“The need to see patients as people is very real. It is an ideal that will drive healthcare transformation.” Mandi Bishop (@MandiBPro).
Mandi Bishop prefers to be called the Chief Evangelist rather than the Chief Executive Officer. Her new start-up, Aloha Health, launched this past July and she is making considerable traction. I caught up with Ms. Bishop in New York at the MedStartr Momentum, an Equity Crowd Challenge, where she won the People’s Choice award.
Here is an edited transcript of our conversation.
How do you see non-clinical insights adding value to value-based care metrics?
Bishop: As our industry moves from volume to value and from fee for service to more programs like comprehensive joint replacement and bundled payment methodologies such as ACOs models. These types of shared savings programs involve shared risk. When you have a capitated payment structure where you are being asked to manage the care of an entire population, including people with a set number of funds. Obviously, you have to find ways to be very effective in that care delivery. You have to understand all the things about that population, and each patient as an individual to help him or her to help themselves become healthier. In turn, this saves money for organizations through improved health outcomes.
What types of data would be considered as the “other 95%”?
Bishop: The first 5% of the health data happens in the clinical setting. The remaining data is what we refer to as the “other 95%” and is what happens at home, at work, and in the environment. It is also about one’s mental health, state of mind and one’s situational context. All this information is about who you are contextually. How you live can help you as an individual with the inherent benefits of giving the provider and payer an understanding of who you are. From a value-based care perspective, we have to become more efficient and more effective. As a healthcare industry, it’s imperative that we understand all the barriers and challenges that patients have in order to help them manage their own health. What is Aloha Health doing? Aloha Health is shedding light on all of the other the data points.
Would you consider your approach similar to databases like Claritas, Prizm, and Nielsen lifestyle data to better understand consumer behavior? What types of data do you intend to use?
Bishop: It is similar, though I can’t disclose the data sources. I can share that we have a baseline dataset comprised of two commercially available data source are similar to what you mentioned and one open data set. We are going to leverage as much open data that is currently available. The NOA puts out climatology data, so you can deduce the effects of weather patterns on your asthmatic patients. Another benefit is being able to take the weather patterns in conjunction with the neighborhood safety patterns. Traffic patterns and the availability of public transport can predict not just readmissions but, for someone who is diabetic and asthmatic, you might want to predict how or when they are going to be able to get exercise outside. Community data like healthcare is publicly available at the hyper-local level. Aloha Health’s three core datasets consist of one of that is purely commercial, one which layers the commercial data within health data and the third is environmental.
How will your database align with population health?
Bishop: We intend to develop what the clinical relevant social determinants are, and what is clinically relevant to each of the disease states. We are starting to look at diseases that are aligned very well with population health and care management programs such as diabetes and cancer. When we talk about diabetes, we look holistically at all of the factors that would be clinically relevant to a diabetes patient. Our conjecture is that it is not enough to know whether a diabetic is insulin dependent or non-dependent or whether the patient is prediabetes or has diabetes. At Aloha Health we want to know which social determinate factors are clinically relevant to managing the patient in his or her entirety.
How will the “clinically relevant social” insights be accessed by the provider?
Bishop: All information will be available via web service. Our social determinant database will be integrated on to the EHR plus a care management platform. The provider will be able to purchase insights on all his patients as a data overlay, and the information will be integrated into the EHR as an added custom workflow. It is a service where you, as the patient, go to the doctor, and your doctor will be able to pull up your chart along with custom patient and community level insights for every encounter.
“The need to see patients as people is very real. It is an ideal that will drive healthcare transformation.” Mandi Bishop (@MandiBPro)
Mandi Bishop prefers to be called the Chief Evangelist rather than the Chief Executive Officer. I caught up with Ms. Bishop in New York at the MedStartr Momentum, an Equity Crowd Challenge, where she won the People’s Choice award.
Here is an edited transcript of our conversation.
How does Social Determinants Of Healthcare (SDOH) data relate to me as a patient?
Bishop: SDOH attributes are available both the individual patient level and a “high propensity that this is you” level via micro-segmentation. Optimally, there will be personalization of information where personalization is possible and micro-segmentation profiles for when it is not. Also, we are not trying to give the doctor more data since we think that is a big part of the problem. “What about your lifestyle” matters which respect to you as a patient, and we at Aloha Health convert that data into insights. When the doctor pulls up an encounter, based on our models, the EHR is populated with the insights that are available about you and your conditions.
As a workflow example, I pull up your encounter. Aloha then pings the Aloha insights section and gets all this information about you. This is the use case we are going after: a diabetic patient and this is the demographic information we are going after about that person. Pertinent and clinically relevant information would be pulled up about you and on your profile. We are only showing things that matter. The fact that you are a 40-year woman is information the doctor already knows. But the fact that you are a single mother, who just got divorced 3 weeks ago, is caring for an elderly parent, and has all of these other “things”, all of these “things” would influence your ability to have an insulin pump.
What makes SDOH data a must have for patient engagement and patient-centered care?
Bishop: We can really take a look at data at the individual level as well as the population level. Everything that we are building is probabilistic in nature and will include stochastic models. I will use myself as an example to better understand our initiative. I have gained 20 lbs. in a year and my blood pressure had gone up 10 points. If I went to see my doctor right now without knowing anything except that my BMI has gotten substantially high worse. Perhaps my blood work is going to be poorer than it was last time since I was on a paleo diet a year ago. I might appear to be on the verge of becoming hypertensive or pre-diabetic if you just took those things on the “outside”. But if there was an understanding that I just got married, I traveled 300 out of 365 days last year, and I am running a new a startup. Then all of these mitigating factors can create an opportunity for context.
On average, a doctor has 7 minutes with you. If your doctor has only 7 minutes with you to find out all of the mitigating circumstances about your life that might influence your care plan and you engage in your health. If doctors could develop personal relationships with their patients it would be great, but the mass majority don’t. If my case load is 300 diabetic patients, the likelihood that I have enough time to get to know them as individuals and understand each patient’s life in context is minimal. Think about the cancer patients and something as simple as transportation. If I had to go for infusion treatments 3 times a week and I can’t get there, how would my doctor know this? Aloha Health would make it possible.
