it contains and why.

Two judgements shape every brief. The first concerns recommendations. Evidence teams often describe implications and leave the choice to decision-makers, whose choice also depends on costs, values and feasibility. The Cedar Valley planning team asked directly how to launch the program, so the brief makes a labelled recommendation about the design of the launch, based on the gap in the evidence. The second concerns applicability, the extent to which findings are likely to hold in the decision-maker's setting. Only 6 of the 42 studies were Canadian, and the scan supplies what the studies cannot say about British Columbia.

3.4 Worked Example: The Cedar Valley Evidence Brief

The brief below is the version the evidence team presented to the planning team in week 12. It fits on two printed pages. All content is fictional and illustrative.

The brief follows the pattern in the cards. Its key messages carry the certainty of each finding, its table shortens the one in Section 2, the scan adds who runs programs locally and how they are funded, and its final line states how current the evidence is, which Section 4 develops.

Introduction

Every review describes the evidence as it stood on its search date. The Cedar Valley databases were searched on 10 February 2026, and any study published after that date is absent from the report, the brief and the summary of findings table. A review can go on being cited long after newer studies have changed the picture. Shojania and colleagues (2007) followed a sample of 100 systematic reviews of interventions and found that the median time before new evidence signalled a need to update was about five and a half years, while almost a quarter of the reviews needed updating within two years of publication.

Lesson 1 introduced living reviews as one member of the family of reviews. This section covers updating, surveillance and living reviews, and it closes the course by showing how the Cedar Valley final report draws on every stage.

4.1 Why Reviews Go Out of Date

A review can become outdated in several ways, and only some involve new studies. Open each card.

An update is a new version of an existing review that adds new studies or data, applies new methods, or both, while keeping the same question or a closely related one. When the question changes substantially, for example from older adults to all adults, the result is a new review that may draw on the old one. The distinction matters for reporting, because an update states what changed since the previous version, whereas a new review starts from its own protocol.

4.2 Deciding Whether to Update

Garner and colleagues (2016), writing for an international panel on updating systematic reviews, proposed that teams answer a short series of questions before committing to an update. The first asks whether the review is still current, meaning that it addresses a question that matters to decisions and used methods that remain acceptable. The second asks whether there are new methods or new studies, data or other information. The third asks whether adopting the new methods or including the new studies would be likely to change the findings, the conclusions or the credibility of the review. A review is updated when the answer to all three is yes. The figure adapts their framework.

The third step requires judgement. A new small trial that agrees with eight earlier trials is unlikely to change a conclusion that group-based programs may reduce loneliness slightly. A first randomized trial of a community connector program would be likely to change the certainty for that comparison, because the existing evidence comes only from non-randomized studies. Teams can make the judgement more consistent by writing down, in advance, the kinds of new evidence that would trigger an update. These pre-specified update triggers also protect against updating only when new results are welcome.

4.3 Watching for New Evidence

A team cannot answer the second question without some way of noticing new studies. This ongoing monitoring is called surveillance, and it uses tools that earlier lessons introduced. Database platforms let users save a search and receive email alerts when new records match it, which turns the review's own search strategies into a monitoring system. Citation alerts, which Lesson 5 described, report new papers that cite the key included studies. Trial registries such as ClinicalTrials.gov, the ISRCTN registry and the World Health Organization's International Clinical Trials Registry Platform show studies that are under way before they are published. AI tools can help sort the alerts, with the verification that Lesson 6 required for any AI output. Each check is recorded with its date and result, so that the team, and later its readers, can see how current the review is.

The Cedar Valley team wrote its surveillance plan into the final report. Because the student intern's placement ends with the review, every task is assigned to a permanent member of staff.

Environmental scans go out of date faster than reviews, because programs open, close and change their funding within months. The scan's currency is therefore reported separately from the review's search date, and the yearly refresh is planned from the start.

