HP's planned commitment to provide funding will support efforts to find trends and patterns in clinical, genetic, and environmental data that solve clinical challenges. In the partnership, LPCH will contribute physician scientists with clinical expertise and HP Labs will offer researchers with data analytics skills.
Since 2007, both organizations have developed groundbreaking research using bioinformatics tools and the hospital's digitized patient records which has already produced significant progress in identifying signs of an oncoming cardiac arrest in children.
With fewer than 30% of patients leaving the hospital alive after such an event, LPCH and HP jointly worked toward developing an approach for automated risk assessment screening in non-ICU pediatric inpatients using existing electronic clinical data. They identified high-risk markers and patterns in the electronic medical record (EMR) data that can signal an ongoing clinical deterioration, a first step towards automatic real-time identification of pediatric patients at elevated risk for cardiac arrest.
Among the data examined, 85,091 clinical laboratory tests were measured, and 69,795 clinical events were recorded. Data that included everything from systolic blood pressure to clergy visits were analyzed, as well as liver function tests, pain control, infection/immunology, and medication dosing.
According to Sharad Singhal, senior scientist at HP Labs, LPCH and HP researchers used the data to establish a system of prompts for deployment of rapid response teams (RRT) consisting of trained medical personnel available to evaluate, stabilize, and if necessary, transfer to a higher level of care patients who are considered to be at high risk.
According to a LPCH report, the implementation of a RRT "was recently shown to reduce hospital-wide mortality and non-ICU code rates in our children's hospital, saving the lives of 33 patients over the first 19 months."
"The system our researchers developed together was designed to use sophisticated analytics to give predictive notice to doctors and nurses so they could address the problem right away," Singhal said in an interview.
He also said HP's funding would be used to purchase new technology to sustain the joint research effort.
"The technology will include both hardware and software, including advanced HP high-performance computers that will process innovative HP predictive analytics to analyze clinical and operations data to ultimately improve treatment protocols and safety procedures," Singhal said.
Additionally, the partnership has developed a cutting-edge computing system to detect hidden patterns and extract useful information from the vast amounts of medical data now available from gene sequencing, clinical testing, and lab reports.
"For example, these sophisticated data mining capabilities have led to new blood tests that a doctor can use to predict how a child with cancer will react to chemotherapy drugs. This enables the doctor to calculate the safest formula and dosage for that patient," Singhal added.
Christopher Longhurst, chief medical officer at LPCH, said the hospital, which uses Cerner's EMR system, has fully digitized almost all aspects of inpatient care, and noted that researchers and hospital leaders are now asking how they can use these large quantities of data to drive further advancements in patient care and hospital operations.
"For example, a novel partnership now underway is investigating the use of real-time dashboards with patient safety and care quality indicators derived from the electronic health record to improve adherence to evidence-based guidelines," Longhurst said.
Nominate your favorite mobile health or telehealth application as part of our Mobile Application Smackdown competition.