The joint venture aims to accelerate the diagnosis process for people with rare diseases. Image credit: HVH Patient Precision Analytics website
Havas Health and Vencore formally debuted HVH Patient Precision Analytics, a joint venture soft-launched in October that pairs a communication powerhouse with a provider of defense-department-grade predictive analytics.
That description of Vencore is no exaggeration: It has long supplied organizations in the intelligence and defense communities with an array of machine-learning and predictive-modeling support. The pairing with Havas Health, then, would appear to be the first of its kind, even as healthcare and pharma players continue to embrace the data revolution with more than their usual degree of gusto.
The goal of the venture is to meaningfully accelerate the diagnosis process for people with rare diseases.
HVH differs from existing pharma/healthcare analytics plays in that it doesn’t seek to collate, curate, or otherwise collect reams of data. Rather, HVH is data-agnostic; it is set on “aggregating and analyzing the journeys of known patients,” as the company puts it in press materials, then providing tools that can help companies in and around the rare-disease space identify undiagnosed patients.
There aren’t too many organizations that can match Vencore in terms of depth of ability in, and commitment to, analytical precision. As HVH COO Jeff Ceitlin quipped, “If [Vencore] can use data and analytics to find bad guys in Afghanistan, they can use it to find [undiagnosed] patients.”
The HVH methodology has already passed its first test. In a peer-reviewed study published in The American Journal of Pharmacy Benefits, the company outlined the processes it used to analyze the characteristics of patients in its database who were already being treated for rare genetic disease hereditary angioedema (HAE). From those characteristics, HVH created a model and used it to probe the database for HAE patients who had not yet been identified.
The study was noteworthy on several levels, but none so important as providing proof of the scalability of the HVH model. “If you can do it in rare disease, you can do it in other places,” Ceitlin noted, even as he described the process as “needle-in-a-haystack math.”
Challenges abound, less around the technical details of HVH’s methodology as in the disruption it will create in numerous corners of the set-in-its-ways healthcare business. While industry players might fancy themselves analytics experts, many don’t know what they don’t know. “‘Vector-machine-learning’ doesn’t roll off the tongue real easily when you’re talking about pharma,” Ceitlin acknowledged.
Too, one can envision a scenario in which physicians chafe at accepting guidance provided by HVH algorithms. Experts in their fields generally don’t believe they need the help. Asked about this potential conflict, Ceitlin merely shrugged. “I guess you frame it for them in moral/ethical terms. ‘If you can find these patients, why wouldn’t you try?’”
As for where HVH might turn its attention in the future, Ceitlin identified the payer marketplace as an area of keen interest. Using its algorithms and data provided by payers, the company could attempt “to predict which patients will become the sickest, then hyper-prepare them in advance,” he explained.
HVH will also likely turn its attention outside the U.S. and intensify its interaction with rare-disease communities. Ceitlin noted that HVH has already met with the CEOs of Global Genes and CureDuchenne.
HVH debuted with approximately 20 full-time employees, as well as additional support from Havas and Vencore on an as-needed basis. The company is headquartered in King of Prussia, Penn., with additional office space in Boston and New York.
Havas Health first announced the partnership with Vencore in February 2016.