Accelerating Literature Reviews with Large Language Models (LLMs): An Evaluation of Performance and Efficiency

Gautam R¹, Anwar S², Srivastava T¹, Pandey R²

1. ¹ConnectHEOR, London, United Kingdom
2. ²ConnectHEOR, Delhi, India

ISPOR 2026, May 2026, Philadelphia, PA, USA.

Independent evaluation across four literature reviews. Screening efficiency improved, while extraction accuracy ranged from 9% to 60% and complex reviews remained challenging.

Presentation code: MSR190

Evaluation of AI-Assisted Multi-Row Data Extraction for Meta-Analysis Using EasySLR

Sharma S¹, Kamboj G¹, Malik A², Rathi H²

1. ¹Skyward Analytics, Gurugram, India
2. ²EasySLR, Gurugram, India

ISPOR 2026, May 2026, Philadelphia, PA, USA.

Evaluation across nine reviews. Outcome-data recall reached 100% in seven reviews and 60% and 79% in two; complex reporting formats required human review.

Presentation code: MSR235

Assessing the Performance of Artificial Intelligence for Title-Abstract and Full-Text Screening in Systematic Literature Reviews

Kamboj G¹, Sharma S¹, Malik A², Rathi H²

1. ¹Skyward Analytics Private Limited, Gurugram, India
2. ²EasySLR, Gurugram, India

ISPOR Asia Pacific 2026, September 2026, Bangkok, Thailand.

Evaluation across four SLRs. Full-text sensitivity ranged from 83.3% to 100% and accuracy from 91.7% to 100%. Performance varied by review, with human oversight remaining necessary.

Presentation code: SA5

Can Artificial Intelligence Reliably Perform Title-Abstract Screening for Systematic Literature Reviews?

Sharma S¹, Kamboj G¹, Malik A², Rathi H²

1. ¹Skyward Analytics, Gurugram, India
2. ²EasySLR, Gurugram, India

Public abstract for ISPOR Europe 2026, November 2026, Vienna, Austria.

Evaluation across three SLRs using GPT-5 mini. Sensitivity ranged from 96.4% to 100%, with moderate precision and a continuing need for human oversight.

Presentation code: MSR110

Evaluating Automated Deduplication: A Performance Comparison of EasySLR and EndNote in Multi-Database Systematic Literature Reviews

Barman P¹, Sharma S¹, Aggarwal S¹, Kamboj G¹, Rathi H¹

1. ¹Skyward Analytics, Gurugram, India

Public abstract for ISPOR Europe 2026, November 2026, Vienna, Austria.

Comparison across five SLRs found similarly high deduplication accuracy for both tools. Inconsistent bibliographic formatting remained a challenge.

Presentation code: MSR167

Can AI Generate Comprehensive, Robust, and Efficient Search Strategies for Systematic Literature Reviews?

Aggarwal S¹, Kamboj G¹, Malik A², Rathi H²

1. ¹Skyward Analytics, Gurugram, India
2. ²EasySLR, Gurugram, India

Public abstract for ISPOR Europe 2026, November 2026, Vienna, Austria.

Search strategies evaluated in PubMed across three completed SLRs retrieved all studies in predefined reference sets with fewer citations. Human review remains necessary.

Presentation code: MSR221

A Human-centred Approach to Conducting AI-assisted Systematic Literature Reviews Using EasySLR

Kamboj G¹, Malik A², Sharma S¹

1. ¹Skyward Analytics, Gurugram, Haryana, India
2. ²EasySLR, Gurugram, Haryana, India

Authorea preprint. January 16, 2026.

Artificial Intelligence in Systematic Reviews: An Investigation Into the Impact of Eligible Studies Being Excluded by Artificial Intelligence

Bishop E¹, Sanderson A¹, Reddish K¹, Carr E¹, Shires C¹, Jelveh Z¹, King K¹

1. ¹York Health Economics Consortium, York, United Kingdom

ISPOR Europe 2025, Glasgow, Scotland (2025-11). Value in Health, Volume 28, Issue 12, S2 (December 2025).

Presentation code: MSR36

Give Us the Tools and We Will Finish the Job: How Accurate Is AI for Study Selection in Systematic Reviews?

Bishop E¹, Sanderson A¹, Reddish K¹, Carr E¹, Shires C¹, Jelveh Z¹, King K¹

1. ¹York Health Economics Consortium, York, United Kingdom

ISPOR Europe 2025, Glasgow, Scotland (2025-11). Value in Health, Volume 28, Issue 12, S2 (December 2025).

Presentation code: MSR116

Assessing the Accuracy of Artificial Intelligence in Conducting Systematic Literature Reviews

Radotra A¹, Kamboj G¹, Aggarwal S¹, Singh R¹, Rathi H^1,2

1. ¹Skyward Analytics, Gurugram, Haryana, India
2. ²EasySLR, Gurugram, Haryana, India

ISPOR Annual International Meeting 2025. May 13th - 16th, Montreal, QC, Canada. Value in Health, Volume 28, Issue S1.

Presentation code: MSR111

Can AI reliably replace human reviewers in systematic literature reviews?