Are you going to take SDOH data and build out models for common use cases?
Bishop: Yes. This is part of what we are doing right now. We have clinicians as part of our founding team, and they work with our clients as an extension of our primary research. We are also working with their clinicians to further develop and validate the clinical use cases. Also, we are validating the relevance of all these social determinants on datasets by disease state.
An opportunity space for us, once the data are integrated into the EHR, would be to address the most likely conditions that may influence the typical intervention or would influence the care pathway. What do you do for a homeless diabetic that doesn’t have a refrigerator? What is routinely prescribed for a diabetic patient may not be the answer for a different set of circumstances. Our goal is to be able to surface all those deviations in pathways according to the clinical relevance of social determinants. Aloha Health would provide a recommendation based on a validated clinical decision support algorithms, which we are building as part of this effort. We are social. As founders, we all really believe in patients as partners. We believe in collaborative care delivery, and we are really interested in engagement. We know this data matters.
Learn more about Mandi Bishop and her team at Aloha Health.
“The prospect of integrating disparate sources of information into a multifaceted canvas of patient experiences is a tantalizing one, yet basic concerns with the usability of electronic health records, the availability of health information exchange, and chronic lack of time, knowhow, and funding have all contributed to keep big data on the back bench.” Source: Top 4 Emerging Tech Trends in Healthcare Big Data Analytics by Jennifer Bresnick, Health IT Analytics Journal June 27, 2016
Predictive Analytics World moderator, Jeff Deal recited Ms. Bresnick’s statement to the panelists and asked “Is that statement fair? Is analytics still on the backburner in healthcare? How are we going to get adoption, get beyond ad hoc activities and weave it into the culture of our organizations?” Guest expert panelists: Martin Kohn, MD, Chief Medical Scientist at Sentrian, Wasim Malik, Ph.D. from the Harvard/MIT Laboratory for Neuromotor Signal Processing, and Nephi Walton, MD, from the Washington University School of Medicine and co-founder of Brain Spin, all shared their personal views, experiences and challenges with an audience of data patriots. Here are some of the session’s takes-aways for data scientists, statisticians or persons interested in getting a green light for setting up analytics in the everyday practice of healthcare.
Seeing Top Value: The Clinician vs. The Administrator
“My viewpoint is that analytics is not on the back burner and is not on the front burner either. There are a lot of important issues that need to be understood. And if we, as data science people, can understand them and work with clinicians and hospital administration then there are not a lot of road blocks,” expressed Dr. Malik. From his perspective, healthcare’s clients are in two distinct camps. One group is the hospital administrators and CFOs who are primarily concerned with dollar savings and economics. Their focus is usually not on innovation but rather on cost savings. “When show dollar value improvement using predictive analytics to an individual hospital usually people will listen to you,” he told the audience. Clinicians, the other group, look to see what is demonstrated from analytics. Making a change or improvement in their procedures, work flow or clinical practice is key. “Data should show some promise of clinical value. Once you do that, it is not that generally challenging,” he said with respect to engaging clinicians.
Pinpointing the Motivation for Building Models
Dr. Martin Kohn believes that we haven’t done a good job in addressing real world issues. Also, healthcare is an inherently changing system because there are so many mixed motivations. He reflected back to the days when he was an emergency physician. The goal was reducing emergency room visits. “In a fact,” he stated, “ER medicine opposes a lot of these initiatives where you have to get permission from your doctor to go to an emergency room. So when you are doing things to threaten some else’s self-interest, there is going to be a lot of push back”. The challenge, as he sees it, is to demonstrate that which produces value for the end user. It is one of the reasons why his company, Sentrian, does not work with individual physicians who are in fee for service models since they have less interest in reducing encounters, but ACO-like organizations do. Dr. Kohn highly recommended that you need to direct what you are developing in order to show real value for the target audience.
Barriers for Data and Implementation
Barriers for analytics projects start with trying to get all the data. In Dr. Malik’s experience critical things like clinical data, particularly in the form of EHR data, is noisy. “Typically when we work with EHR data, people have to sit down and manually curate and clean up the data. We spend 80% of our time doing janitorial work. It is not fun but has to be done to make sure. Otherwise, it is garbage in and garbage out,” he emphasized.
According to Dr. Walton, Washington University has a lot of data. Securing resources to build models can be worrisome. But perhaps the greatest challenge, once models are built, is implementing them into the infrastructure. “The universal answer – No. You have to fight through layers of bureaucracy to actually get something into the system,” volunteered Dr. Walton. He mentioned that another handicap is a thinly staffed IT group, and this situation limits the available resources for model implementation. Since it is a cost center, the people who run may not see the benefit of what you are doing in order to make a priority. “You have to translate what you are doing again into dollars or something that the other stake holders or other people understand in order to implement it. That is a major challenge,” he stated.
Establishing Trust in the System: Starts with Collaboration
When Dr. Walton first proposed changes to the standard genetic testing algorithms, he got a lot of resistance from physicians. Having given the situation some thought, he realized they were on the peripheral of the project and not at the center making self-invested decisions. They had never even seen the data. From his own experience, he imparted “If you show them a better way to do something, a way that they can believe in then make them apart of the solution. I think you can gain some acceptance. Tell them this is the data and ask how you [the clinician] want to fix it. You want them to be a part of the conversation. Making sure their concerns are addressed brings them into the conversation. That can improve your models.”
The Predictive Analytics World conference took place at the Jacob K. Javits Center in New York City on October 23-27, 2016. Find out more about the Predictive Analytics World.
The internet has a rather detailed picture of the health of the population, coming from digital sources through all of our connected devices, including smartphones. This is digital epidemiology: the idea that the health of a population can be assessed through digital traces, in real time. Digital Epidemiology: Tracking Diseases in the Mobile Age. M. Salathé, J. Brownstein et al.
