AI & Real World Data May publications

A new month means a new round-up of the top ten publications in the fields of AI and Real World Evidence, kindly collated by Benjamin Hughes, SVP at IQVIA. 

Pharma-published:

Roche: Speculates that COVID-19 will accelerate the use of RWE in access and evidence/data based decision making.

Life after COVID-19: R WE going to help?

Pfizer: On FDA oncology approvals (16-19), with 6 of 93 approvals levering RWE – 3 levered Flatiron data, one iKnowmed EMR, one EAP data, and one pooled RWD/RCT

Role of real-world evidence for oncology product registration in the United States: A review of approvals by the U.S. Food and Drug Administration from 2015 to 2019.

Academia and industry published:

COVID mortality: HDRUK lever the UK Caliber (CPRD/HES) network to estimate risk factors and total UK excess mortality from COVID depending in different rates of Ro

Estimating excess 1-year mortality associated with the COVID-19 pandemic according to underlying conditions and age: a population-based cohort study

US Department of Health and Human Services: On guidance of Clinical Decision Support (CDS) in EMR and the need to have full commercial separation from organizations such as pharma

Estimating excess 1-year mortality associated with the COVID-19 pandemic according to underlying conditions and age: a population-based cohort study

COVID infection: PHE/Welcome lever the UK GP surveillance database to identify risks factors for COVID (e.g., male, older, black, urban, deprived, obese, non-smoker, CKD)

Real- world assessment of the clinical impact of symptomatic infection with severe acute respiratory syndrome coronavirus (COVID-19 disease) in patients with Multiple Myeloma receiving systemic anti-cancer therapy.

External controls: A primer for researchers on synthetic and external control arms in randomised clinical trials

Synthetic and External Controls in Clinical Trials – A Primer for Researchers

COVID spread: AI model via modified auto-encoder (MAE) rather than SEIR model, estimating with late or limited intervention a potential 300m total cases and 9m global deaths

Forecasting and Evaluating Multiple Interventions for COVID-19 Worldwide

Fair Data: Various (FDA, NIH, EMA) combine with pharma to publish on the principles of FAIR data and the need for CDEs (Common data elements or models) in implementation

FAIR Data Sharing: The Roles of Common Data Elements and Harmonization

AI & firm performance: A review of 500 (non-healthcare) case studies and a reflection that AI can create immense value, but only if firms re-configure their business processes

Influence of artificial intelligence (AI) on firm performance: the business value of AI-based transformation projects

Privacy by design: A team look at a variety of latest techniques, and also demonstration of a block-chain powered distributed analytics networks for IoT healthcare apps

Enabling trustworthy personal data protection in eHealth and well-being services through privacy-by-design