Cloud-based platforms allow sleep clinicians to work from anywhere with an Internet connection. Increasingly, they also allow a means to increase compliance of therapies such as CPAP, and some can even predict therapeutic outcomes before they occur.
When it comes to getting patients on CPAP, sleep clinicians and techs can only do so much. They can educate patients and help them find the right device and mask in the office, but even the most dedicated sleep clinician can’t follow a patient home to watch their guidance pan out. Now, some manufacturers are using cloud-based data storage to help providers keep a virtual eye on their patients’ progress.
Several CPAP manufacturers have developed proprietary software that helps clinicians and durable medical equipment providers (DMEs) zero in on patients who are having trouble adhering to CPAP therapy with the goal of improving compliance.
For example, Philips’ Care Orchestrator (an update to its previous software, EncoreAnywhere) is part of its “ecosystem” that also includes the patient-facing app DreamMapper, says Tim Murphy, general manager of new business solutions at Philips. The system works by using data that Philips has already collected from its devices to inform algorithms that help discern patients who need help and who would be receptive to counseling.
Murphy says the cloud-based software can tell a clinician, for example, “of these 10 patients that you’re treating, the patients at the top of the list are exhibiting attributes that say they are more likely to be productively responsive to an outreach by you. They’re more apt to take your advice and change their behaviors than patients 8, 9, and 10, who are showing behaviors that say they are probably less likely to take your advice and take action.” This allows the home care provider to focus their time on individuals who are more likely to be receptive to clinical intervention.
This type of patient monitoring is called “management by exception,” and it appeals to busy DMEs who struggle with reimbursements, says Amy Cook, director of health care informatics at ResMed. DMEs are “starting to really look at how they can drive more efficiencies within their practice,” she says. “So whereas they might have been fine looking at every patient every day in the past, in order to drive efficiency in their business, the idea of management by exception is becoming more attractive.”
ResMed offers a cloud-based platform called AirView to track patients’ progress, which can be paired with its management by exception program U-Sleep. The combination of AirView and patient-facing app MyAir improves compliance dramatically compared with only one monitoring system alone or no monitoring at all, the company says.
If a particular patient is struggling, clinicians or DMEs can tap into these platforms to communicate with them or to even make remote pressure adjustments to their devices.
While Philips and ResMed’s platforms are specific to their devices, Somnoware is an independent software that integrates with 150 systems. Somnoware too employs predictive analytics to estimate which patients will have trouble complying with PAP therapy. The software starts with an initial estimate of a patient’s risk factor based on demographics, past compliance with other types of therapies, and severity of sleep apnea, among other variables.
“It’s a dynamic model that is based on machine learning,” says Raj Misra, chief data scientist and vice president of marketing at Somnoware. “So what happens is as we get more and more data on the patients, it improves the accuracy of the prediction. At day 0, the baseline prediction is fairly accurate in most cases. But as we start getting daily usage data for the patient, we take that data and use machine learning to update the model and update and refine the predictions.”
The company isn’t boxed into PAP therapy only, Misra adds.
“Sleep apnea is a disease state,” he says. “There’s nothing in our platform that prevents us from looking at other forms of therapy.” Oral appliances, for example, could be Wi-Fi-enabled to transmit usage data to providers. “In terms of our market share, obviously the vast majority of our user base is centered around PAP because that is the 800-pound gorilla in the industry, but our solution definitely works with every kind of therapy.”
Another company focusing on predicting patient success from the get-go is App-Nea, which makes MAD FIT, an algorithm that determines appropriate titration for patients’ oral appliances from the very first fitting. Pankaj Singh, DDS, CEO of App-Nea and a practicing dentist, developed the algorithm retrospectively by looking at results from his own practice. He sees the next phase of MAD FIT, which has been beta-tested and is in the process of a soft rollout, as a prospective study.
“It’s [been] very subjective how a dentist determines that start position, and it’s not very scientific,” Singh says. “What this algorithm does is determine the optimal titration point as a starting point taking into account over 100 different data points that are very easily accessible and gathered by the dentist.”
The data fed into the algorithm depends on what the dentist has available for that particular patient; the more data, the better, but the algorithm works with what the dentist has, Singh explains. “They’re as simple as dental and orthodontic data points extracted from x-rays if they have it, panoramic x-rays, CT scans, and sleep studies,” he says.
ResMed and Philips also rely on the extensive library of data they’ve collected to help manage patients in their respective portals.
“[Care Orchestrator] is benefiting from the 10-plus years that Philips Respironics has been engaged in cloud-based clinical management,” Murphy says. “That history is built upon strong sleep physicians and even now respiratory physicians. It benefits from the value of that 10 years, plus over 5 million connected devices, and 2 and a half billion sleep patient nights that are available to ensure and harvest, to build what now we know as Care Orchestrator.”
Cook notes that ResMed recently recorded its billionth night of sleep data, which will help the company provide actionable data to customers—data that can be used to pinpoint problems and implement solutions. A study the company released earlier this year that found patients with central sleep apnea were more likely to be compliant on auto servo-ventilation compared with CPAP came about as a result of this data collection, she says.
All manufacturers see their products as time-saving devices, even though at first blush one might think that logging on to check on patients would add to a clinician’s workload.
“We’re really about giving them the ability to deliver a very positive clinical outcome and experience for the patient and giving them an opportunity to be as efficient as possible, knowing that their economics and their time are being constrained and pulled in a lot of different directions,” Murphy says.
Rose Rimler is associate editor of Sleep Review.