As a Harvard Medical School Professor and the Boston Children’s Hospital Chief Innovation Officer, the plights of patients and the hurdles in care are Dr. John Brownstein’s starting points for questions and discovery. When the Community Transportation Association study reported “an estimated 3.6 million patients the United States miss at least one appointment due to lack of access to transportation,” Brownstein was determined to make this challenge his own. This fall, he launched the first customizable patient-centric digital transportation system – Circulation – a new vision for non-emergency medical transportation. As a Klick Health Muse attendee and having had the privilege to speak with John Brownstein, Ph.D., co-founder of Circulation, I would like share what I learned about his journey as an epidemiologist, public health educator, and innovator.
Social Media’s Big Data: Preventing Epidemics and Tracking Drug Safety
Digital Epidemiologists think in terms of “digital phenotype” to understanding the health of individuals. Uncovering critical information about what is happening at the population level is collectively called “digital exhaust”. These digital traces that are left behind, help track local outbreaks around the world. “In fact, you don’t need surveys, just mine what people are saying online. We combine social media to get real insights as to what is happening on the ground: facts and sentiment. The ability to understand risk and population health is fantastic with these emerging technologies,” opened Dr. John Brownstein at the 2016 New York City Klick Muse event.
Social media mixed with disparate sources of health data, was how Brownstein began solving public health risks. In 2006, he designed HealthMap, a publicly available, online real-time disease surveillance system, with the capacity to monitor emerging public health threats such as Dengue fever, Zika Virus, Influenza and Ebola. “Data Mining was actually the first indication of hemorrhagic fever in Kenya and how it spread across West Africa. And even more recently using emerging technologies, big data, data mining from social networks and online chat groups, we have been able to show the expansion of Zika Virus as it emerged across South America and Central America,” shared Brownstein.
Getting insights into issues at the population level that doesn’t come through any traditional channels makes social media a priority and a necessity. As a way to troll for drug adverse events, Brownstein created real-time a pharmacovigilance app that mines twitter and other social networks. The FDA now uses Medwatcher to monitor safety signals from drugs. He also developed StreetRX, an app that allows for anonymously reporting and a crowdsourcing approach to capturing black market prices for prescription drugs. The information aids in detecting which prescription drugs are being abused. “What are we going to do once we know about this risk? What will be actionable to take from the risks that we know in populations?” are Brownstein’s theme questions.
Uber Health takes on Public Health: On-Demand Vaccine Delivery
Two years ago, Brownstein had been running the Vaccine Finder to organize vaccine supply for Walmart, CVS and Walgreen’s. With over 70,000 locations over the country, 60% of adults were still not getting their annual influenza flu shots, which is a lifesaving vaccine. The percentage of non-vaccine users were also far worse in younger age groups. “You know that patients don’t have a lack of faith in the vaccine. It is a lack of awareness and education. Even though the vaccine is a block away, people aren’t willing to walk a block to get it. This is when we came up with the idea of Uber Health. Why not get vaccines to the people just like people getting deliveries of anything on demand? Why can’t that be transformative in healthcare in as well?” shared Dr. Brownstein.
The story began one day in 2014 with the launch of a 4-hour on-demand campaign to prevent the flu. How? The answer was Uber. They put a few little tabs on the Uber app saying Uber Health. In four cities, New York, Boston, Chicago and Washington DC, Uber customers had a $10 flu delivered with a click. “It was extremely successful. We didn’t have enough supply of nurses to meet the demand. People were taking selfies while getting vaccines. Going viral, we got really excited selfies coming back at us of people getting vaccines. This was really a transformative moment for us because we then realized that you are actually having an impact on a public health level with on-demand service.” recalled Brownstein.
A post-treatment survey revealed 42% of the vaccine recipients did not have a one the prior year, and over 90% believed that an on-demand access was important in their decision. Once again there is a bigger concept. In fact, this past year across 40 cities, Uber Health showed the same level of enthusiasm within this population. On-demand healthcare is now being pursued by a majority of companies in telemedicine with growing number offering house calls and drug delivery.
Circulation: A Cost-Saving and Patient-Centric Transportation System
On-demand healthcare is the key to patient access and engagement. “Our question is why Uber can’t be for healthcare?” asked Brownstein. The Circulation co-founder had been grappling with the flaws and risks of the Non-Emergency Medical Transportation (NEMT) system for quite some time. Under the status quo, sick, disabled low-income patients encountered many frustrations such as: limited ride options, 48 hour bookings prior to pick up, waiting over 3 hours for ride to finally arrive and third-party services lacking new mobility software. The consequence – missed appointments that threaten a patient’s well-being, which can lead to terrible outcomes on the part of the patient. “So this is really a concern not just to individual but the healthcare system overall which spends billions of dollars currently on transportation services. Over S5 billion [Medicare & Medicaid $3 billion and Health plans $2 billion] is spent on moving patients around, getting patients to appointments, getting them out of hospitals and getting them to clinical trials. So now we have a huge amount of cost and low patient satisfaction. There has to be a better way to do this. So we have a new vision for transportation- Circulation. Taking modern approaches to transportation and bringing them into healthcare,” announced Brownstein.
The official partnership of Circulation and Uber launched in late this September. The digitized patient-centric transportation platform was piloted at Boston Children’s Hospital, Mercy Health System in Pennsylvania and Nemours Children’s Health System in Wilmington, Delaware. Succeeding in integrating the current work flow of operations, the system demonstrated interoperable interfaces with patients’ electronic health records, which also met HIPAA requirements for compliance. On-demand scheduling in real-time ensured patient arrival information was received by physicians, nurses, and caregivers; and rides were billed and payment reconciliation fulfilled on the backend. The intuitive cloud-based software platform brings on-demand convenience and efficiency to the world of NEMT so that the growing number of pediatric patients and geriatric patients will be able to show up for doctor appointments. On-demand transportation – Circulation – another win-win for digital epidemiology.
More information on Circulation.
