big data

Archive for big data

Sidelined by mindlines?

Evidence-based medicine (EBM), a movement that emerged roughly 30 years ago, advocates for the use of current best evidence from high quality research studies in healthcare decision making. This logical and straightforward way of delivering healthcare often fails in modern day practice. One simple reason that clinicians cannot execute point of care decision making with EBM is due to the overwhelming volume of scientific evidence that is ever changing and available within severe time constraints. A more pervasive reason is found in the way clinicians practice and incorporate knowledge into their daily work – they tend to follow what ethnographers Gabbay and Le May have coined as mindlines: collectively reinforced, iterative, internalized, and tacit guidelines. Clinicians’ practice is primarily influenced by trusted colleagues, mediated by cultural and organizational features of their practices, and is constantly refined as knowledge-in-practice-in-context.

Through my own wandering through various clinical settings, I have often heard phrases from respected clinicians including “there is evidence…and then there is actual practice.” The five part concept of EBM appears intuitively important in a science-based profession – define the problem, search for sources of information, critically evaluate that information, apply the information to the patient encounter, and evaluate the efficacy of the application of that information for that specific patient. It seems that an exciting opportunity would be data analytics enhanced by artificial intelligence that could search high volume clinical research and identify patient-matching criteria in order to assist clinician judgment on relevant treatment protocols.

How much of this is naïve rationalism? Upon evaluating a typical clinical scenario, what I used to think was a clear set of facts in a one-dimensional reality is now more like an interaction of temporary realities of patients, clinicians, researchers, and guideline/policy makers. Mindlines are therefore:

  • More than intuition.
    Mindlines that clinicians abide by undergo a validation process despite being mainly tacit. They are built off of shared sense-making in the local settings of patient care, which leads to coherence and negotiation with real-time environmental influences. They provide for more accuracy than the reductionist tools and beliefs of EBM.
  • More patient centered.
    Mindlines allow for incorporation of valid knowledge to occur from the patient’s perspective, as opposed to the paternalistic model of clinician knowing all and only being able to derive more information from EBM.
  • Meaningful and effective.
    Mindlines are not very far off from the way typical high performers solve problems – they consciously and unconsciously adjust their frameworks through contextual experience, colleagues, and the physical world. EBM can negotiate with these frameworks, but likely can never replace them.

The paradigm of mindlines offers insight into the way clinicians practice and how western medicine operationally works in an environment with varying expectations from the patient and the overall industry where innovative work is being attempted. The secular trend for the future hopefully will be the risk-adjusted incorporation of EBM with assistance from artificial intelligence into the tacit world of clinical medicine.

Posted in: big data, Clinical Research, Research

Leave a Comment (0) →

Dispatches from the Canadian E-Health Conference: The same but different

Bear statue in VancouverThe annual Canadian E-Health Conference was held in Vancouver, BC last week. I had the opportunity to speak about the work we’re doing at Wellpepper in applying machine learning to patient-generated data, and in particular the insights we’ve found from analyzing patient messages, and then applying a machine-learned classifier to alert clinicians when a patient message might indicate an adverse event. Our goal with the application of machine-learning to patient generated data is to help to scale care. Clinicians don’t need to be alerted every time a patient sends a message; however, we don’t want them to miss out if something is really important. If you’d like to learn more about our approach, get in touch.

My session was part of a broader session focused on ‘newer’ technologies like machine-learning and blockchain, and some of the other presenters and topics definitely highlighted key differences between the US and Canadian systems.

Aside from the obvious difference of Canada having universal healthcare, there were subtle differences at this conference as well. While the same words were used, for the most part: interoperability, usability, big data, and of course blockchain and AI, the applications were different and often the approach.

Interoperability: Universal doesn’t mean one

Each province has their own system, and they are not able to share data across provinces. Unlike the UK which has a universal patient identifier, your health records in Canada are specific to the province you live in. As well, apparently data location for health records is sometimes not just required to be in Canada, but in the actual province where you reside and receive care. As for interoperability, last we heard, British Columbia was doing a broad roll out of Cerner while large systems in Alberta were heading towards EPIC, so Canada may see the same interoperability challenges we see here if people move between provinces.

Privacy: The government is okay, the US is not

What’s interesting is as a US company, is that whenever we talk to health systems in Canada they bring up this requirement, but as soon as you mention that the PIPEDA requirements enable patients and consumers to give an okay for out of Canada data location they agree that it’s possible. Regardless, everyone would rather see the data in Canada.

