The advent of the Electronic Health Record (EHR) came with promises: a world where clinically valuable patient information became easily accessible to everyone involved in the chain of care. EHRs, we were told, would democratize data and facilitate easy sharing between hospitals, primary care providers (PCPs), specialists, device manufacturers, and patients themselves, paving the way for improvements in communication, operational efficiencies, cost savings, and, most of all, patient outcomes. It was the power of the internet applied to medicine.
But eight years after the Affordable Care Act mandated “meaningful use” of EHRs in 2014; the American health system still struggles to realize the fulfillment of those promises. Why? In a word: “interoperability.” Interoperability became the buzzword capturing the zeitgeist of challenges in the post-paper era.
Simply described, it is the idea that health record interchange between EHR systems should act as more than a glorified digital fax – one that is only accessible in the darkest corners of an EHR. Interoperability standards dictate that shared information (even between disparate systems) should not only be live, cloud-based, and easily accessible but be integrated at a deep level, to the point that imported data could enrich a patient’s health profile, predict risk categories, provide contextual information, etc.
Achieving Interoperability in Sleep Medicine
The ache for well-applied EHR interoperability is perhaps felt most poignantly in the world of sleep medicine. Providers outside of the direct clinical sleep community, like PCPs, cardiologists, dentists, neurologists, and hospitalists, are all awakening to the reality that sleep disturbances like obstructive sleep apnea (OSA) have direct consequences on the diagnoses they specialize in.1, 2, 3, 4, 5 Consequently, testing for sleep disorders are growing in popularity and are ordered from this broadening base of clinical environments.
The consequence of such a multi-layered system is that new complexities arise. Sleep centers struggle to keep up with the flow of information coming from all these sources, and workflows are largely manual and spread across 5-6 various healthcare record platforms to track patient screening, scoring, diagnosis, treatment, and compliance. Many sleep labs that are fed up with the difficulties are turning to so-called “middleware” to address these challenges and achieve true interoperability.
What is Middleware and How Does it Help?
At its core, middleware is software designed to communicate with other platforms. In sleep medicine, it is being used to connect EHRs, sleep devices, schedulers, scoring platforms, DME ordering, shipping logistics, etc. – all in one place, essentially “hacking” interoperability out of systems that weren’t designed for it. Doing this has several positive benefits.
#1: Automating Evaluation to Treatment
When a patient is being evaluated for sleep apnea, sometimes even what is thought to be a routine diagnosis can be complex, with a pathway that involves, at minimum, first clinical screening, home or in-laboratory sleep test orders, scoring, results communication, insurance authorization, and obtaining durable medical equipment such as positive airway devices. This workflow often involves many routine manual tasks that take time and human capital to get through, placing additional stress on technologists and other staff and thus limiting the volume a lab can support.
Middleware can centralize all this information and put many of these tasks on autopilot. Unfortunately, it isn’t until implementing a solution like this that lab operators realize how much time these rote duties used to take. With the current sleep technologist shortage, the value of freed time goes a long way in improving operations and making space for revenue-generating procedures.
#2: CPAP Adherence and Reimbursement
Evaluating a patient’s success with therapy remains an enormous challenge for sleep labs. Effective follow-up programs require human capital and interruptions in the course of a patient’s daily life. When most sleep clinicians think about digitizing the EHR, PAP follow-up is the main benefit they hope to achieve. And for a good reason – a 2018 2-million-participant study showed the incredible impact of remote PAP monitoring: 75% adherence to CMS utilization criteria (at least 4 hours of use in > 70% of nights during 30 consecutive days in a 90-day period).6
Thankfully, the last few years have seen the rise of cloud-based telemonitoring to make follow-up more effective and improve reimbursement. Care Orchestrator from Philips is an example of this, but third-party companies like Somnoware are paving the way for integration with all of the devices or software a lab could possibly use. These platforms not only automate initial patient evaluation but can also track device usage, compile utilization dashboards for entire patient populations, and monitor the effectiveness of various follow-up strategies.
Dennis Hwang, MD, is the medical director at the Kaiser Permanente San Bernardino County Sleep Disorders Center and co-chair of sleep medicine for the Southern California Permanente Medical Group, with a particular practice interest in realizing the promises of interoperability not just in the future but now. After positioning themselves to capture an enormous amount of sleep diagnostics, PAP devices, and patient-reported data, Dr. Hwang encountered a familiar dilemma: “The problem with data is if you don’t have the tools to do something with it, the data becomes overwhelming.”
In response to this, the network of sleep centers adopted Somnoware to unify and centralize all the disparate information while also automating certain adherence follow-up tasks to trigger based on a specific set of live patient behaviors. A time-motion study showed that by using an automated risk-identification method, an 83% efficiency improvement in patient care was realized compared to a manual process.