Having been successful for decades maintaining a traditional product-driven model, pharmaceutical companies are feeling their way toward a new business model: patient-centricity. Hospitals, providers, payers and other health organizations have transitioned towards outcomes rather than products and services for several years now. Pharma’s comparatively recent patient-focused approach has cultivated a strategic shift from brand to disease. This cultural change has spurred on partnerships and collaboration testing new models for patient services, payer engagement, and pharmacy engagement. A suggested read on this subject is Heidrick & Struggles’s whitepaper ‘Walking the Talk’ in Patient-Centric Pharma.
At the World Congress on EHR and E-Prescribing Summit, Dr. Usman Iqbal discussed how EHRs have created a tremendous value for the pharma industry as they work towards patient-centricity. As the Senior Medical Affairs Leader at AstraZeneca, he shared his own experience and spoke to the audience on how the concept of patient centricity gets translated into meaningful practice. For the pharma ecosystem to succeed, three key areas must be addressed:
- An entity needs to be in place that synthesizes the EHR database of patient information and translates data to support actual decision making.
- Identification of barriers to care in order to proactively spot opportunities and build solutions around a target product profile.
- Most important is the strategy. The plan is to then apply it to chemical or commercial decision support systems or a patient access programs.
Patient Journey Detailed in EHRs
Dr. Iqbal deems these three aspects (listed above) as the fundamentals to truly deliver patient centricity to the pharma ecosystem. Intricate to this process is access to patient EHR data, and at this juncture, the concept of the patient journey comes into play. Historically, pharma has opted for high-level patient assessments, which are extracted from primary research, market research, and focus groups. In today’s environment, a plethora of tangible and detailed information actually enables a level of decision confidence to define a commercial strategy or define a development strategy. Over the last few years working on this concept of patient journey, pharma has tried to pull together 5 key stages of the patient journey. The five key stages include:
- Understanding the treatment landscape
- Disease management
- Experience in health outcomes
Patient screening and diagnosis require knowing the barriers to access that impact a patient’s ability to get screened. Fortunately, Big Data and EHRs are a significant help in capturing this information. Dr. Iqbal contends that knowing the actual epidemiology of a disease is useful for making a diagnosis and fully understanding all the patient touch points. “Much of the time the patient journey starts with treatment but 50% of the time 50% of the population doesn’t get diagnosed or may be missed diagnosed. You are losing a lot of your focus by starting with treatment. You need to start with awareness, screening and assessing the diagnosis landscape. And that takes a lot of variable details and that is where EHRs are very helpful,” expressed Dr. Iqbal.
EHRs are providing pharma with the whole 30,000-foot view of the treatment landscape and disease management, which is a very important piece to understand. The information helps to address questions such as: What is the standard of care in today’s world? What are the unmet needs? How can we position the unmet needs towards a target product profile that correctly addresses those needs? Dr. Iqbal asserted, “All of this information along with a patient segmentation, and most importantly what the prescribing practices are, can be gleaned systematically from EHRs and positioned towards clinical, commercial and medical care strategy.”
Patient Centricity Needs Real World Outcomes
If you are truly patient-centric, you need to know what the outcomes are as well as their differences in terms of clinical trials versus real world outcomes. When you are talking about clinical trials the measurements are marked by clinical endpoints. However, Real World outcomes cover economic endpoints, patient medication adherence, the patient-reported outcomes and health-related quality of life. EHRs detail the full picture at the time the patient gets access to the drug. “You can have an approved drug but in the end, someone has to pay for it. And those payers are putting in all those prior options in the background. Behind those prior options, you need to have to effectiveness data. How does your drug compare to the competition? Unless you have this information on the treatment landscape and disease management, it is really hard to keep up against the competition. If you have this information all along, it won’t empower the patient. But within the institution, it will help you establish better patient-centric designs that have the right information, the right population, the right endpoints, and the right competitors. So you won’t get rejected when you try to get access,” pointed out Dr. Iqbal.
Fit for Purpose Databases: Links Patient Centricity
EHRs only provide information on episodic care. Accessing several different types of databases to track patients across all sorts of care, opens channels to more knowledge and information on patient experience and provides a complete picture of the patient journey. Linking unique patient identifiers is where all the advancement is happening in Big Data. In the pharma world, companies are joining with cutting edge data providers to create, customized “fit for purpose” databases. The goal is to have the ability to link all of these disparate datasets and make sense of them in order to track the actual patient experiences. “What Pharma is investing in, is actually the partnering with data providers to create fit for purpose databases, which is geared for one disease or even one indication. It requires a lot of investment but it provides you with a complete view of the patient. From diagnosis to treatment, outcomes to genetics, depending on whatever data you need and you can build it, “described Dr. Iqbal. An excellent example of a fit for purpose database is the Oncology Services Comprehensive Electronic Records database, also referred to as OSCER. Learn more about the OSCER database. More about the work and background of Dr. Usman Iqbal.
A Spotlight on Intervention and Analytics with Damian Mingle, Chief Data Scientist, WPC Healthcare
On the eve of the Sepsis Alliance’s gala, attendees from national healthcare organizations walked the red carpet, posed for celebrity-style photos and partook in festivities at a trendy Times Square club – all to honor sepsis heroes, support survivors and remember those who had succumbed. WPC Healthcare’s Chief Data Scientist Damian Mingle received recognition as a finalist for the 2016 Sepsis Alliance Heroes Awards for his work applying data science to detect early warning signs. WPC Healthcare’s soon to be released Sepsis Risk Index (an early warning system), built from a hospital facilities patient data and more than 20,000 variables extracted from a vast external data resource, has the potential to be the model of choice among all the current scoring models.
Annually, there are over a million cases of sepsis with a mortality rate of 28% to 50% in the United States. In recent conversations with friends, family and associates, a six degrees of separation phenomenon surfaced. So many of them knew a person who encountered sepsis, who either died or recovered. The fact remains that sepsis mortality can be prevented. The Sepsis Alliance is making this public health crisis heard and acted upon. September is Sepsis Awareness month. Visit the Sepsis Alliance organization, which brings awareness, knowledge, intervention and prevention to all health providers and organizations as well as the public. Read the stories of the survivors and families who lost a loved one and how they are devoting their lives to prevent new tragedies.
