Solving data integration key to success for PMI
Disparate data from a variety of sources needs to be combined with information from EHRs info to reveal new insights, says NIH’s Francis Collins.
The Precision Medicine Initiative, the Obama administration’s effort to map the genomes of a million Americans and make the data available to researchers, is about much more than genomics, according to National Institutes of Health Director Francis Collins, MD.
Achieving the vision of the PMI will require significant advances in data integration, and NIH is focusing efforts to make advances in this arena.
NIH plans to create a participant technologies center to leverage the ability to collect data from mobile phone and sensor technologies for the study. Collins acknowledged that one of the challenges that the PMI cohort must address is how to synthesize EHR, genomic, mHealth, and wearable sensor data.
“We’ve not been able to do that kind of integration of all those data types together with electronic health records,” he added. “There are no laws of physics here that will get in the way of our being able to achieve that. It’s just going to be a very tough data wrangling problem.” However, Collins argued that “if you can make it work for a million people, well you ought to be able to make it work for 330 million” Americans.
“This is not just the Human Genome Project 2.0,” said Collins this week at an event on big data and precision medicine hosted by Bloomberg Government and software vendor SAP in Washington. “There’s a lot of other information that we need to collect about individual lifestyle, diet and exercise, behaviors, socio-economic status and such things as environmental exposures. If you really want to understand health, you can’t look at it from just one perspective. You’ve got to put all these together. That’s what we aim to do.”
The goal is to enroll 79,000 cohort participants directly or through recruiting partnerships with healthcare provider organizations by the end of this year, and ultimately one million or more U.S. volunteers by 2019 for the landmark longitudinal study.
“We will ask them to make available their electronic health records, information about them that can be derived from those records in terms of medical diagnoses, lab results and what medications they are on,” said Collins. “We will ask them potentially to be involved in wearing all kinds of sensors that keep track of what’s happening to them over the course of time.”
Vanderbilt University Medical Center was recently awarded an NIH contract to establish a pilot exploring the optimal approaches and systems for engaging, enrolling and retaining participants from across the country in the PMI cohort. The university is working in collaboration with Verily (formerly Google Life Sciences) to enable remote participation for “anybody in the United States to raise their hand and say ‘I want to be part of this,’ ” according to the NIH director.
Under the Direct Volunteers Pilot Studies, Vanderbilt will create and optimize a prototype informational website that is engaging to a diverse array of potential volunteers, and develop an interface for obtaining consent and basic enrollment and health information using methods that are efficient, effective and secure.
Collins described Verily’s role in the pilot as providing their expertise in how to set up a web-based platform that is user-friendly and actively engages volunteers for this large-scale, long term study. “We want this to be not a clunky website that everybody gets tired of after the first half hour.” By September or October, he expects the online site to be operational.
Unlike other studies, he said that the PMI cohort will entail two-way communication with participants about what is learned. “We aim to give them a lot of information back about themselves,” asserted Collins. “This is different than a lot of research projects where you give consent and maybe a blood sample and you never hear anything. That’s not how this is going to be. We’re going to have constant contact back and forth.”
At the same time, he emphasized that there will be “intense attention” paid to ensuring the privacy and security of health data shared by volunteers “because if we screw that part up, we will lose the confidence of the American public quickly.”
Achieving the vision of the PMI will require significant advances in data integration, and NIH is focusing efforts to make advances in this arena.
NIH plans to create a participant technologies center to leverage the ability to collect data from mobile phone and sensor technologies for the study. Collins acknowledged that one of the challenges that the PMI cohort must address is how to synthesize EHR, genomic, mHealth, and wearable sensor data.
“We’ve not been able to do that kind of integration of all those data types together with electronic health records,” he added. “There are no laws of physics here that will get in the way of our being able to achieve that. It’s just going to be a very tough data wrangling problem.” However, Collins argued that “if you can make it work for a million people, well you ought to be able to make it work for 330 million” Americans.
“This is not just the Human Genome Project 2.0,” said Collins this week at an event on big data and precision medicine hosted by Bloomberg Government and software vendor SAP in Washington. “There’s a lot of other information that we need to collect about individual lifestyle, diet and exercise, behaviors, socio-economic status and such things as environmental exposures. If you really want to understand health, you can’t look at it from just one perspective. You’ve got to put all these together. That’s what we aim to do.”The goal is to enroll 79,000 cohort participants directly or through recruiting partnerships with healthcare provider organizations by the end of this year, and ultimately one million or more U.S. volunteers by 2019 for the landmark longitudinal study.
“We will ask them to make available their electronic health records, information about them that can be derived from those records in terms of medical diagnoses, lab results and what medications they are on,” said Collins. “We will ask them potentially to be involved in wearing all kinds of sensors that keep track of what’s happening to them over the course of time.”
Vanderbilt University Medical Center was recently awarded an NIH contract to establish a pilot exploring the optimal approaches and systems for engaging, enrolling and retaining participants from across the country in the PMI cohort. The university is working in collaboration with Verily (formerly Google Life Sciences) to enable remote participation for “anybody in the United States to raise their hand and say ‘I want to be part of this,’ ” according to the NIH director.
Under the Direct Volunteers Pilot Studies, Vanderbilt will create and optimize a prototype informational website that is engaging to a diverse array of potential volunteers, and develop an interface for obtaining consent and basic enrollment and health information using methods that are efficient, effective and secure.
Collins described Verily’s role in the pilot as providing their expertise in how to set up a web-based platform that is user-friendly and actively engages volunteers for this large-scale, long term study. “We want this to be not a clunky website that everybody gets tired of after the first half hour.” By September or October, he expects the online site to be operational.
Unlike other studies, he said that the PMI cohort will entail two-way communication with participants about what is learned. “We aim to give them a lot of information back about themselves,” asserted Collins. “This is different than a lot of research projects where you give consent and maybe a blood sample and you never hear anything. That’s not how this is going to be. We’re going to have constant contact back and forth.”
At the same time, he emphasized that there will be “intense attention” paid to ensuring the privacy and security of health data shared by volunteers “because if we screw that part up, we will lose the confidence of the American public quickly.”
More for you
Loading data for hdm_tax_topic #better-outcomes...