#3: The Great Leveraging of Data
Telemonitoring with good EHR integration and rule-based automation only scratches the surface of what’s possible in truly interoperable environments. Artificial Intelligence (AI) and Machine Learning (ML) offer us a glimpse into an even more exciting future where meaningful conclusions can be drawn from across the entire patient journey.
In a conversation with Somnoware’s CEO, Subath Kamalasan expressed excitement over the operational outcomes they’ve been able to achieve but noted that the future of applying AI to sleep is much more compelling. He notes that, since patient behaviors are one of the most critical components to successful sleep apnea treatment, AI’s highest value proposition for patients and clinicians will be in its ability to predict those behaviors and make contextual whole-person treatment recommendations.
Somnoware is building towards a future where software-generated recommendations can be made and then automated based on the specific follow-up or PAP habits of each patient. Is a text 21 days after beginning therapy more effective than an email to engage your patients and increase adherence? How are a patient’s comorbidities affecting their sleep? These are the types of questions that AI applied to aggregated, anonymized patient data could answer in an automated fashion.
But going even further, the term “whole-person” here matters – Subath’s team wants to see interconnected data touching each specialty in a two-way flow of information: CPAP adherence data used to continually update cardiovascular health scores, neurological stroke events used to predict apnea risk, etc. These elements represent the most promising ideals of the interoperability dream put into practice.
Interoperability is Coming Sooner than We Expected
Interoperability is needed to realize the promises offered by EHRs, but sweeping change across an entire healthcare system is complex. A 2019 paper by Lehne describes the challenge:
“Most of today’s medical data lack interoperability: hidden in isolated databases, incompatible systems and proprietary software, the data are difficult to exchange, analyze, and interpret. This slows down medical progress, as technologies that rely on these data – artificial intelligence, big data or mobile applications – cannot be used to their full potential.”7
The most notable organization looking to make a difference is Health Level Seven (HL7). This international non-profit has created the most widely agreed-upon set of interoperability standards known as FHIR (Fast Healthcare Interoperability Resources).
But while getting everyone in American healthcare on board is a challenge, industry giants like Kaiser Permanente, the VA, and Wellstar Health have leveraged companies like Somnoware to unify their experience, save time, enhance diagnostics, and, most importantly, improve patient outcomes. It’s encouraging that broad interoperability in healthcare is coming, but it is even more exciting to think about the doors it may open for tomorrow.
Nathan is a Nurse Practitioner and Freelance Writer based in West Palm Beach, Florida. He sees disabled veterans struggling with sleep disturbances in the context of multiple comorbidities. You can find him at clinicious.com or on his LinkedIn profile.
- Yeghiazarians Y, Jneid H, Tietjens JR, Redline S, Brown DL, El-Sherif N, Mehra R, Bozkurt B, Ndumele CE, Somers VK; on behalf of the American Heart Association Council on Clinical Cardiology; Council on Peripheral Vascular Disease; Council on Arteriosclerosis, Thrombosis and Vascular Biology; Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation; Stroke Council; and Council on Cardiovascular Surgery and Anesthesia. Obstructive sleep apnea and cardiovascular disease: a scientific statement from the American Heart Association. Circulation. 2021;144:e56–e67. doi: 10.1161/CIR.0000000000000988
- Quan SF, Schmidt-Nowara W. The Role of Dentists in the Diagnosis and Treatment of Obstructive Sleep Apnea: Consensus and Controversy. J Clin Sleep Med. 2017;13(10):1117-1119. Published 2017 Oct 15. doi:10.5664/jcsm.6748
- Salas RE, Chakravarthy R, Sher A, Gamaldo CE. Management of sleep apnea in the neurology patient: Five new things. Neurol Clin Pract. 2014;4(1):44-52. doi:10.1212/01.CPJ.0000442583.87327.5d
- Culpepper L, Roth T. Recognizing and managing obstructive sleep apnea in primary care. Prim Care Companion J Clin Psychiatry. 2009;11(6):330-338. doi:10.4088/PCC.08m00725
- Sharma S, Mukhtar U, Kelly C, et al. Recognition and Treatment of Sleep-Disordered Breathing in Obese Hospitalized Patients May Improve Survival. The HoSMed Database. Am J Med 2017; 130:1184.
- Cistulli PA, Armitstead J, Pepin JL, et al. Short-term CPAP adherence in obstructive sleep apnea: a big data analysis using real world data. Sleep Med. 2019;59:114-116. doi:10.1016/j.sleep.2019.01.004
- Lehne M, Sass J, Essenwanger A, Schepers J, Thun S. Why digital medicine depends on interoperability. NPJ Digit Med. 2019;2:79. Published 2019 Aug 20. doi:10.1038/s41746-019-0158-1
Source: Sleep Lab Magazine Sept/Oct 2022
Be the first to comment