Damian Mingle and with him, Ruth Smith, Director Business Development, arrived in from Tennessee, and luckily we had slotted just enough time for a short interview. Here is a transcript of our conversation edited for clarity.
What was the initiative for taking on Sepsis Prevention project? Answer: The need to know more.
Mingle: We were consulting on a project with a prominent healthcare organization [ identity soon to be released in upcoming publications] that had a Hospital Engagement Network program, and we were encountering a lot of information around harms events. The president of a major division in a hospital system, who wanted to know more about the sepsis population that he had, called us into a meeting. He felt that a simple report wouldn’t be able to answer his questions. This hospital is actually doing really well with sepsis mortality, and are at almost half of national average.
Was there a reason for the hospital’s interest in pursuing an analytical solution for Sepsis? Answer: Apply a new data science approach.
Mingle: They have already implemented a lot of statistical process control. They have wonderful protocols in place and did a lot of training and education. Yet, they still wanted to move the needle. We took that as a real compliment. They asked us how we could put a data science approach to this. We had already been thinking along these lines of building an infrastructure or architecture that allowed us to easily ingest data at the right time to make these kind of probabilistic statements about a patient. After five months working on this project, we came out of the data lab and shared with the client a little bit of what we had. It was a compelling presentation for them to say let’s go ahead and deploy this at a partner facility.
With respect to the deployment, what kind of advanced, interfacing technology is required in terms of electronic health records and interoperability? Answer: An HL7 feed.
Mingle: When we developed the solution we asked — what was the lowest amount of interaction we could require from a facility? The minimum for us is just an HL7 feed from the data that is already in place. Every hospital in American has it deployed. We ask them to “fork the data” which means instead of sending the data in one direction we can push or pull that data. Our company provides all the processing, all the normalizing and all the storing [not just the algorithm, but cloud-based back end]. Some hospitals are more sophisticated than others, but we want it to be universally approachable for them. Honestly, they turn on the data and we deliver the answer (or the score) back. The technology is text or an email, all encrypted, and they can make a judgement call quickly based on empirical observation.
How does your model compare to the other models that are in place? How would explain the model effectiveness in terms of the improvement score?
Mingle: Quite honestly, the major difference between our solution and other solutions is that we are not incorporating any vital signs or any clinical information. We are not using clinical values so there is no white blood count, respiratory count and heart rate. That is a big difference. The other thing that is kind of interesting, is when we go in to a client meeting, we usually have to educate a little bit on data science in general and what accuracy means. A lot of people ask about accuracy that doesn’t take in to account the imbalance you would see in a positive and not positive population for sepsis (the presence and lack of presence of sepsis). Our model actually identifies patients who do and do not have sepsis. When we talk about area under the curve (the correct identification of sepsis and those without sepsis), which is what really are excited about, it takes an imbalance and right sizes it in a way that we can talk intelligently about prediction. Most of us like the MEWS or TREWScores out of Johns Hopkins that accounts for 70% AUC [area under the curve]. We just did presentation in Nashville yesterday of our new system, where we are at 94% AUC, which is pretty high. As you might imagine, we get a lot of questions about how we can do this with a model that does not include vital signs or any other clinical data.
Where do the models have the most impact on treating sepsis? Answer: Definitely teaching hospitals.
Where we see the most impact at teaching hospitals. Most hospitals, from a sepsis care [event can cost] about $35,000 on up. For a teaching hospital, it’s about $168,000 on a per case basis because they are running so much lab work trying to figure out what they need to do. It is a good training set for a teaching hospitals [to use a data science model]. In fact, we did some post analyses, where we looked at a series of [physician] attendings. Those who encountered sepsis more often had a lower mortality rate in their patient population versus those who hadn’t encountered sepsis. I think it is because they just hadn’t encountered it a lot. Letting someone develop their medical intuition around sepsis while they are protecting the population seems like a really a big win.
Here in New York we instituted Rory’s law for prescreen. Illinois has also done the same. What are your thoughts? Answer: To have more non-invasive, more intelligent screening.
I am a fan of screening that is not invasive and in a way that’s a little more intelligent. It is hard to imagine, we have this dataset that has been around for forty years. We are using it in a way of an-old –problems, new angles approach. It is a low cost way to not have any selection bias. We run it on everybody that is coming through the emergency department or direct transfer, and process the information back to the clinician. WPC Healthcare reprioritizes the information for the health provider to determine if the patient needs to be seen by a doctor. The result is 3 to 4 hours of time saved. In some cases you are dealing with late stage sepsis, shock or severe sepsis. From a mortality perspective, these people are already half dead. Even if you have a world class protocol in place with a world class doctor at the top of his or her class, chances are not in their favor so —time — is the only thing that really matters.
Tom Ahrens,PhD,RN, one of the recipients of a 2016 Sepsis Alliance Heroes award, has an educational approach to managing sepsis. Do you foresee the blending of forces using both education and analytics? Answer: Both components, education and analytics, are needed. Time is of the essence for sepsis survival.
Smith: Absolutely. There is a clinical component to this that we don’t replace. We’re just the indicator on the front end. In the hospitals that we are working with, they are using their normal protocols to take out information and work it into their clinical work flow. So what we are really doing is gaining hours from the front end of identifying sepsis that helps then treat the clinical side of it effectively. Both sides of the component are needed and gets us additional lift in terms of saving lives, it is the time component in helping sepsis.
Congratulations Damian Mingle on his breakthrough project. Many thanks to Sepsis Alliance board members and special tribute to all 2016 Sepsis Hero winners. And we can’t forget to thank to all nurses on frontline defense. Our Sepsis Stars whole-heartedly deserved their night on Broadway!!
Interested in discovering more about Chief Data Scientist Damian Mingle, WPC Healthcare and their analytical solutions please visit their website.
- Sepsis is the body’s overwhelming and life-threatening response to infection which can lead to tissue damage, organ failure, and death. Sepsis is a medical emergency that requires early detection and treatment for survival.
- Sepsis can be treated but it must be suspected first.
