Literature and evidence reviews

The evidence table, every cell sourced.

An operator screens your papers and public PubMed and ClinicalTrials.gov results against your criteria, then extracts the outcomes. Every cell carries its page and passage; your scientist signs off.

Talk to salesStart free
  • ISO 27001
  • GDPR
  • Every run auditable
Review 0412Halvard Pharma · Medical Affairs
Illustrative review

The table, every cell sourced

Oral therapies vs placebo in adults with moderate-to-severe plaque psoriasis: PASI-75 at Week 16

  • PopulationAdults, moderate-to-severe plaque psoriasis
  • InterventionOral small molecules
  • ComparatorPlacebo
  • OutcomePASI-75 at Week 16
  • DesignRandomised controlled trials, 2015–2026
  • Your library41 PDFs from SharePoint › Medical Affairs › Psoriasis
  • PubMed318 records, public search API
  • ClinicalTrials.gov27 registered trials, public API
  • 386 found
  • 336 unique, screened
  • 12 included
  • IncludedStudy A (2023)RCT in adults; PASI-75 at Week 16 reported · p. 4, Methods
  • ExcludedStudy K (2022)Outcome reported at Week 12 only · p. 5, Results
  • ExcludedStudy N (2024)Open-label extension, no placebo arm · Abstract
StudyNArmPASI-75, Week 16Source
Study A (2023)RCT, phase 3612
Study B (2024)RCT, phase 2238
Study C (2025)RCT, phase 3740
Study A (2023) · p. 6, Table 2

At Week 16, 58.4% of patients receiving KX-204 200 mg achieved PASI-75, compared with 9.1% receiving placebo (Table 2).

Not checked yet

Waits for a yesUpload Review 0412 to SharePoint › Medical Affairs › Reviews

Illustrative review: fictional studies, compounds and numbers that show the output's shape.

Enterprises running operators on Alomana

What the operator does in one review

The reading, screening and extraction behind an evidence table, done the way a careful reviewer would check it.

  1. The questionTurns your question into criteria (population, intervention, comparator, outcome, design) that you approve
  2. The searchReads the PDFs you hold and searches PubMed and ClinicalTrials.gov through their public APIs
  3. The screenIncludes or excludes each record against your criteria, with the reason and the page
  4. The tableExtracts design, population, arms and outcomes into Excel, each cell with its page and passage
  5. The checkRe-reads each value against its passage and flags disagreements, such as an abstract that differs from its table
  6. The draftWrites the summary in your Word template, every statement with its reference

How the operator works a review

  1. Connect

    SharePoint or Drive with your papers, plus a folder of past reviews and SOPs.

  2. Screen and extract

    Your criteria, every decision and cell traced to its page.

  3. Sign off

    Your scientist checks, edits and approves. Then it uploads and logs it.

A reviewed table, not a verdict.

The operator works from what your team can read and stops at a draft for your scientist.

  • Reads the papers you hold, under your own licences
  • Searches public PubMed and ClinicalTrials.gov
  • Every cell carries its page and passage
  • Your scientist decides what is included and signs off
  • Evidence table (Excel)
  • Screening log with reasons
  • Draft summary in your Word template
  • Reference list (RIS)

Weekly watch: plaque psoriasis, oral therapies

The same searches run on a schedule and draft a digest. Posting it waits for a yes.

Mondays, 07:00
  • Registered trials2 new on ClinicalTrials.govClinicalTrials.gov public API
  • Papers3 new on PubMedPubMed public API
  • Competitor news1 phase 2 readout in a press releaseWeb search

Digest drafted, each item with its link and the passage it rests on.

Waits for a yesPost the digest to #medaffairs-psoriasis on Slack

Nothing leaves the team without a yes.

  • Uploads and messages wait for a person's yes
  • A declined action is not retried
  • Every run logged: input, output, tools, user
Run record · illustrative review
Input
Review 0412 question and 41 PDFs from SharePoint
Tools
PubMed and ClinicalTrials.gov public APIs, Excel, Word
User
Dr. L. Brandt, medical affairs lead
Output
Evidence table, screening log, draft summary, reference list
Action
Upload to SharePoint · waiting for a yes
Measured on
  • Hours per review
  • Weeks to a first evidence table
  • Values corrected at QC
  • Reviews per analyst per year
Works with
  • SharePoint
  • Google Drive
  • Confluence
  • Slack
  • Public PubMed and ClinicalTrials.gov
  • A folder of past reviews and SOPs
  • Your literature database via MCP
Built-in operators
  • Clinical Trial Analyst
  • Drug Safety Monitor
  • Regulatory Affairs operator
Trusted for
  • ISO 27001
  • GDPR
  • Never trains on your data
  • Single-tenant for enterprise

A custom operator typically goes live in up to 4 weeks.

Questions evidence teams ask

What is a literature review operator?

An autonomous agent that works a review from your question to a draft. It turns the question into criteria you approve, screens each record with a reason, extracts the outcomes into a table with the page and passage behind every cell, and drafts the summary for your scientist to check.

Which sources does it search?

The PDFs you give it, including papers you hold under your own licences, and PubMed and ClinicalTrials.gov through their public APIs, plus public web pages. It has no index of its own and no Embase or Scopus access; a licensed database connects only through your own MCP server.

How do our scientists check its work?

Every inclusion decision, extracted value and summary sentence shows the paper, page and passage it came from. The operator flags values that disagree between sources. Your scientist checks, edits and signs off before anything is shared.

Is it a validated systematic review tool?

No. It follows your protocol and keeps a screening log with a reason for every decision, but it is not validated software. Your team decides what is included and signs off the table and the review.

Can it watch the literature for us?

Yes. A scheduled workflow can run the same searches every week for an indication, target or competitor and draft a digest with links and passages. Posting it to Slack or SharePoint waits for a person's yes.

Is our research data private?

Alomana is ISO 27001 certified and GDPR-compliant, and your data is never used to train AI models. Enterprise deployments run single-tenant, in a dedicated instance, and every run is logged with input, output, tools and user.

Bring one review question from last quarter.

Talk it over with sales: what the operator needs from your systems, and how it goes live.

Talk to salesStart free

Alomana is a secure enterprise AI platform where autonomous agents, called operators, run business processes end to end. Every run is fully auditable.