Last fall, the Delaware Health Information Network (DHIN) announced a collaboration with the Lexington, Mass.-based biospecimen collection company iSpecimen, with the goal to allow hospitals and labs in the DHIN network to repurpose remnant clinical specimens into medical research programs. According to both organizations, the specimens involved are samples that would otherwise be discarded once patient testing is complete. The organizations also announced Maryland-based Union Hospital as the first DHIN member to participate in the joint program.
Indeed, through this program, healthcare information from participating DHIN members is de-identified and matched to corresponding remnant specimens at their provider sites. The specimens and their data are then searchable in real-time using iSpecimen's technology. When matches are made, iSpecimen's technology instructs laboratory personnel at participating provider sites to pick and ship the specimens to iSpecimen's research customers instead of discarding them as they usually do once clinical testing is complete, explains Jan Lee, M.D., CEO of DHIN.
Lee notes that the iSpecimen initiative began at Brigham and Women's Hospital in Boston, as its hospital leaders were looking for ways to further their own research. They then realized that they had specimens that are being collected for clinical purposes right in front of them that will get thrown away as soon as tests have been run. What’s more, they realized that some of those specimens probably meet the requirements of their research work, Lee says. “So rather than go out and pay a specimen bank to order specimens that fit certain conditions, it makes sense to first check to see if you have any in your own hospital that are about to get thrown away. That’s where it all got started, and they used the information that was in their own hospital electronic health record (EHR) system to identify the specimens that would be of research interest,” Lee says.
Lee further explains the process: “Let’s say you’re looking for a specimen of female diabetics over the age of 50 with a hemoglobin A1C (HbA1c), for example. What you would want to be able to do is query your data sources and see if you have any specimens sitting around in your lab that meet those conditions. And then if the answer is yes, go snag that one, and the researcher doesn’t need any of the PHI [protected health information]—just that this is a specimen from a female diabetic over the age of 50 with an HbA1c over 9. So [the specimen] is de-identified before being provided to the researcher, and it’s a way of turning trash to something that has potential to further medical research.”
Jan Lee, M.D.
The idea, Lee, continues, is to be looking for earlier and more accurate diagnoses, or treatment modalities for conditions of interest. With genomics becoming a hotter topic all the time, and the whole idea of personalized medicine, the more you can identify what specific personal traits are correlated with successful treatment with this agent versus that agent, the better you’ll be, Lee says. “Right now you can say that 75 percent of patients have a good outcome with XYZ, but what about the 25 percent of those who don’t? What’s different about them? You can either shrug your shoulders and say life is random, or you can dig in to try to understand what is different about those 25 percent of patients that don’t see the benefit. What would give them the benefit? If not this, then what? The availability of a larger pool of specimens that meet very particular defined characteristics will accelerate the ability to tease out the ones that aren’t in the middle of the bell curve,” she says.
Indeed, iSpecimen executives began to see the applicable benefits for other hospitals and lab settings, in addition to Brigham and Women's. Since that initiative, it has set up as a commercial company and has been offering this service to other hospitals, Lee notes. “They approached us,” Lee recalls. “As the Delaware state HIE [health information exchange], they realized that we have data that goes beyond any one hospital. And you need the data to identify the specimens that you are interested in out of the state HIE. We have data from every hospital and every commercial lab in Delaware. If they used our data source to identify the specimens of interest, they will get a higher hit rate. So that was why they approached DHIN,” she says.
From DHIN’s side, Lee says that the organization was very excited about the collaboration because its senior leaders believe that there is value beyond point of care decision making in the aggregation of all the clinical data. “Clearly, the first and most critical primary use of the aggregated data is to support point-of-care decision making, creating that longitudinal health record that spans geographies, time and place, and aggregates data about a patient so you don’t have to go on a treasure hunt every time you need to know things that have happened in the past in order to make good decisions for a patient,” Lee says. “That’s always been our primary mission and purpose,” she adds.
A Sustainable Business Model
Hospitals choosing to take part in the specimen collection program will be making a critically important contribution to the acceleration of biomedical research and also benefit from reduced DHIN fees, which may be put back into hospital operations, according to DHIN officials. Data senders will pay a transaction fee when they send the data into DHIN, which then delivers it to the ordering provider on their behalf. What’s more, DHIN also archives it in the searchable longitudinal community health record so clinical users of the system have the ability to search for a patient and see the aggregated data of all the places they may have received treatment, Lee says.
“What we would really love is for the day to come when no one has to pay to send us data because the users of the data are generating sufficient revenue to support the operation of the network,” Lee says. “Part of way we are approaching it now, with our collaboration with iSpecimen and Union Hospital, is to say that whatever fees we collect, and the revenue we collect out of the revenue sharing agreement we have with iSpecimen, will largely go to offset the participation cost for the hospital in the network,” Lee explains.
Officials also point out that the patient has a central role in this program. Only samples from patients who have expressly consented to participate are considered for an active research project. All samples from patients who have not provided consent are excluded and suppressed from consideration, they note. “We did insist that we would only agree to partner if the patients know that their specimens are potentially being used in this way,” Lee says, adding that one of the things that has to be [better] addressed is collecting consent on the front end. “You don’t know what might be a specimen of interest until you have done queries and identified it. There is something creepy about the idea of someone taking your bodily fluids or tissues and running the studies without you knowing about it. You have to let them know that the specimen they give could be of use to researchers, and operationalizing that is something we need to improve,” Lee admits.
Currently, amongst DHIN patient care organizations, only Union Hospital is on board with the iSpecimen program, though Lee notes that a regional pathology lab has shown strong interest. There are many challenges in getting more organizations linked up with the collaborative, Lee says. “It largely comes down to informed consent. We have approached the other hospitals in our network, and the response has been tepid in some cases, and actively opposed in other cases. One lab said they think it will be a burden to collect the upfront consent and hold the specimens; there are mechanics involved that made them uninterested. They may or may not come around in time. Everyone has so much on their plate, this industry is in such a state of volatility, so something that is extra and isn’t necessary is tough to ask of people,” she says. Still, Lee has a goal of getting one new organization a year on board until everyone has agreed to join, further noting that legal agreements and procedural issues take a lot of time. “Of course, we would love to see all of the hospitals and labs participating, but that’s not [practical] right now.”