- We don’t know yet exactly why sepsis occurs. Important to limit your exposure to infections by hand washing, caring and cleaning wounds and getting vaccinations.
- If you are worried about sepsis, call 9-1-1. Studies suggest early care in an ambulance can increase survival. Tell health care providers, “I am concerned about sepsis.
Drugs and Devices: What Matters Most is the Patient
“I got an email from someone who was neither a doctor, nor a payer, nor an insurance broker, but just someone who had read our piece on the catheter breakage. She wrote that she had been a patient for many years and absolutely dependent on indwelling medical devices. She read our piece on how we are able to identify problems early, and she realized how incredible it was – we were truly promoting patient safety and welfare. I sent this woman’s email to the entire team, with a note saying ‘this is why we get out of bed in the morning’,” shared Libbe Englander, PhD, Founder, and CEO of Pharm3r.
Dr. Englander formed her company just five years ago. Pharm3r (pronounced “Farmer”) (@Pharm3r) actually cultivates very large datasets by accessing multiple sources of information on end-user experience of medical products, allowing people to track adverse events and compare product effectiveness. A software company at its core, proprietary programs use natural-language processing and artificial intelligence to mine, extract and combine data. This allows companies to find trends in adverse events, and identify product problems and patient sentiment. Having had the pleasure to meet with Dr. Englander in her New York office, here is a transcript of conversation, edited for clarity.
Where do you think Pharm3r has made the most impact on public health?
Dr. Englander: It’s hard to narrow down–we’ve been having a busy and brisk couple of years! We just put out a case study on catheter tip breakage, Pharm3r Case Study: Predicting Medical Device Recalls. It’s one of many, but because it was just published, I think it makes for interesting reading. Cook, voluntarily recalled their Beacon® Tip catheters. They felt that there was a higher-than-acceptable amount of tip breakage.
We were curious about how the catheters still on the market compared, went to our system and immediately pulled up our data. We ran it through Boomerang NLP™, our natural language processor. It turned out that Angiodynamics’ catheter had a higher rate of tip breakage. Subsequently, that product was recalled.
NB:Dr. Englander also shared that they will soon be publishing a paper on the TAVR (Transcatheter Aortic Valve Replacement) and MitraClip® clips both for valve disease treatment. Pharm3r will compare their data with highly curated, very expensive patient registries. So far Pharm3r data has been replicated in these registries, which serve as a good quality control. Importantly, they will soon be able to share the comparative effectiveness of each of the TAVR valve products, which are not publicly reportable from registries. Dr. Englander: “Many companies are making, in some cases, lifesaving instrumentation and drugs, but there are side effects with everything. The sooner they are detected, the sooner an engineering and corrective measure can be instituted. Companies are very interested in making sure they are the first to know.”
How does one go about reporting an adverse event in MAUDE and FAERS?
Dr. Englander: Anybody can report an event – a doctor, a patient, family member, a concerned citizen- and that, in a sense, is the real power of these databases. It is tremendous that there is a way that those voices can be heard. I think it is so significant that anybody can fill out a MAUDE (Manufacturer and User Facility Device Experience) report or an FAERS (FDA’s Adverse Event Reporting System) report. They are so easy to do – they can be done electronically, by hand. Our job as a company is to translate that voice into meaningful information. We feel that theses databases capture the end user voice in a very democratic, wisdom-of -the-people fashion.
How has your technology and products improved adverse event reporting?
Dr. Englander: One of the things that we can do is start looking with natural language processing at the MAUDE narrative, which is a very rich source of information. You not only have a huge amount of structured data on the product, which is submitted by a patient or medical professional, but also often a very large ad important narrative text that is associated with each one of these reports. We have a patent-pending natural language processor, which allows us to read effectively those 5 to 6 million MAUDE reports. We are able to look at product problems and adverse events in a precise and sophisticated fashion. If you are a patient and about to get a valve replacement, using our product, you can now know which valves are associated with adverse events. Each report is accessible to the public, but to put them together and do a comparative analysis, you need our system. Our best-use case is to inform patients, doctors, payers, governments where there are trends, adverse events and comparative effectiveness that are meaningful.
What do you think are the advantages and limitations of the OpenFDA Data Initiative?
Dr. Englander: The OpenFDA initiative, requiring and encouraging people to report adverse events in the FAERS database for drugs and the MAUDE database for devices, is, I think, an extremely credible and well-intentioned project. The Open FDA is a great initiative is to start playing with. The idea of saying ‘let us try to put an understanding of the end user experience in everyone’s hands’ is a well-intentioned one. However, the current iteration of the OpenFDA, is also deeply flawed, as it does not reliably reflect the underlying dataset for a couple of reasons. There is a lot of miss-matching of names, carelessness in terms of how things are dated, and a lack of suppleness in extracting the exact adverse events. We wrote a white paper, OpenFDA, raw FAERS and Pharm3r’s PandoraPlus™ that deals with this in depth. On a few time series test cases, the time series were really not representative of the underlying data. As a sophisticated computer science company, we take the raw data sets and look at the underlying data in a much more nuanced, careful way.
Has Pharm3r been engaging with the FDA on Post Market Surveillance?
Dr. Englander: Yes, and I have been very impressed with the people I talked to [at the FDA] so far. Their challenge is that they don’t want false positives. They don’t want to yank things off at any bad signal. I feel they are very much trying to do the right thing. They are exploring the best way of doing post-market surveillance. It is an important problem. The better the technology you have, the faster and the more accurately you can see the problems. We are giving a webinar on the September 20th under the auspices of the FDA to discuss our technology on this. I think it is a public health issue. It is a comparative effectiveness payer issue. What we do is really critical. A lot of our work is for product liability insurance companies. They are insuring the companies making these products, so it about how you price that risk and how you take that risk. Ideally, the insurers and brokers are in a position to encourage these companies to be even more proactive about avoiding product problems. I really think we are at a moment in time.
