Over the years he has also sailed, hiked, biked or hitchhiked to the four corners of the world. After becoming a father, he put away his sailing gear and, with an engineer co-founder who knew how to make things like computers work, founded Xerus Medical with the aim of improving medical care by leveraging personalized patient data.
Dr. Gelinas is the CMO of a molecular diagnostic start-up that uses next-generation sequencing to elucidate the complex interactions of humans and their microbial companions.
Thank you so much for joining us! Can you tell us a story about what brought you to this specific career path?
As a young physician, the critical care specialty fascinated me. Critical care doctors have to make quick and critical clinical decisions based on often very incomplete information. They have access to an enormous amount of data and have to rapidly triage this information to make the best decisions for a given patient. As a junior physician, I often found it difficult to make sense of so much information quickly. With time and experience not only did the data become less intimidating but I started wanting access to even more. I think this happens because clinical experience creates a sort of pattern recognition expertise and that pattern recognition tends to become better with more of the right kind of data. It is well known that computers can be extremely good at pattern recognition and I (and many others) thought that ‘putting the clinical experience in a box’ would be highly desirable for patients and society at large. As we at Xerus Medical started working on this project, we noticed that the type of data experienced clinicians and ‘trained’ computers thrive upon overlap but are often very different and complementary.
A ‘clinical irritant’ that greatly influenced my current career path was that as a clinician there was no way for me to obtain objective feedback about how my patients evolved when they were no longer in my care. Siloed care makes it easy to lose track of our patients. Very negative outcome is usually accessible but more subtle outcomes (especially longer term) are more difficult if not impossible to obtain. It’s hard for a clinician, computer or any larger system to systematically improve granular care without having comprehensive and well-organized longitudinal feedback.
If you can’t measure it, you can’t improve it. — Peter Drucker
Xerus Medical was founded to collect and analyze the type of patient-centric data that humans and computers need to understand and improve clinical care.
Can you share the most interesting story that happened to you since you began your career?
As an anesthesiologist and critical care physician, I’ve had the privilege of being part of some extremely joyous as well as to many very difficult moments in people’s lives.
The most pertinent (if not necessarily entertaining) story for the perspective of this article would be how physicians process radically new medical information. Therapies which are accepted as being ‘absolutely necessary and standard of care’ can and must be changed or discarded when confronted with overwhelming proof that they don’t work. One of the barriers to healthcare workers adapting innovative new ways of doing things is the lack of feedback that proves to clinicians that new therapies perform as well as anticipated and don’t have unforeseen negative consequences.
As a young Intensivist, I became an expert in the analysis of the data derived from an invasive monitoring tool called the Pulmonary Artery Catheter (PAC). Over the years I along with other clinicians suspected that PACs seemed to cause significant mechanical and infectious complications and started to question their use. Recruitment for trials attempting to evaluate the usefulness of these diagnostic tools was notoriously difficult to do because many clinicians found it ethically unacceptable not to use PAC in critically ill patients. Eventually, trials showed no benefit (or even harm) in patients who were routinely monitored using PACs, but surprisingly most clinicians wouldn’t change their practice. Eventually, PAC use slowly started to decline, and then they abruptly stopped being routinely used. I think physicians would have transitioned away from this invasive technology a lot faster if they could have seen that outcomes weren’t negatively affected if they relied less on invasive hemodynamic monitoring. I also painfully learned not to build up so much expertise in something you believe doesn’t have much of a future and that advocating against something which is part of your professional claim to fame is not a good career builder.
It took the medical system almost 20 years to gather the data necessary to show that most patients weren’t benefiting or were harmed by PAC monitoring technology. Because of the way data is currently collect and managed we could now potentially reach these types of important conclusions in a small fraction of the time it historically took us. Modern data management and analysis will radically improve the way we design and perform many types of clinical trials. Quality improvement will become deeply integrated with medical care, and we will look back amazed that it was not always so.
Can you tell us about your “Big Idea That Might Change The World”?
Medical care is currently in the early stages of being massively transformed by the same IT forces that are disrupting so much of the human experience. In some way’s healthcare is trying to catch up to the consumer market. Leveraging medical data to improve care and make it more efficient is seen as being one of the next big innovation and commercial opportunities. Consumer and medical data are also starting to intersect and merge. Giant IT companies now routinely know more about patients than the doctors directly caring for them.
