Move from a broad research question to a reviewable scoping-review draft
This guide explains the complete AccuScope workflow, from choosing the study source and preparing the research question to retrieving evidence, scoring studies, reviewing the structure, generating evidence tables, and writing the final report draft.
What AccuScope does
AccuScope is a guided workflow for scoping reviews and evidence-mapping projects. It helps turn a research topic into search criteria, candidate studies, ranked evidence, a report structure, evidence tables, and a cited draft that remains reviewable by the research team.
It is designed for broad questions where the goal is to map what evidence exists, identify themes or evidence patterns, and prepare a structured report. It is not a substitute for researcher judgement on the final question, inclusion boundaries, selected studies, or published conclusions.
| What you provide | What AccuScope prepares | What you review |
|---|---|---|
| Topic or research question | PICOS fields and filters | Scope, criteria, and search intent |
| Search source or uploaded study list | Article details ready for scoring | Retrieved or imported study metadata |
| Final study selections | Structure, tables, and draft report | Citations, evidence tables, and narrative |
The complete AccuScope workflow
The workflow can run automatically from the start panel, but each major output remains visible and reviewable. You can pause, revise, retry failed stages, and continue manually from the relevant step.
- 1
Define the topic and source
Choose online search or uploaded study list, then enter a focused research question that can guide the workflow.
- 2
Prepare evidence
Generate and review search strings, retrieve hits and article details, or import study metadata from a spreadsheet.
- 3
Score and select studies
Run scoring to rank studies against the project criteria, then manually confirm the final selection for writing.
- 4
Review the report structure
Generate a proposed structure, inspect sections, references, and planned tables, then edit the outline if needed.
- 5
Generate evidence tables
Prepare the tables defined by the structure and check rows, columns, values, and abbreviations before writing.
- 6
Write and export
Generate the final draft, review citations and narrative, and export the report for researcher editing.
Choose how studies enter the review
AccuScope supports two source modes. Choose the source before starting, because the study-source choice is locked after the review begins. This prevents a later switch from mixing search-derived evidence with an uploaded study list in the same project state.
| Source mode | Use it when | What happens next | Researcher review |
|---|---|---|---|
| Search online for studies | Starting from a topic and letting AccuScope build database-ready search steps. | AccuScope autofills fields, generates search strings, retrieves PubMed hits, and fetches article details. | Search fields, search strings, translated strategies, hit counts, and retrieved article metadata. |
| Upload study list | Starting from a study list already assembled by the team or client. | AccuScope imports study metadata and moves directly toward scoring and selection. | Required titles, recommended abstracts, citation metadata, and whether each imported study belongs in the project. |
Write a research question that can guide search, scoring, and writing
The start panel asks for a topic or research question. AccuScope uses it to fill PICOS fields and filters, then carries that context into search generation, article scoring, structure proposal, table generation, and writing.
Include context that changes project decisions
- Population or setting
- Intervention, exposure, phenomenon, or concept of interest
- Outcomes, themes, or evidence areas to map
- Study designs or publication types that matter
- Time window or geographic limits, where relevant
- Text availability expectations, such as abstracts or full text
- Any important exclusions or scope boundaries
Review the fields that control retrieval and prioritisation
Autofill is a draft. Before generating searches or scoring imported studies, review the structured fields and filters. These values define the project boundary used throughout the workflow.
| Field group | What it controls | Review guidance |
|---|---|---|
| Population | The group, disease area, setting, or evidence population being mapped. | Make the population specific enough for search and scoring, but not narrower than the intended scoping question. |
| Intervention or concept | The intervention, exposure, phenomenon, or concept used to focus the evidence map. | Include synonyms or related concepts when they are important to retrieval. |
| Outcomes | The evidence areas AccuScope should prioritise and later discuss. | List outcomes, themes, or domains rather than only a generic phrase such as clinical evidence. |
| Outcome types | Treatment patterns, clinical efficacy, safety, economic, HRQoL, and epidemiology emphasis. | Select the evidence domains that should be represented in scoring, structure, tables, and writing. |
| Text availability | Whether the project favours studies with abstracts, free full text, or full text. | Use these filters intentionally because writing depends heavily on usable abstracts or text. |
| Study and publication filters | Study design and article-type boundaries for retrieval and scoring. | Balance recall and precision. Broad filters may bring noise; strict filters may miss relevant mapping evidence. |
| Language, geography, and dates | Limits that shape the search result set and interpretation. | Apply these only when the scoping question or protocol requires them. |
Generate, review, and retrieve database evidence
In online-search mode, AccuScope turns the reviewed fields into editable search steps. The generated strategy should be treated as a draft search strategy, not an irreversible decision.
- Enter the research question and start AccuScope or run autofill.
- Review PICOS, outcome types, study design filters, publication types, language, geography, and dates.
- Generate search strings and review each step description and query.
- Edit steps directly, add or delete steps, or use the Excel download/upload path for larger revisions.
- Generate Embase and Cochrane translations when you need translated strategies for those platforms.
- Retrieve PubMed hit counts, then retrieve article details from the final result set.
- Export search strings, hit summaries, or article details when you need a project record outside the app.