Pharm3r is a computer science company that is linked to sector expertise looking at medical products. They help their clients think about effectiveness, risk, pricing models, risk models and ultimately the effect on the patient and the end user. The company has been successful in identified major datasets and has done a highly competitive job about collecting, and correlating and analyzing them. They do everything from building predictive models to looking at trends and comparative product effectiveness. To learn more about Dr. Libbe Englander, Pharm3r, and their products click here.
Hallmarked as a solution to improve healthcare quality, cost and safety, studies are showing health technology is up against a “digital divide” when it comes to patient engagement. At the Internet Governance Forum, Pew Research Center’s Lee Rainie, Director of Internet, Science and Technology Research presented the Fact Tank Report discussing the “digital divide” that exists in 2016. The report documents that lower income, less educated, non-white, seniors and rural communities are the least likely to have home internet, home broadband, mobile connectors and smartphones. This summers’ medical publications, the Journal of the American Medical Association and the Journal of the American Board of Family Medicine, released studies where demographic and socioeconomic data marked the root causes to limited or no access to digital technology, thus hindering the benefits and improved outcomes it can bring to the neediest and most costly populations. Here are the highlights from each study.
Trends in Seniors’ Use of Digital Health Technology in the United States, 2011-2014, a research letter submitted from Harvard Medical School’s Brigham and Women’s Hospital, appeared in the August 2, 2016, JAMA. Authors, David M. Levine, MD, MA, Stuart Lipsitz, ScD, and Jeffrey A. Linder, MD, MPH,FACP made mention that this study, based on the National Health and Aging Trends survey (NHATS), was exempted from the Partners HealthCare Human Research/IRB Committee. The research team included participates to the longitudinal NHATS survey in 2011. The participants were re-surveyed annually on everyday (nonhealth) and digital health use until 2014. The research team acknowledged that this may be the first nationally representative study to examine trends in the adoption of digital health technology by seniors age 65 years and older who are community-dwelling Medicare beneficiaries.
Here are some the reported statistics from the study:
- In 2011, the total number of participants was 7609. The mean age was 75 years (SD, 7.4); women were the majority with 57%. By 2014, the number of participants decreased to 4355 seniors (1430 due to deaths and 1824 lost to follow-up).
- In 2011, senior cell phone usage was 76% and computer usage was 64%. 2012, 2013 and 2014 saw little or no significant change. In comparison, the general population had approximately 90% internet use and owned cell phones.
- In 2011, the seniors reported everyday technology such as internet usage at 43% and email/texting at 40%. Less than 20% of the seniors used internet banking, internet shopping, social network sites (from 2013 data) and tablets (from 2013 data). Yet, by 2014, marginal but statistically significant everyday technology usage started to increase.
- In 2011, the rates digital health usage was low with 16% for health information, 8% for prescriptions, 7% for clinician interactions and 5% for managing insurance online. By 2014, the proportion of seniors who use digital health technology for health information, contacting a physician, or filling prescriptions increased 4% from 21% to 25%.
- 2011 socioeconomic and demographic data categorized by digital health modalities had shown due cause for a “digital divide”. Participants who were older in age, Latino and “other” race/ethnicity, divorced and poor health had lower use rates of digital health technology. And variables associated with greater use included college education, higher annual income, taking medications, and more comorbidities.
“Digital health is not reaching most seniors and is associated with socioeconomic disparities raising concern about its ability to improve quality, cost, and safety of their healthcare. Future innovations should focus on usability, adherence, and scalability to improve the reach and effectiveness of digital health for seniors,” wrote the authors in their closing discussion.
Original research on Patient Portal Use and Blood Pressure Control in Newly Diagnosed Hypertension from the Department of Family and Community Medicine at the Saint Louis School of Medicine was published in the July-August 2016 edition of the JABFM. The research team, William Manard, MD, Jeffrey F. Scherrer, Ph.D., Joanne Salas, MPH and F. David Schneider, MD, conducted a study to determine whether a patient portal use was associated with controlling blood pressure in hypertensive patients. Inadvertently, a covariate adjustment to their outcomes model presented the socioeconomic factors as being the influencers on controlling blood pressure (BP), not patient portal use
Here are some of the reported statistics from the study:
- A study sample of 1,571 patients, ages 21 to 89, with an incident hypertension diagnosis between 2008 and 2010, was identified from an academic medical center primary care patient data registry.
- Portal use and the incident BP control were tracked for all patients from 2011 to 2015.
- Cox proportional hazard models were used to estimate the association of portal use and BP control.
- Sociodemographic variables included race, sex, age, marital status, socioeconomic status index, and clinic type, volume of healthcare utilization, smoking, depression, obesity and comorbidity index.
- The results of the first Cox proportional hazards model, which adjusted for age only, showed that portal users were more likely than nonusers to achieve BP control.
- Interestingly, when a second model was built adjusting for all the sociodemographic variables, portal use was no longer associated with controlling blood pressure.
- A “digital divide” was evidenced given that race showed only 28% of non-whites used the portal; also in socioeconomic status with the lowest tier having 18% portal usage and 19% in the lower-middle segment.
On an analytic note, transparency reporting has become exceedingly important in research. Kudos to Dr. Manard and team for reporting both the unadjusted models and the covariate adjustment. Their analysis uncovered that “a health disparity exists in the use of patient portals and its benefits for BP control”. The authors recommended further “research to determine which sociodemographic groups would benefit most from access to patient portals and what conditions and what outcomes are most sensitive to improvement via portal use”.
