A Truly Integrated Approach to Health Information Processing Could Provide a ~500% ROI

Integrated Health Information

If you are like me and you read something like this first thing in the morning you would probably choke on your coffee forcing back the laughter! However, the digital transformation in health care is taking place today, and the pace of change is quickening. More and more location-based data in population health is being leveraged, mobility is being employed as a physician economic behavior modification tool to support shared savings, pay for performance and risk contracts, and lastly sensor data is making its debut! This digital transformation is an incredible shift for a market that has historically focused a good portion of its financial and human resources on the traditional implementation of electronic health records and health information exchanges to support meaningful use. This new effort to digitally transform healthcare is going to require new business models, architecture shifts and most difficult of all a cultural shift away from the traditional information technology mindset.

No matter if the ROI in a new integration strategy is 500% or 75% the justification for healthcare to move away from traditional models of doing it yourself interfacing to integration as a service makes more sense now than any other time in history. The need to integrate more trading partners such as physicians and post-acute care networks, reduce the number of FTE's working on integration, eradicate application downtime due to interface needs and, improve performance and cost over legacy middleware solutions are all large contributors to the ROI on a platform service solution for integration. This type of integration need does not support the single dimensional legacy interface technologies or the project based approaches to delivering integration performance. Health care has to embrace smart built, intelligence based, consumer quality, rapid innovations that completely redefine how providers, payers and consumers of medical care communicate. This permanently changes the integration landscape from point to point communications to the real inclusion of all functions required creating one pane of glass for all integration services on one platform as a service offering.

At a minimum, this single pane of glass for integration should consist of service and micro-service API layer, machine learning and in-memory associative data construct, natural language processing, business process management, event management, rules for stateful and stateless models, and a visualization layer for streaming, static and, mashup analytics. So this description begs the question as to why you need all this technology to interface two applications? The answer would be simple; you don't. But that is not what we are suggesting either. The key to true interoperability in health care is less about moving data from one container to another and more about doing stuff with data as soon as it becomes available versus waiting for a traditional database to acquire, prepare and, transform the data for a retrospective view. The importance of this is analogous to flying a commercial airliner on air traffic controller and weather information that is minutes’ old! I don’t know about you, but at 18,000 feet in the middle of summer flying through stormy weather while trying to land at Newark International Airport in New Jersey, I want that pilot to have up to the Nanosecond information with the intelligence added to provide only the information that the pilot needs to help land that aircraft safely! The pilot does not need the weather for JFK and LaGuardia just the information for Newark so adding that intelligence around information processing is critical to landing the plane. So why are patient care and the business of running a medical organization any less important? Lives are at stake, any mistake can be devastating, and it requires the aggregation and presentation of information from a myriad of sources in one easy to use a single pane of glass for the pilot to perform his/her job safely which is no different from a physician or care provider. 

So how can the claim be made that a competent interoperability strategy can provide a ~500% return on investment? It is not trivial but relatively straight forward. It does, however, require health care enterprises of all sizes to change their perspective on integration and move away from in-house solutions over time to an integration as a service model. This is counterintuitive to the usual thought process that occurs in any size healthcare enterprise but a shift that is none the less critical to reducing IT spend and rationalizing application portfolios to right size the information assets the organization in question maintains. For the healthcare enterprise to re-balance the cycle of innovation to meet the demands of a consumer grade, intelligence driven, risk bearing, and agile business, this is no longer a question of "IF" but how fast can we get there! To attain any level of hard dollar ROI in healthcare, we have to start thinking about integration as a way to optimize and innovate our environment leveraging existing investments in EHR and population health tools and completely automating creative integration processes.

Focusing our integration efforts on creating reusable assets that provide a single pane of glass “interface” to the internal and external information sources required to operate the healthcare enterprise is the first step. Moving away from traditional interface techniques over time to simple Micro-Service based components that can be easily assembled to reduce the overall development lifecycle and go to market exponentially faster than conventional methods. Redirect current IT resources toward innovative solution versus commodity based interface work. Reduce linear processing associated with traditional interface environments which will reduce hardware costs over time. Decouple integration logic from applications creating a lighter more agile integration leveraging cloud and intelligent devices to propagate information immediately where and when it is needed for a particular purpose. As healthcare enterprises continue to add new tools to their IT architectures and depend more and more on cloud-based solutions with mobility as part of the equation, it becomes critical to achieving real interoperability as efficiently between distributed information sources.

So how does this answer the ridiculous ROI claim? Simple! The healthcare enterprise can integrate more information sources and applications faster with the intelligence built in to deliver fast and relevant data with fewer human resources and reuse those integration services over and over in the future. This speeds the time to integration, reduces required staff, drives higher productivity, reduces downtime, reduces hardware and software costs versus traditional methods. The ROI period is dependent on the individual organization and the aggressiveness with which they convert their current portfolio of integration point and applications. If the enterprise chooses to outsource this function including infrastructure to a cloud-based vendor the time to return on investment can be compressed dramatically. The methodology to measure the actual return and benefit should be documented and set forth with either cloud or on premise approaches and has to be benchmarked against current performance.

Steps should include savings from the reduction of:

  • Staff
  • Hardware
  • Maintenance
  • Support

And improvements in:

  • Productivity
  • Revenue
  • Shift from maintenance to innovation
  • Shift from CAPEX to OPEX

Based on this focus the estimated payback should be targeted for 36 months again depending on the size and complexity of the enterprise.

ROI calculations should be targeted for the following measures:

  • Net present value based on three-year investment based on the original sum with a 10-15% return to account for missed opportunity cost. This provides support for the cost of capital and internal rate of performance.
  • “Information Downtime” is the number of hours/days that critical information for application features and functions to operate is unavailable multiplied by the number of users/clinicians affected.
  • Productivity loss calculated as salary burden multiplied by application unavailable and or information downtime
  • Revenue opportunity loss is the information downtime multiplied by the average revenue per hour
  • Revenue loss/opportunity is defined as the data downtime impact on the end user's productivity/impairment 

Calculations of user productivity and lost revenue opportunity are somewhat subjective so every enterprise should choose a reasonable fraction of the number derived in the ROI exercise to apply to their final savings estimate. Also, ROI should be calculated starting the first full month after implementation of the new integration architecture.

The pressure on the healthcare enterprise large or small to use ever increasingly complex clinical and business applications is only going to increase. With that comes the need to integrate efficiently the myriad of data sources required to support the trade and clinical processes for which those technologies are purchased. As healthcare enterprises become more and more dependent on these applications to support at risk models of contracting and care the achievement of interoperability in an expedient and efficient manner is outpacing the ability for organizations to survive in a do it yourself model. Add to this the desire and need to engage through a mobile channel, have self-service analytics and appropriate business and clinical alerts and the reasonable approach to the required interoperability is to partner with a reputable cloud vendor who can remove this burden entirely.