What integration and what systems?
One of the biggest challenges in health information technology today is not do we need healthcare system integration, but rather identifying what is the highest priority problem we have and need to engage in?
By now, most healthcare organizations have either implemented an electronic medical record (EMR), some have even implemented a couple, or are buying a new system as we write this. Health information exchanges (HIE) are out there, as are electronic health records, as well as personalized health records. These data records are all now part of meaningful use and the everyday lexicon of healthcare providers.
So what now?
Business Intelligence? Analytics? Predictive Modeling? Simplifying regulatory compliance processes? Every organization that we consult with is working on most of these issues right now. The question many are asking now is what can we do now to improve our operational performance and prepare for the future challenges associated with risk based care, and the transition from fee-for-service to a risk model like an accountable care organization (ACO) or a clinically integrated network (CIN).
The answer is in looking more closely at the specific problems people within the organizations are having that can be solved with the right focus and some integration work. People we talk with are trying to meet meaningful use and one of their biggest challenges is how to get patients to engage in their own healthcare decision process by using a patient portal. We have heard stories of healthcare systems calling patients asking them to log on and use the portal in an attempt to meet the meaningful use criteria. If you have to call patients to get them to use their own personal health record then there is probably something missing in what is presented.
Patient engagement critical for success
Our observations at this point indicate that most health systems and payers are convinced now that patients must engage in their own healthcare situation or outcomes in quality and cost are not going to go in the direction they need to. In our experience, the reason patients don’t engage is they either:
- don’t understand the data they are given, or
- it doesn’t translate to actions they can take or will take.
Patients tell us they want trends and choices that lead to simple actions that will also provide them feedback. Things like:
- lab results, for a diabetic, trending over time to show improvement or degradation can be motivating for that patient to act.
- integration to information that tells them what actions to take for improved outcomes, such as a prescribed diet in their electronic health record that they discussed with their doctor or care coordinator, and that information is displayed without searching for it.
Patients want choices about which action to take first and that must also be part of the data set they have access to which means integration to the application being used by their care coordinator is important.
Improved outcomes stem from healthcare systems integration
As part of getting better outcomes, it is essential that the patients’ care coordinators also have the data they need. Optimally, this includes the entire electronic health record integrated into a single record and normalized for high productive and high quality use. The care coordinator should also have the exact same view of the personalized health record that the patient has to create improved collaboration. So what is the current problem with patient data? Most HIE’s in the healthcare environment have been designated as the working database to gather all the data and normalize it for presentation. But, HIE’s do not have all the data needed to do this.
In many cases, HIE’s have very little patient data other than potentially lab results, for distribution. Why? Healthcare systems integration is challenging, and this difficulty is clearly evident in the results we have seen in the market so far. As a system-agnostic, yet multi-vendor experienced systems integrator, Orchestrate Healthcare is asked on a daily basis to do integration projects.
Challenges to successful integration
When there is an interface engine involved, and HL7 level integration, it is straightforward and relatively easy to accomplish. Throw in biosensors, tightly focused analytics solutions, and smart monitoring, and the integration suddenly becomes much more difficult. The variety of systems just in EMR alone makes integration very demanding and the lack of progress in useable data can be tied directly to that problem alone.
Help is on the way, IF you believe that CCD, CXA, and CDA are going to be the holy grail of simple integration. Early indications are that many CCD’s, and their associated derivatives, are not easily consumable between vendors. In many cases, you have to send the CCD as a PDF like document that doesn’t allow for discreet data integration – which then means you either have to parse the data, through another level of integration, or you have to force the user to work with results that are flat and not trendable; which defeats our objective of giving care coordinators and patients data that will be actionable not just data.
In working with HIE’s themselves, either stand-alone or owned by healthcare systems, they are committed to delivering greater value to their customers. Many are now being asked to provide:
- regulatory reporting assistance
- care coordination data including ADT alerts
- personal health record level information
- dash boards and reporting to administrators
With these demands, the lack of data they really need to have aggregated, but do not have due to integration issues, coupled with the need for healthcare itself to improve on cost and quality, the priority of WHAT to integrate WHEN for patients and care coordinators becomes especially challenging.