- PewResearchCenter Internet, Science & Tech Digital Divides 2016 by Lee Rainie
- Trends in Seniors’ Use of Digital Health Technology in the United States, 2011-2014, August 2, 2016 JAMA 2016,316(5) 538-540 doi 10. 1001/jama 2016.9124
- Patient Portal Use and Blood Pressure Control in Newly Diagnosed Hypertension, July-August 2016 J Am Board Fam Med 2016;29:452-459
“Where ever there are bundled payments today, physicians are typically required to report out on patient-reported outcomes,” says Dr. Todd Johnson, CEO of Noble.MD. The Pennsylvania-based digital health company provides a technology platform called Theo – defining a new standard as the interface for capturing patient-reported data. At the MedCity Converge Conference, Dr. Johnson shared his predictions on how the MACRA payment reforms, Meaningful Use and Bundled Payment programs would act as catalysts for incentivizing patient-generated data. At our interview, he foresees today’s new healthcare programs, such as bundled payments, and digital wellness tracking, making it a requisite for doctors to gather information directly from patients. However, the real turnkey is to give actionable information back to the patients so they can be proactive about preventive care and staying out of the hospital. “But what we found was that doctors and nurses just don’t always have much time. We’re helping healthcare providers to save time by more efficiently capturing the information they [the physicians] need to pass on [to their patients] for these programs,” claims Dr. Johnson. Interview highlights with Dr. Todd Johnson (gently edited) starts here:
How are reimbursement programs trending towards requiring patient information?
I think that what MACRA will do is provide a launching pad for increased importance for capturing this data across all specialties. With MACRA in place, a physician will have to either participant as part of a bundled payment or will have to participate in the MIPS program. This new metrics incentive program is going to consolidate PQRS and all of the other quality surveys, and is going to necessitate that doctors have a standardized, useful and actionable as well as efficient means of capturing all this data. In 2019 the [MACRA] law becomes effective, so for the next several years, capturing and benchmarking that data will become very important. Plus, ensuring that every health system that their patients are receiving care and are self-reporting the care back to the physician, also is done in manner that puts them [providers] in a good position to remain stable financially.
How have physicians been able to transition to this new mindset?
I think few physicians have time to really consider it. When they do consider it, they are very concerned about the requirements that are being put on them, and especially how they are going to meet the requirements.
Is the Theo platform a tool to help reach the MACRA and MIPS levels?
Yes. Yet, no one knows what the actual MIPS question set will be, but we are currently working with our clients to design a patient reporting quality management system. We are capturing the patient-generated information and developing a baseline to understanding where they are now. Once the CMS publishes the surveys that they’ll use, we can rapidly move to help to doctors.
What have you seen so far as the benefits of collecting patient-generated data? Have you been able to perform any studies?
We have been able to prove how the use of our system improves patient and health outcomes. Yes, we have run studies across the patient population spectrum from pediatrics all the way through elder care. Our tool has helped a number of hospital systems demonstrate a real positive impact on their patients. Children’s Hospital of Philadelphia (CHOP) is a client of ours, and were an early user of the Theo platform. The head of vaccinations at CHOP really wanted to understand why patients or the parents of the patients were opting out of HPV vaccinations and other really important vaccinations. What the CHOP team was able to find out from conducting study [via the digital Theo platform] was that a significant portion of patients were opting out because they didn’t understand what the potential ramifications were of not having their child vaccinated. Also, they didn’t understand the wealth of data behind the safety of vaccinations. By presenting that data to them in a very easy to understand manner, the CHOP teams have been able to increased HPV vaccination rates by a significant level. This study will be presented at the CDC conference this fall and is to be published in a peer review journal.
At Nobel.MD, have done the same for other healthcare organizations and in many of different populations. With a large Miami primary care practice, we were able to increase flu vaccinations rates by 37% and increase rates of cancer screenings by 10% to 20%. We are able to increase the number of patients who were interested in and willing to undergo advanced care planning. With a signed and stored document, an advanced directive in put their EMR over the period of that project. All of these are opportunities is what everyone is trying to do now to improve patient care and decrease overall system costs.
Can the platform help providers meet Meaningful Use?
The Theo platform can help physicians meet Meaningful Use. What we have seen is that with most EMRs, everyone has a patient portal, but very few of them are effective. The best example of that is MU 2015 stage 2, when CMS had to lower the bar to a single patient having access to a portal system. But the reason why is that EMRs weren’t built for patients to use them. EMRs were built for physicians to use. Physician and nurses are well trained and can easily pick up these types of complex systems. They understand the terminology used in the EMR, which are often at times very detailed oriented. They are also OK with a screen with 1,000 different numbers on it. Patient interfaces, on the other hand, need to be kept extremely simple. For elderly patients especially for whom sight may be becoming an issue, large buttons are created. They need to capture as many patients can and make it as useful as possible. Vocabulary on patient interfaces are written on a 4th grade or lower grade reading level. We need to do this to make it usable by everyone. That’s where many of the EMRs attempts at patient interfaces have stumbled.
How does Theo’s customizable platform facilitate the exchange of data from the patient to physician and back to the patient?
While in the waiting room or in the exam room, the patient is handed an iPad, usually in a private environment before the doctor arrives. Preset questions, mainly multiple choice, guide the patient through the survey. Disease status and changes in health, ability to taking care of oneself and signs of depression, the success or challenges of following the care protocol can be directed by the provider on Theo’s customizable platform. Noble.MD makes sure the platform captures correct questions for the selected patient types or protocols, along with specific document preparation directed in Theo. For example, if the patient is an elderly primary care patient, the doctor or nurse can choose to have the patient run through a data input needed for a basic annual wellness visit. If an orthopedic surgeon’s patient is about to have a hip or knee replacement. Theo can reach out to that patient through email or text message on how the patient is following the doctor’s orders. Simple checks on whether patients are taking their medications as the doctor requested leading up to their surgery, on the day before surgery, did they remember to not eat or drink anything before the procedure. Theo can keep the patient engaged leading up to a procedure, plus give information back to the patient. To goal is make sure the patient understands and remembers the importance of following directives for optimal results. Theo digitally facilitates the medical team and patient to work together.
Any comments you would like to add about the digital health innovation?
What we see as hindering the progress is a reticence to innovate because it is seen as uncertainty. There is a fear of the unknown when you are adopting a new solution. What we’re seeing since the ATA was passed in 2010, and now that we are 6 years into this, is the solutions that are surviving, growing and thriving are ones that are proven because we have the data behind it. Ask for measurements of results, the ROI, but don’t fear innovation. And don’t fear digital health. It is a very important part of the future. It is helping to save time and make life easier for many healthcare professions to better take care of their patients.