What was possibly the most striking example of a difference in privacy was from one of my co-presenters in the future technologies session, who presented on a study of homeless people’s acceptance of iris scanning for identification. 190 out of 200 people asked were willing to have their irises scanned as a means of identification. This identification would help them access social services, and healthcare in particular. The presenter, Cheryl Forchuk from the Lawson Health Research Institute said that the people who participated didn’t like to carry wallets as it was a theft target, that they associated fingerprinting with the criminal justice system, and that facial identification was often inaccurate due to changes that diet and other street conditions can make. When I tweeted the 95% acceptance rate stat there were a few incredulous responses, but at the same time, when you understand some of the justifications, it makes sense. Plus, in general Canadians have a favorable view of the government. The presenter did note that a few people thought the iris scan would also be a free eye exam, so there may have been some confusion about the purpose. Regardless, I’m not sure this type of identification would play out the same way in the US.

Reimbursement: It happens, just don’t talk about it

The word you didn’t hear very much was reimbursement or when you did, from a US speaker the audience looked a bit uncomfortable. The funny thing is though, that physicians have billing codes in Canada as well. It’s just that they are less concerned about maximizing billing versus being paid for the treatment provided and sometimes even dissuading people from over-using the system. Budgets were discussed though, and the sad truth that money is not always smartly applied in the system, and in a budget-based system, saving money may decrease someone’s future budget.

Blockchain: It’s not about currency

Probably the biggest difference with respect to Blockchain was the application, and that it was being touted by an academic researcher not a vendor. Edward Brown, PhD from Memorial University suggested that Blockchain (but not ethereum based as it’s too expensive) would be a good way to determine consent to a patient’s record. In many US conferences this is also a topic, but the most common application is on sharing payer coverage information. Not surprisingly this example didn’t come up at all. If you consider that even though it is a distributed ledger, a wide scale rollout of Blockchain capabilities for either identification or access might be more likely to come from a system with a single payer. (That said, remember that Canada does not have a single payer, each province has its own system, even if there is federal funding for healthcare.)

“E” HR

Physician use of portalFor many of the session the “E” in e-health stood for EHR, which while also true in the US, the rollout of wide scale EHRs is still not as advanced. Cerner and EPIC in particular have only just started to make inroads in Canada, where the a telecommunications company is actually the largest EHR vendor. In one session I attended, the presenter had done analysis of physician usage of a portal that provided access to patient labs and records, but they had not rolled out, what he was calling a “transactional” EHR system. Physicians mostly accessed patient history and labs, and felt that if the portal had prescribing information it would be perfect. Interesting to see this level of access and usage, but the claim that they didn’t have an EHR. What was also interesting about this study is that it was conducted by a physician within a health system rather than an academic researcher. It seemed like there was more appetite and funding for this type of work within systems themselves.

Other Voices: Patients!

Patients on the mainstageDuring the interlude between the presentations and judging for the well-attended Hacking Health finals, and on the main stage, presenters interviewed two advocate patients. While they said this was the first time they’d done it, both patients had been at the conference for years. So while the mainstage was new, patient presence was not, and patient advocate and blogger Annette McKinnon pushed attendees to go further when seeking out engaged patients. Noting that retirees are more likely to have the time to participate in events she asked that they make sure to seek out opinions for more than 60 year old white women.

There was also an entire track dedicated to First Nations Healthcare. Think of the First Nations Health authority as a VA for the indigenous people of Canada, which incorporates cultural differences and traditional practices of the First Nations people. The track started and concluded with an Elder song and prayer.

Manels

Speaking of diversity, I didn’t witness any manels.

Best Quote

 

Posted in: big data, Clinical Research, Health Regulations, Healthcare Disruption, Healthcare Research, Healthcare Technology, Healthcare transformation, Interoperability, M-health, patient-generated data

Leave a Comment (0) →

Are Women Better Surgeons? Patient-Generated Data Knows The Answer

As empowerers of patients and collectors of patient-generated data, we’re pretty bullish on the ability for this data to show insights. We fully admit to being biased, and view things through a lens of the patient experience and outcomes, which is why we had some ideas about a recent study that showed female surgeons had better outcomes than male surgeons.

The study, conducted on data from Ontario, Canada, was a retrospective population analysis of patients of male and female surgeons looking at rates of complications, readmissions, and death. The results of the study showed that patients of female surgeons had a small but statistically significant decrease in 30-day mortality and similar surgical outcomes.

Does this mean that women are technically better surgeons? Probably not. However, there is one sentence that stands out to a possible reason that patients of female surgeons had better outcomes.

A retrospective analysis showed no difference in outcomes by surgeon sex in patients who had emergency surgery, where patients do not usually choose their surgeon.

