Jun 2, 2021

Most top pharma companies have "poor digital culture"

pharma
Digitaltransformation
Data
PredictiveMaintenance
2 min
New research shows that 87% of the top 300 pharma companies in Europe and the US have a poor digital culture that inhibits innovation

 A new survey has revealed that 87% of pharmaceutical manufacturing companies have a poor digital culture, and are aware it holds back digital transformation and undermines revenues. 

The research was carried out by software provider Aspen Technology among 300 senior pharmaceutical industry decision-makers in the UK, US, Germany, France, Spain and Sweden.

Just 13% of organisations surveyed fit the description of “digital culture leaders”, with confidence in adopting new technologies such as artificial intelligence (AI). 

Two-thirds of respondents (66%) admitted their company’s digital skills with tools such as AI and machine learning (ML) are either poor or merely adequate, leading to lower revenues. By comparison 38% of digital leader companies saw revenues increase by up to 10% over the last 12 months during the pandemic. Other findings include: 

  • 50% of respondents admit they are hesitant or lack confidence about AI and ML adoption
  • 35% say their companies have decentralised the kind of decision-making that generates agility and innovation
  • 25% admit they struggle with centralised and bureaucratic processes

David Leitham, senior vice president and general manager pharma at Aspen Technology, commented on the results: “The pandemic has shown how manufacturers relying on rudimentary IT such as spreadsheets cannot carry the industry forward.” 

"Companies that are unable to tap into data insights or establish predictive maintenance, along with other AI-powered technologies will not achieve the speed of development and rollout the market now demands."

Leitham added that data is vital for organisations to adapt to the market. “Digital leaders have a better sense of what data can do. Data is the thread that runs through organisations’ ability to innovate, use AI and ultimately to respond to changes in demand or market disruption effectively.”

Share article

Jun 15, 2021

 NHS trials test that predicts sepsis 3 days in advance 

sepsis
MachineLearning
clinicaltrial
blooddisorder
2 min
Queen Alexandra Hospital is trialling a new sepsis test by Presymptom Health that uses machine learning to detect the onset of the disease

A new test that can predict sepsis before the patient develops symptoms is being trialled at a National Health Service (NHS) hospital in the south of England. 

Clinicians at Portsmouth’s Queen Alexandra Hospital are leading medical trials of the blood test, which they hope will help them save thousands of lives a year. 

The test is being developed by government spin-out company Presymptom Health, but the research began over 10 years ago at the Defence Science and Technology Laboratory (Dstl). This included a study of 4,385 patients and more than 70,000 samples, the largest study of its kind at the time. 

From the samples taken, a clinical biobank and database were generated and then mined using machine learning to identify biomarker signatures that could predict the onset of sepsis. The researchers found they were able to provide an early warning of sepsis up to three days ahead of illness with an accuracy of up to 90%.

Unlike most other tests, Presymptom Health identifies the patient’s response to the disease as opposed to detecting the pathogen. This is an important differentiator, as sepsis occurs as a result of the patient's immune system’s overreaction to an infection or injury, which can then cause life-threatening organ dysfunction. 

Worldwide, an estimated 49 million people a year contract sepsis, while in the UK almost two million patients admitted to hospital each year are thought to be at risk of developing the condition. If Presymptom's test is effective, it could save billions of pounds globally and improve clinical outcomes for millions of sepsis patients.

The initial trials at Queen Alexandra Hospital will last 12 months, with two other sites planned to go live this summer. Up to 600 patients admitted to hospital with respiratory tract infections will be given the option to participate in the trial. The data collected will be independently assessed and used to refine and validate the test, which could be available for broader NHS use within two years. 

If successful, this test could also identify sepsis arising from other infections before symptoms appear, which could potentially include future waves of COVID-19 and other pandemics.

Dr Roman Lukaszewski, the lead Dstl scientist behind the innovation, said: “It is incredible to see this test, which we had originally begun to develop to help service personnel survive injury and infection on the front line, is now being used for the wider UK population, including those fighting COVID-19.”

Share article