Use an uploaded study list when evidence is already assembled
Upload mode skips database search and starts from a study metadata file. This is useful when a team already has a curated candidate set or wants AccuScope to score and write from a predefined list.
| Input | Requirement | Why it matters |
|---|---|---|
| File type | Excel or CSV | Provides a tabular study list that can be imported into the project. |
| Title | Required | Creates the main study identity used in scoring and tables. |
| Abstract | Recommended | Studies without useful abstracts may be skipped for writing or table generation. |
| Metadata | Authors, year, journal, DOI, PubMed ID, article type, keywords, and MeSH terms are supported | Improves citation quality, traceability, and review context. |
Use scoring to prioritise, then confirm the final selection manually
Scoring ranks studies against the review criteria and creates an auto-selection set. The final selection is saved separately from raw ranking, so researchers can remove auto-selected studies or include lower-ranked studies before writing.
| Item | What it means | How to use it |
|---|---|---|
| Score | A project-specific prioritisation signal calculated from the criteria. | Sort and inspect studies, but do not treat score alone as eligibility. |
| Rank | The relative ordering of scored studies. | Start review from the highest-ranked studies, then inspect borderline or lower-ranked records as needed. |
| Auto-selected | A system-proposed inclusion set based on ranking and scoring rules. | Use it as a starting point, then remove or add studies manually. |
| Manual selection | The researcher-confirmed set used for structure, tables, and writing. | Confirm this set before generating downstream outputs. |
| Warnings | Signals that a study may be unsuitable or missing required information. | Investigate warnings before selecting a study for writing. |
Generate the outline before tables and writing
The report structure creates the section plan, reference allocation, and table allocation used by the later stages. A study must be selected and have usable abstract content before it can contribute to writing.
- Complete scoring and confirm at least one eligible selected study.
- Propose the structure and wait for the structure status to complete.
- Review the overview, sections, reference entries, and planned tables.
- Edit section titles or descriptions when the outline needs a clearer scoping-review structure.
- Export the structure when you need a Word copy for review outside the application.
- Continue only when the structure reflects the question and selected evidence.
Prepare evidence tables before final writing
Evidence tables are generated after the structure is complete and before the final report draft. Tables are based on the selected studies and the table plan in the structure.
| Table output | What to check |
|---|---|
| Table title and number | Confirm that it matches the section and research question. |
| Columns and descriptions | Check that the requested evidence concepts are clear and not duplicated. |
| Rows and cell values | Verify that extracted values are plausible and trace back to selected studies. |
| Abbreviations | Review abbreviation definitions before exporting or using the final draft. |
Generate and review the final AccuScope draft
Final writing starts only after selected studies, report structure, and evidence tables are ready. The completed report can include narrative sections, bullets, search summaries, evidence tables, citations, and reference entries.
| Output area | Review focus |
|---|---|
| Written sections | Check that each section answers the scoping-review question and does not overstate evidence. |
| Citations and references | Confirm that cited studies are selected, relevant, and correctly represented. |
| Search summary | Review search steps and hit counts for transparency. |
| Evidence tables | Compare table values with the generated table step before exporting. |
| Exported report | Use the export as a draft for researcher editing, not as a publication-ready manuscript. |
Know when earlier changes make later outputs stale
AccuScope preserves intermediate outputs so you can inspect them, but later stages depend on earlier decisions. When upstream inputs change, rerun the affected stage before relying on downstream output.
| Change | Likely impact | Next action |
|---|---|---|
| Research question or PICOS fields change | Search strings, scoring criteria, selection relevance, structure, tables, and writing may no longer match. | Regenerate or rerun the affected stages from search or scoring onward. |
| Search strings change | Hit counts, article details, scoring, selection, and writing outputs may be stale. | Retrieve fresh hits and article details, then rerun scoring. |
| Uploaded study list changes | Scoring and all downstream report outputs may reflect the old study set. | Rerun scoring and review the final selection. |
| Manual selection changes | Structure, evidence tables, and report writing may not reflect the selected studies. | Regenerate the affected report outputs before relying on them. |
| Report structure is edited | Evidence tables and writing should follow the revised outline. | Save the structure, regenerate evidence tables if needed, then write the report. |
Running a reliable AccuScope project
| Do | Avoid |
|---|---|
| check_circle Write a research question with clear population, concept, and evidence domains. | cancel Start from a vague topic and accept every generated field without review. |
| check_circle Choose search mode or upload mode based on the real evidence source. | cancel Mix database retrieval and client-provided study lists in one project state. |
| check_circle Review generated search strings and translations before retrieval. | cancel Assume generated queries are final search strategies. |
| check_circle Use scoring to prioritise review work. | cancel Treat scoring as the final inclusion decision. |
| check_circle Confirm selected studies before structure and tables. | cancel Write a report from an unchecked auto-selection set. |
| check_circle Review structure and evidence tables before final writing. | cancel Generate the final report before checking table allocation and citations. |
Before final writing
- check The research question accurately describes the scoping-review objective.
- check The study source mode matches how evidence should enter the project.
- check PICOS fields and filters have been reviewed after autofill.
- check Search strings or uploaded studies have been checked for scope fit.
- check Article details are available, or imported studies include usable titles and abstracts.
- check Scoring has completed and stale scoring warnings are resolved.
- check The final selected study set has been manually reviewed.
- check The report structure has been reviewed for sections, references, and planned tables.
- check Evidence tables have been reviewed before final writing.
- check The written draft will be checked by a researcher before external use.