So how do we go forward from here?
How do we prioritize what to integrate? It may seem controversial, but the answer is probably not the entire electronic health record of data from all sources. That flies in the face of recent incentives to integrate everything that has information about a patient into a single file. The best scenario would be to have all the data and then sort it so that specific users can get what they need at any time and place. This course of action has proven to be impractical. However, through discord, we have learned far more about what data we really need to have to be impactful in improving care.
Simple data served in near real-time seems to be the solution most want right now. Blood pressure, blood glucose, heart rate, pulse ox, and a few simple social behaviors such as: what causes the patient anxiety and when do those conditions typically occur. But, these examples present serious challenges in the healthcare systems integration process.
How can we improve healthcare systems integration?
How do you capture data from a blood pressure cuff, identify it to a particular patient, store it, normalize it, and make it available in a trend. This is true for any bio-data we would encounter, but it has to be integrated for healthcare and patients to make the kind of progress needed for improved health. For patients with chronic disease, which is by far the largest group by number and costs, seeing differences in bio-data is really the only measurement typically looked at. We want to prevent negative medical events from occurring and these events can be more predictable as trends are charted in the patients’ bio-data.
Making the challenge even greater is the fact that most measurable bio-data either has to be manually integrated through data entry, or have custom integration done for each type of device or system gathering the data. This challenge must be met at some point, now or in the future. The value of this data is clear and impacts the care paradigm right now. Clearly capturing would be a very useful step provided that integration isn’t impossible. If we cannot integrate this data and we haven’t been successful in integrating the necessary data in HIE – then are we constrained to simply using EMRs to create electronic health records by exchanging data with other EMR’s from the same vendor? IE: Epic to Epic?
Developments in integration will advance care
A recent development in integration work is the implementation of newer level personalized health records (PHR). These advances capture data from devices, monitors, and sensors and then can feed the data back to HIE’s or EMR’s as long as those systems can consume the data. Most newer PHR’s employ CCD or CDA standards, but this does not insure consumption by HIE’s or EMR’s. Recently, we’ve been working with more customers who have engaged with us to deliver this type of integration based on our knowledge of interface engines and HIEs. Though strenuous, it is possible and really a matter of deciding what you want integrated and asking for the most qualified healthcare systems integration vendor to get it done.
In the case of HIEs, we often see users ask for only the things they are aware that an HIE can supply, when in reality they should be asking for what they need – and then having the HIE quote the custom work necessary. This is not the typical direction as custom integration work by an HIE is not the norm. We know from our experience that there are ways to do this and the cost estimates should be looked at versus the cost of not having the data formatted and integrated the way it would be most beneficial. Without having the data needed, you are either working around it and the associated work-around costs will appear elsewhere, or your care cost is higher because you don’t have enough information to treat your patients at a more specific level that the data would allow.
Focus needs to be on the patient and provider
In conclusion we believe, from many different experiences in the market, that healthcare integration strategies in late 2014 and 2015 must focus on the patient and the care coordinator/provider. As part of this, data must be provided to the patient in a way that can actually be understood and used by them. The care coordinator needs a larger data-set than the patient, but for collaboration their needs are similar. To accomplish this, successful integrations must include bio-data that:
- hasn’t been captured directly in the past, and
- may not be from the patient as a source.
Further, integration needs to have a heightened focus on just the data that can drive changes in behaviors, care processes and outcomes.
At Orchestrate Healthcare we proudly claim the title as being the best integrator in the US. Our customers ask more from us on a daily basis in healthcare systems integration, and we deliver standard HL7 integration as well as CCD, CXA, CDA, and now custom integration. If your organization has a mindset of the value of integrating the right data to the right places at the right time, then the transition to more custom integration needed to get greater impact now is your next step towards more beneficial outcomes. Decide what data you really need, regardless of what you think is possible, then engage with us to help you get there in either a strategic discussion, a tactical one, or both.