Povsic M¹, Armitage EL¹

1. ¹AMICULUM, Bollington, Cheshire, UK

World EPA Congress 2025, March 5th - 6th, Amsterdam, Netherlands

Presentation code: MSR74

Assessing GPT-3.5 Turbo's performance in primary screening across diverse types of systematic literature reviews

Rathi H^1,2, Malik A², Behera DC², Kamboj G¹

1. ¹Skyward Analytics, Gurugram, Haryana, India
2. ²EasySLR, Gurugram, Haryana, India

Abstracts accepted for the 2nd Global Evidence Summit, Prague, Czech Republic. Cochrane Database of Systematic Reviews 2024; (1 Suppl 1):555.

Bridging the Gap Between Theory and Reality in Systematic Literature Reviews Using Artificial Intelligence

Rathi H^1,2, Malik A², Behera DC²

1. ¹Skyward Analytics, Gurugram, Haryana, India
2. ²EasySLR, Gurugram, Haryana, India

Value in Health. 2024 Dec 1;27(12):S467.

Presentation code: MSR149

Use of Large Language Model (LLM) for Full-Text Screening in Systematic Literature Reviews: A Comparative Analysis

Rathi H^1,2, Malik A², Behera DC², Kamboj G¹

1. ¹Skyward Analytics, Gurugram, Haryana, India
2. ²EasySLR, Gurugram, Haryana, India

ISPOR Annual International Meeting 2024. May 5th - 8th, Atlanta, GA, USA

Presentation code: MSR28

A Comparative Analysis of Large Language Models (LLM) Utilised in Systematic Literature Review

Rathi H^1,2, Malik A², Behera DC², Kamboj G¹

1. ¹Skyward Analytics, Gurugram, Haryana, India
2. ²EasySLR, Gurugram, Haryana, India

Podium presentation at ISPOR. November 12th - 15th, Copenhagen, Denmark, Presentation code: P21

Best Practices for Navigating New Venture Development Programs that Adopt the Lean Startup Methodology: A Scoping Review on U.S. National Science Foundation’s Innovation Corps™

Kim Y¹

1. ¹Industrial and Applied Research Division, Office of Research, The City University of New York, USA

IntechOpen. Published September 2, 2026.

Uses EasySLR’s De-Duplicator with manual confirmation. Screening was performed in Rayyan.

Pegunigalsidase Alfa Elfabrio® as a Long-Term Enzyme Replacement Therapy in Adults With Fabry Disease: A Systematic Literature Review

Kamboj G¹, Sharma S¹, Aggarwal S¹, Papadopoulos G², Crowley S², Rathi H¹

1. ¹Skyward Analytics, Gurugram, India
2. ²Lucid Health Consulting, Sydney, Australia

ISPOR Europe 2025. Value in Health 28(12), Supplement 1, S77.

Uses EasySLR for title/abstract and full-text screening. Research by Skyward Analytics and Lucid Health Consulting.

Presentation code: CO178

Evaluating the Impact of Intermittent Fasting on Anthropometric, Cardiovascular, and Metabolic Outcomes: A Targeted Literature Review

Radotra A¹, Kamboj G¹, Kumar N¹, Rathi H¹

1. ¹Skyward Analytics Pvt. Ltd., Gurugram, India

ISPOR Europe 2024, Barcelona, Spain. Value in Health 27(12), S47.

Title/abstract and full-text screening were performed in EasySLR. Research by Skyward Analytics.

Presentation code: CO174

Effectiveness of the interprofessional community-based rehabilitation versus standard care on quality of life in community-dwelling adults’ post-stroke: systematic review protocol

Sarmago JRT¹, Rejollo MR¹, Ortega L¹, Sulit B¹, Espina-Lavilla JA¹, Dones VC III^2,3, Gabor MGA⁴

1. ¹Cebu Doctors’ University, Mandaue City, Cebu, Philippines
2. ²University of Santo Tomas, Manila, Philippines
3. ³JBI Philippines Center for Health Research and Movement Science, Manila, Philippines
4. ⁴JBI, School of Public Health, Adelaide University, Adelaide, Australia

JBI Evidence Synthesis. Published online July 28, 2026.

Plans to use EasySLR for deduplication and screening in a review of rehabilitation after stroke. This publication reports the protocol, not completed review findings.

Factors influencing pharmacovigilance practices among health care professionals in lower-middle-income countries: a scoping review

Cesar JHA¹, Nacua CRV¹, Michael RR¹, Jabines JMC¹, Dawa JNG¹, Lizarondo L²

1. ¹Cebu Doctors’ University, Mandaue, Cebu, Philippines
2. ²JBI, The University of Adelaide, Adelaide, Australia

JBI Evidence Synthesis. Scoping review protocol, published online August 31, 2026.

Specifies EasySLR for screening records from the primary searches, with Rayyan used for citation-tracking records. Describes planned use.

How and when is implementation science considered in gene therapy research? A scoping review protocol.

Barois Z, De Geest S

OSF scoping review protocol. Registered August 7, 2026.

Specifies EasySLR as the second independent reviewer for literature screening.

Integrating an AI-Assisted Tool as an Independent Reviewer in Scoping Reviews: Reflections on an Optimized Workflow

Le Thi TG^1,2

1. ¹Department of Pediatrics, Dr. von Hauner Children’s Hospital, LMU University Hospital Munich, Germany
2. ²Stiftung Kindergesundheit, Munich, Germany

OSF study protocol. Registered March 16, 2026.

Plans EasySLR-assisted screening and extraction alongside human reviewers in a scoping review of healthy snack sales in German schools. The protocol reports no EasySLR funding and payment of standard subscription fees.