This would lead us to believe that there is something about the relationship between the patient and the provider that is resulting in better outcomes. We have seen this at Wellpepper, while we haven’t broken our aggregate data down by gender lines, we have seen that within the same clinic, intervention, and patient population, we see significant differences in patient engagement and outcomes between patients being seen by different providers.

Some healthcare professionals are better than others at motivating patients, and the relationship between provider and patient is key for adherence to care plans which improve outcomes. By tracking patient outcomes and adherence by provider, using patient-generated data, we are able to see insights that go beyond what a retroactive study from EMR data can show.

While our treatment plans, and continued analysis of patient outcomes against those treatment plans go much further than simply amplifying the patient-provider relationship, for example with adaptive reminders, manageable and actionable building blocks, and instant feedback, never underestimate the power of the human connection in healthcare.

Posted in: Adherence, Behavior Change, big data, Clinical Research, patient-generated data

Leave a Comment (0) →

In Defense of Patient-Generated Data

There’s a lot of activity going on with large technology companies and others trying to get access to EMR data to mine it for insights. They’re using machine learning and artificial intelligence to crawl notes and diagnosis to try to find patterns that may predict disease. At the same time, equal amounts of energy are being spent figuring out how to get data from the myriad of medical and consumer devices into the EMR, considered the system of record.

There are a few flaws in this plan:

  • A significant amount of data in the EMR is copied and pasted. While it may be true that physicians and especially specialists see the same problems repeatedly, it’s also true that lack of specificity and even mistakes are introduced by this practice.
  • As well, the same ICD-10 codes are reused. Doctors admit to reusing codes that they know will be reimbursed. While they are not mis-diagnosing patients, this is another area where there is a lack of specificity. Search for “frequently used ICD-10 codes”, you’ll find a myriad of cheat sheets listing the most common codes for primary care and specialties.
  • Historically clinical research, on which recommendations and standard ranges are created, has been lacking in ethnic and sometimes gender diversity, which means that a patient whose tests are within standard range may have a different experience because that patient is different than the archetype on which the standard is based.
  • Data without context is meaningless, which is physicians initially balked about having device data in the EMR. Understanding how much a healthy person is active is interesting but you don’t need FitBit data for that, there are other indicators like BMI and resting heart rate. Understanding how much someone recovering from knee surgery is interesting, but only if you understand other things about that person’s situation and care.

There’s a pretty simple and often overlooked solution to this problem: get data and information directly from the patient. This data, of a patient’s own experience, will often answer the questions of why a patient is or isn’t getting better. It’s one thing to look at data points and see whether a patient is in or out of accepted ranges. It’s another to consider how the patient feels and what he or she is doing that may improve or exacerbate a condition. In ignoring the patient experience, decisions are being made with only some of the data. In Kleiner-Perkin’s State of the Internet Report, Mary Meeker estimates that the EMR collects a mere 26 data points per year on each patient. That’s not enough to make decisions about a single patient, let alone expect that AI will auto-magically find insights.

We’ve seen the value of patient engagement in our own research and data collected, for example in identifying side effects that are predictors of post-surgical readmission. If you’re interested, in these insights, we publish them through our newsletter.  In interviewing patients and providers, we’ve heard so many examples where physicians were puzzled between the patient’s experience in-clinic or in-patient versus at home. One pulmonary specialist we met told us he had a COPD patient who was not responding to medication. The obvious solution was to change the medication. The not-so-obvious solution was to ask the patient to demonstrate how he was using his inhaler. He was spraying it in the air and walking through the mist, which was how a discharge nurse had shown him how to use the inhaler.

By providing patients with useable and personalized instructions and then tracking the patient experience in following instructions and managing their health, you can close the loop. Combining this information with device data and physician observations and diagnosis, will provide the insight that we can use to scale and personalize care.

Posted in: Adherence, big data, Clinical Research, Healthcare Disruption, Healthcare Research, Healthcare Technology, Healthcare transformation, Interoperability, M-health, patient engagement, patient-generated data

Leave a Comment (0) →

HIMSS 2017 Recap: What’s Hot and What’s Hype

Wellpepper had a great HIMSS 2017 Conference with a very busy booth in the Innovation Zone, a panel on the current state of innovation, and a talk on Delivering Empathy Through Telehealth. Here are a few of our thoughts on the conference compiled from our team.Empathetic Care Through Telehealth

Cognitive and AI: Hype

Starting with Ginni Romety’s keynote, Cognitive and AI were definitely the buzzwords of the conference. Everyone is excited about the promise but it seems like the current status is not ready for takeoff. First, there’s a lot of work to get data out of the EMR, and second, no one seems quite sure what the killer use case is going to be. Immediately before HIMSS, MD Anderson announced that after a $62M investment they weren’t seeing value in IBM Watson and were pulling out of the program. That did not stop them from co-presenting with Mayo Clinic and Watson at the conference. The main use case seemed to be shortening the time to identify cancer patients for clinical trials from 30 minutes to 8 minutes. Another example, which just highlights the sorry state of clincial technology, was to use Watson on top of Epic to help staff figure out how to use features. During the session, Mayo CIO Christopher Ross referred to Watson as a toddler. While all of this was disappointing, it’s heartening that for once healthcare is on trend with the rest of the tech world, and possibly pointing to an accelerated evolution of health IT.

