Five Ways Analytics in Digital Health Tools Will Change Healthcare


There are many reasons cited why digital technologies hope to improve of patient care as well as the state of healthcare itself. They include improving efficiencies, patient safety, and cost. However, as has been seen with the most ubiquitous face of digital health technology, the EHR, these promises remain unfulfilled. One significant barrier to the utility of digital technology has been the heretofore unlinked status of ‘sterile’ data with analytical tools which can bring it into the world of clinical relevance to both the provider and patient. Analytics have been utilized in other sectors of society including retail, social and finance for decades. They drive efficiencies and outcomes at Amazon, IBM, telecoms, FedEx, financial institutions, and sports. Yet the millions of bits of discrete data amassed every minute in healthcare are warehoused in a contextual vacuum. To add insult to injury, even when utilized in hospital patient satisfaction surveys, bundled payment programs, and physician performance measures, the results are transmitted to healthcare enterprises and providers (who are eager to affect improvement based on these metrics) only after months (and up to a year) later. Analytics can be seen as mission control of digital technologies, putting all the pieces together in order to assure ultimate success of the vision. The filtered data needs to be delivered in real-time and incorporated into operational and clinical workflows without having to be mined. Barriers to the adoption of analytics were identified in a joint study by IBM and MIT. The biggest ones were: inability to get the data, the culture does not encourage the sharing of data, lack of understanding of the benefits of analytics, competing administrative priorities, and lack of executive sponsorship. It should be noted that this study was performed in 2010. Nevertheless it is the opinion of this author that these same barriers remain obstacles today. I will discuss some reasons why analytics will ultimately change healthcare.

  1. Analytics will deliver value to electronic health records (EHRs). EHRs were developed to help improve and integrate the flow of clinical information.  However, they were designed as billing tools which also met regulatory specifications.  They do not follow clinical workflows. The American Medical Association has called for design overhaul of EHRs to improve usability.  Clinical decision support is rudimentary and can vary widely in its breadth and depth of use. The discussion of the utility of analytics with EHRs is not new. I suggested what this might look like in healthcare in a piece I wrote in 2011, with pilot studies using predictive analytics have been done.
    1. Analytics can improve clinical workflow. It is intuitive that analytics can improve workflow. Actually determining this by way of metrics has been a challenge. One interesting study from the University of Michigan “focused on measuring clinicians’ ‘time expenditures’ among different clinical activities rather than inspecting clinical ‘workflow’ from the true ‘flow of the work’ perspective.”
    2. Proscribed therapies and digital health tools. Analytics will recommend, based on available data in the EHR (diagnoses, medications, vital signs, results of tests) treatment and discharge plans as well as digital tools for patients (patient education on diagnoses, medication, and follow-up and care instructions. Case managers (as well as the healthcare provider) who have backgrounds in informatics will review these recommendations. This will close the loop as a human element check.
    3. Population health management. ‘Population health’ is currently the buzz phrase for healthcare enterprises. It encompasses preventive health, outreach programs including telehealth, and the use of data to drive health outcomes. Analytics will facilitate this by analyzing real-time data gathered by EHRs, social media, genomics, and mobile health technologies including apps and remote patient monitoring. Crowdsourcing data, whether it is derived from a worldwide or single institutional database is very powerful.
  2. Analytics will transform Big Data into Actionable Data.
    1. Preventing hospital readmissions is becoming a significant focus of healthcare enterprises because of the financial penalties tied to them via CMS. Remote patient monitoring (RPM) is becoming a significant tool in preventing these readmissions by providing continuity of patient-derived data with the hospital, recognizing actionable trending data before it results in a trip to the ER and a subsequent admission to the hospital. One of the unmet challenges of most RPM systems is to incorporate analytics with the technology, offering suggested changes in lifestyle, care, or other instructions to patients and/or caregivers, or changes in the therapeutic plan to the provider.  This is a far cry from the provider receiving a deluge of useless data for analysis.  This type of analytics can also incorporate clinical decision support based on evidence-based medicine.
    2. Use in clinical trials, post marketing of drugs and devices. Analytics can be extremely helpful in the recruitment and retention of patients in clinical trials. There are a few mobile health technology companies in this space. One not mentioned (by way of disclosure to which I am an advisor) is Parallel6 which utilizes patented technology to keep patients and investigators connected. Post-marketing surveillance of medical devices, new pharmaceuticals, and drugs which transition from prescription to over the counter is critical in discovering adverse reactions and other events not captured during controlled (relatively short-term) approval trials or regulated prescribing.
  3. Analytics will be the key to personalized medicine. Only via analytics can we combine the value of population health data and clinical and digital data from an individual patient in an expedited and accurate fashion. Should all patients with the same cancer receive the same treatment regimen? Analytics can potentially readily address variances of diagnosis and/or treatment of a disease based on geography, race, and genomics.
  4. Analytics will decrease gaps/bias in care (geographic, socioeconomic). It is well-known that geographic variations exist in healthcare utilization and costs. Analytics incorporated into EHRs can utilize best practices seen vis-a-vis pooled data such as this to ‘level the playing field’ with respect to both quality and cost of treatment.
  5. Analytics will decrease the cost of care. The use of analytics is readily seen with its incorporation in apps which provide healthcare cost transparency. Analytics can also help patients interested in medical tourism choose a destination. There are apps which allow patients to compare charges for a given procedure.
  6. I do not pretend to deliver the message that analytics is the Wizard of Oz of healthcare, nor that the successful revamping of our broken system lies solely in IT. As described above, barriers to the use of analytics are not technical but cultural. Organizations like Kaiser-Permanente and Geisinger Health System already realize the value proposition of employing high-grade real-time analytics to drive better outcomes and lower costs. It is important for hospitals to realize that remaining in just survival mode is not an option and that a vision of utilizing cost-effective resources such as analytics can be the best investment for success.
Advertisements

About davidleescher

David Lee Scher, MD is Founder and Director at DLS HEALTHCARE CONSULTING, LLC, which specializes in advising digital health technology companies, their partners, investors, and clients. As a cardiac electrophysiologist and pioneer adopter of remote patient monitoring, he understood early on the challenges that the culture and landscape of healthcare present to the development and adoption of digital technologies. He is a well-respected thought leader in mobile and other digital health technologies. Scher lectures worldwide on relevant industry topics including the role of tech in Pharma, patient advocacy, standards for development and adoption, and impact on patients and healthcare systems from clinical, risk management, operational and marketing standpoints. He is a Clinical Associate Professor of Medicine at Penn State College of Medicine.
This entry was posted in analytics, digital health, healthcare economics, Healthcare IT, healthcare reform, medical devices, mHealth, mobile health, patient engagement, pharma, remote patient monitoring, sudden cardiac arrest, technology and tagged , , , , , , , , , , , , , . Bookmark the permalink.

One Response to Five Ways Analytics in Digital Health Tools Will Change Healthcare

  1. Kim Bellard says:

    David – enjoyed your post, and quoted you in a post of my own on a related topic: http://kimbellardblog.blogspot.com/2015/02/let-my-data-go.html

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s