We at Xerus Medical want to gather and analyze the type of patient-centric data that enables humans and machines to synergistically improve care. We automatically analyze the myriad streams of sensor data, both from consumer and purpose-built medical sensors, integrate them with electronic medical records, and present the information in a meaningful way for clinicians (and patients) so that they can make better decisions that result in better patient outcomes.
How do you think this will change the world?
Even though statements like ‘data is the new oil’ are now somewhat cliché, this does not make them less true. Xerus Medical strongly believe that wise collection and analysis of various ‘uncorrelated’ but patient-centric data streams will revolutionize how and where healthcare workers and patients interact and the therapeutic options they choose. The data we collect and analyze enables clinicians to better understand how post-surgical pain, nausea, sleep disturbances and mobility interact and are often codependent. Patients experiencing significant postoperative pain, nausea and anxiety tend to move less and recuperate more slowly. Pain and nausea treatments can potentially over sedate patients and also decrease mobility and recovery. We believe the effects of sleep disruption on perioperative recovery are meaningfully underestimated in part because we do not routinely measure sleep quality in the perioperative context and do not consider it when trying to understand the post-surgical patient experience. By striving to understand the patient’s surgical experience in a holistic manner we believe we can meaningfully improve not only postoperative pain, nausea, mobility, and quality of sleep but that we can also speed up postoperative recovery and decrease appropriate discharge times.
Keeping “Black Mirror” and the “Law of Unintended Consequences” in mind, can you see any potential drawbacks about this idea that people should think more deeply about?
We don’t have to gaze into the distant future to see the potential pitfalls of individuals collecting and sharing personal data with various commercial service providers. Consumer-focused companies and medical professionals can look at the same information and see very different things. When do mobility and other wearable sensor-derived data stop becoming a consumer fitness product and becomes a medically useful diagnostic tool? Discordant ways of labeling data can and does make a big difference in what can be done with an individual’s data. Healthcare service providers have some of the most stringent privacy standards, but social network and search engine companies handle private data in very uneven ways. Internet service providers are tightly regulated, and some consumer wearable companies like Apple are fairly stringent with their privacy rules while others may not be.
Some insurance companies now routinely ‘request’ various levels of access to personal data. What might happen if the government or insurance companies “insisted” (with or without individual consent) on having access to consumer device data? Could they then micromanage lifestyle choices, even down to an individual’s actions? For example, could patients be denied certain types of care unless they take at least 5000 steps a day for the next month? Insurance companies might insist that you get up from your desk job and walk at least once every hour or else your premiums will go up. Could your employer ask to see how many hours you slept the night before or analyze your walking patterns seven days a week to find out if you were ever under the influence of alcohol or drugs?
Was there a “tipping point” that led you to this idea? Can you tell us that story?
I had just finished working as an anesthesiologist at a daycare surgery center and as I was leaving a nurse told me that she was impressed that my patients generally did not have much nausea postoperatively. I went home feeling proud. The next week another nurse took me aside and said that I should do something to try and minimize my patient’s postoperative nausea — same-day surgery center. The same type of patient population and I changed nothing in my medical practice in that brief time interval. I started giving more anti-nausea prophylaxis and asked the nurses more often how my patients were doing. The answers seemed almost random. One day a nurse mistook me for another anesthesiologist and gave me the ‘nausea rating’ he should have gotten. Most days my patients ended up with so many different post-anesthetic care nurses that it was impossible for me to obtain a realistic evaluation.
It became clear to me that the basic quality of care feedback needed to become widespread if we were to meaningfully and reproducibly improve care.
Based on the future trends in your industry, if you had a million dollars, what would you invest in?
Companies that can efficiently and safely, from a patient’s perspective, integrate vast amounts of clinically relevant data and present the patient/ consumer, and his designated care providers, comprehensible and actionable information.
Which principles or philosophies have guided your life? Your career?
Question everything, read avidly and widely, learn and don’t take conventional wisdom as being the truth.
At the same time keep well in mind that no matter how knowledgeable, informed and good you become, if you aren’t a nice person to work with things will be much harder than they have to be.
Can you share with our readers what you think are the most important “success habits” or “success mindsets”?
Determination, smart, focused work habits and a focus on developing people skills.
Some very well-known VCs read this column. If you had 60 seconds to make a pitch to a VC, what would you say?
Xerus wants to transform perioperative care by applying AI and analytics to multiple uncorrelated but patient-centric data streams. We are particularly interested in understanding the patient’s perioperative journey using wearable and environment sensor. We welcome interest and participation in our venture.