IMG_0611Patient Engagement: Hot

In 2016, patient engagement was also hot, but this year, we’d also say it was real. Buyers visited our booth with checklists of capabilities they wanted to see. Pilots were completed last year, and now they are making platform decisions for patient engagement. We’ve noticed this ourselves in the past 6 months, we’ve seen the patient engagement purchase decision elevated to the C-suite, and the decision being made based on capabilities that will address the needs of all patients and all service lines.

Interoperability: Hot

Compared to the previous year, we saw a lot more talk about interoperability, whether that was EMRs building out APIs and developer programs, the CommonWell Alliance, or talk about how block-chain could be used to both secure and transfer healthcare data. Understanding that data needs to flow with the patient, and also that a heck of a lot of data is being created outside the EMR (in patient engagement solutions for example), is driving a greater commitment to interoperability in the industry.

Healthcare Investment: Hot

The Sharks said so, so it must be hot. The HIMSS Venture+ Investment forum this year had a much more diverse set of pitches than previously, including a social venture. and was won by DiaCardio, a woman-led company from Israel automating evaluation of heart ultrasound.

The Affordable Care Act: Prognosis Unclear

Make no mistake, the potential repeal of the ACA is looming heavy even in health IT. Health systems Boehner, HIMSSare concerned about impact on Medicare and Medicaid revenue. While bundles and value-based care have been quite positively received, the current uncertainty is putting a hold on capital expenditures. (Did we mention that Saas can be accounted for as operating expense?) Possibly the most entertaining speculation on the ACA came from former house speaker John Boehner and former governor Ed Rendell. Rendell suggested that we repeal Obamacare and replace it with the Affordable Care Act. Boehner mused that repealing without a plan would place all the blame and problems with the current system firmly on the sitting government, and recommended that it not be repealed.

The Takeway?

We’re still optimistic. IT is increasingly having a seat at the table within healthcare. Although not all EMR implementations have been seen as a success for clinicians, we are seeing a shift to an expectation of better software for both patients and providers, for data to move smoothly, and the promise of insights and better care when that data can be analyzed and acted on. We’re already looking forward to HIMSS 2018 Las Vegas.

Posted in: big data, Clinical Research, Interoperability, patient engagement

Leave a Comment (0) →

HIMSS17 Sessions of Interest

We are thrilled to attend a number of sessions at HIMSS17 with topics pertaining to Wellpepper’s Vision and Goals!

Patient Engagement

Sessions that impact our ability to deliver an engaging patient experience that helps people manage their care to improve outcomes and lower cost:

Insight from Data

Sessions that impact our ability to derive insight from data to improve outcomes and lower cost:

Clinical Experience

Sessions that impact our ability to deliver more efficient experience for existing workflows and are non-disruptive for new workflows:

 

Posted in: big data, Healthcare Technology, Interoperability, M-health, patient engagement

Leave a Comment (0) →

Our Picks for HIMSS17

himss17-exhibitor-ad-design-300x250-copyHIMSS17 is right around the corner and we at Wellpepper have a lot to be excited about! By empowering and engaging patients, deriving insight from the data we collect, and delivering new value to clinical users without major disruption to existing clinical workflows, we can continue to improve outcomes and lower costs of care. At HIMSS17, we look forward to connecting with friends, partners, colleagues and industry leaders to continue the journey towards an amazing patient experience.

Sessions that we look forward to:

Our CEO and co-founder, Anne Weiler, will be speaking at 2 sessions:

  • Anne will be a featured speaker at the Venture+ Forum, where former competition winners will be sharing how their business has grown, lessons learned and plans for the future. Since being named a winner of the 2015 Venture+ Forum Pitch competition, Wellpepper has continued to bridge the gap between the patient and care team and we are excited to share our progress and vision.
  • Anne will also be presenting a session titled, Designing Empathetic Care Through Telehealth for Seniors, which will explore the role of design-thinking in design empathetic applications to deliver remote care for seniors based on studies completed by Boston University and researchers from Harvard Medical School.