4.4 Living Systematic Reviews

A living systematic review is a review that is continually updated, incorporating relevant new evidence as it becomes available (Elliott et al., 2014; Elliott et al., 2017). Instead of a single search followed by occasional updates, the team searches at a set frequency, often monthly, screens new records as they arrive, and updates the analysis and the conclusions whenever pre-specified criteria are met. The review is published in versions, each with a date, a version number and a summary of what changed. Living reviews were used widely during the COVID-19 pandemic, when treatment trials were reported every week; for example, Siemieniuk and colleagues (2020) published a living network meta-analysis of drug treatments for COVID-19 that informed the World Health Organization's living guideline on therapeutics.

Elliott and colleagues (2017) suggested three conditions under which a living approach is most justified. The review question should be a priority for decision-making. The certainty of the existing evidence should be low or very low, so that new evidence could change the conclusions. And new research evidence should be likely to appear soon. When these conditions are met, a living review keeps decision-makers current at a lower total cost than repeated full updates, because each round of work is small. When they are not met, the continuous effort is hard to justify.

A living review also needs a plan for ending. Teams retire a living review, or move it out of living mode, when the question is no longer a priority, when the certainty of evidence is high enough that new studies are unlikely to change the conclusions, or when new research has stopped appearing. The final version should say that the review is no longer being updated and give the date of its last search, so that readers do not assume it remains current. Cochrane has published guidance on producing and publishing living systematic reviews, and its principles apply to other publishers.

4.5 Should the Cedar Valley Review Become a Living Review?

The planning team asked whether the review could be kept up to date while the program is running. The evidence team applied the three conditions and then considered its own resources.

The team concluded that a full living review was not feasible and chose surveillance with a planned update, focused on the comparison that matters most. If a randomized trial or a large controlled study of a community connector program for older adults appears, the team will update that row of the summary of findings table and reissue the brief within three months; otherwise it will make the formal update decision in February 2027. The team also noted that the health authority's own staggered launch will produce local evidence on the very question the review could not answer, and that the update should include those results once they are available.

4.6 Reporting an Update or a New Version

An updated review must let readers see what is new. PRISMA 2020 applies to updates, and Lesson 7 showed its flow diagram templates for updated reviews, which report the studies included in the previous version alongside those added by the new search.

4.7 Looking Back Over the Course

The Cedar Valley review has now passed through every stage of an evidence synthesis. The final report draws on the work of each lesson, as the table shows, and a smaller review assembles its report in the same way.

The same habits ran through every stage. The team planned before acting, recorded its decisions and reasons, had a second person check the steps most open to error, and reported its work in enough detail for others to repeat or update it. Those habits make a review trustworthy at any scale.

Bringing It All Together

This lesson followed the Cedar Valley evidence team from a completed analysis to the products that decision-makers and the public will read. The full report was planned around PRISMA-ScR, with PRISMA-S for the search, SWiM for the synthesis and PRISMA 2020 items for certainty and for the end matter, so that readers can see what was done, what changed after registration, who paid for the work and what can be shared. Its tables carry most of the results: a characteristics table that describes the 42 studies, a summary of findings table that states what the studies found about loneliness and how certain it is, and an evidence map that shows where evidence is concentrated and where it is missing.

The same findings were then translated for two other readers. The plain-language summary told older adults and their families, in short sentences, that group and one-to-one programs may reduce loneliness a little and that the effect of community connector programs is very uncertain. The two-page evidence brief gave the planning team key messages, findings with certainty, the results of the scan of 18 British Columbia programs, and a labelled recommendation to launch the program in stages across the 24 clinics with loneliness and health service use measured from the start. Finally, the team planned how to keep the review current, choosing surveillance with a planned update over a living review because its small team could not sustain continual searching, and noting that its own staggered launch would add the local evidence that the review could not find.

The final reflection asks you to plan the reporting, the evidence brief and the updating of a new rapid scoping review, drawing on all four sections.

Final Knowledge Assessment

These terms, frameworks and people appear in this lesson on reporting review findings, translating them for decision-makers and the public, and keeping reviews current.