Patient engagement expert Jan Oldenburg, who was featured in our August 2016 webinar, will be speaking at 2 sessions:

  • Jan will be presenting a session titled, The “P” is for Participation, Partnering and Empowerment. This session will highlight what it takes to create a truly participatory healthcare system that incorporates patients and caregivers, using digital health technology to reinforce and support participatory frameworks.
  • Jan will also be presenting a session titled, Importance of Narrative: Open Notes, Patient Stories, Human Connections. This session will focus on how Open Notes enhance the patient’s narrative of their journey through their condition and how this both strengthens the patient-physician relationship and empowers patients to take charge of their illness and wellness.

Christopher Ross, Chief Information Officer at Mayo Clinic will be leading a session on Emerging Impacts of Artificial Intelligence on Healthcare IT. This session will discuss how the advancement of Artificial Intelligence (AI) and Machine Learning (ML) are having a profound impact on how insights are generated from healthcare data.

Posted in: big data, M-health, patient engagement

Leave a Comment (0) →

Population Health and Patient Engagement: A Reckoning Is Coming

Population health and patient engagement should be best friends. To draw conclusions for population health, you need a lot of data, and patient engagement that is, patients interacting digitally with treatment plans and healthcare providers, generates a ton of data. Population health tries to analyze the general to get to the specific and identify patients at risk. Patient engagement starts with the specific patient, and with enough data recorded by those patients, can find general trends.

With patient engagement, the information is real-time. With population health it is backwards-looking. Population health has the richness of the medical teams notes and diagnosis but it is missing the patient perspective. Patient-generated data will have diagnosis if it’s part of a treatment plan prescribed by a physician, but it won’t have the full notes. A blurring of the boundaries between population health and patient engagement presents a way forward to greater insights about both individuals and groups, and can make population health actionable at the individual patient level by providing personalized instructions (with or without care managers).

However, to get to this desired end-state, we need to clear some obstacles, first of which is the idea that patient engagement generates too much data for physicians.

Yes, an individual physician does not want to see or review each data point that a true patient engagement solution generates. However, this information can be extremely interesting to the patient, especially when looking for trends to help self-manage a chronic condition so it is worth enabling patients to collect it. For example, looking at whether certain foods trigger arthritis, or whether certain activities trigger headaches. However, to draw conclusions like this, you must record a lot of data points and in real-time, and this makes physicians nervous. They have enough to do, and not enough time to do it in, so this data cannot add to that workload.

As well, patient-generated data is messy, which can be intimidating, especially in an industry that is looking for deviations from norms. The challenge with patient-generated data is that it can uncover that the long-tail is actually longer than previously thought, that there are sub-groups within previously thought to be homogeneous groups of patients with a similar condition. In the long run, this will result in medical breakthroughs and personalized medicine. In the short run this can be difficult to deal with in the current systems.

the long-tail is actually longer than previously thought

Does that mean that we shouldn’t collect patient-generated data? Not at all. Helping patients track their experiences is a great first step to self-management. Knowing whether they are following a treatment plan, and what their experiences are with that treatment plan can help healthcare systems determine the impact of their instructions outside the clinic.

Although physicians don’t want all this data, healthcare organizations both providers and payers, should want it. Other industries would kill for this type of data. Data scientists and population health managers at health systems should be clamoring for this valuable patient-generated data.

Patient-generated data is usually collected in real-time so it may be more representative of the actual current population. The benefit of real-time collection is that further exploration of the actual patient experience is possible and can be used to prevent issues from escalating. With backwards looking data whatever was going to happen has happened, so you can only use it to impact new groups of patients not current groups.Patient-Generated Data

Finally, patient-generated data is less likely to be siloed, like clinical data often is, because the patient experience is broad and often messy and crosses clinical department thresholds (or more simply, patients are usually treated for more than one issue at a time.) Being relatively new to market, patient-engagement systems are built on modern and interoperable technology which also makes accessing data for analysis easier.

So where will we end up? To our team at Wellpepper, it seems inevitable that influencing and understanding patient experience outside the clinic. If you are making decisions for an individual patient with only a few clinical touch points, this is a very thin slice, often with a specific clinician’s specialty lenses on the actual situation. While healthcare systems are currently dipping their toes in the water on collecting and analyzing this data, if they don’t embrace the whole patient, patients will vote with their feet and pocket books towards organizations that are data and technology driven.

Posted in: Adherence, big data, Healthcare Technology, Healthcare transformation, Interoperability, M-health, patient engagement, population health

Leave a Comment (0) →
Google+