Electronic health records help providers gather vast quantities of data on their patients. But just collecting data isn’t enough to manage their care.

As a result, healthcare organizations are facing new technology choices, looking for applications that can help them take accumulated patient information and do a better job of proactively treating them.

It’s new territory for most providers, and they’re taking two distinct paths. Some are choosing population health management tools from their current EHR vendors. Others are opting for newly developed, specialized applications that take patient information and help care management staff anticipate healthcare issues, and provide pre-emptive treatment.

While reasons vary for selecting vendor products, providers are in agreement on one thing – population health management will force them to use technology to stay ahead of the game.

Picking a known quantity

As Broward Health in south Florida quickens the pace in preparing for population health management, it took a look at “best of breed” vendors, but when it came time to make a decision, the five-hospital delivery system turned to the ancillary tools of Cerner Corp., its electronic health records vendor.

Broward Health issued a request for proposal that contained more than 100 questions, and it received 27 responses, but the system quickly realized it would be difficult to compare all the features of vendors’ systems, and to determine how easily the vendors could integrate with Cerner.

Just as importantly, says Mark Sprada, corporate chief nursing officer, many of the best-of-breed vendors are hurrying to get their products to market, and their products still need additional work, while others were not as sophisticated as they claimed to be.

On the other hand, Cerner could rapidly implement its population health management suite, and Broward could get VIP partner treatment for expanding its contract with Cerner. Further, there’s something to be said for sticking with a vendor you know and trust, Sprada adds.

The timetable calls for the vendor’s HealtheIntent product – expected to be implemented by the end of January – to retrieve data from across Broward and its insurers to identify, score and predict individual patient healthcare risks and steer them to the right treatment programs. The HealtheLife patient engagement and Readmission Prevention platforms should be running by March.

Broward Health is the ninth largest public safety net system in the nation with a large population of vulnerable individuals, including undocumented immigrants and the homeless. It receives $140 million annually from local property taxes to care for vulnerable populations. For example, South Florida has one of the fastest growing HIV rates in the nation, and an important population health management task is to identify those infected and their partners, educate them and make sure they are getting in programs to help pay for medications, Sprada says. Caribbean immigrants are another core target for outreach as they are more prone to hypertension.

Consequently, Broward is engaging in targeted communications to individuals via text and email to encourage visits to get mammograms, substance abuse treatment and physicals, among other services. “We’re really being age- and population-specific with our marketing,” Sprada says.

Accelerating tool development

The Health New England insurance company of five-hospital Baystate Health in Massachusetts is using the

Aerial analytics platform of Medecision for utilization review and care management. Baystate also owns TechSpring, a late-stage technology accelerator, to develop healthcare products in a real healthcare environment, and also operates a health information exchange on which population health management tools can be made available to physicians.

Like any provider moving toward accountable care and value-based reimbursement, Baystate Health has bought software from existing vendors, but it’s also working with entrepreneurs to build new applications, giving developers access to the delivery system’s analytics and informatics expertise.

The organization launched its TechSpring accelerator after receiving a $5 million grant from the State of Massachusetts, says Neil Kudler, M.D., vice president and chief medical information officer at Baystate. TechSpring acts somewhat independently and has partnered with Medecision, which paid for a place to develop its innovation tools, including a clinical care management platform for Baystate’s health plan.

Baystate also is working with TechSpring and Medecision to greatly shorten the time lag between when claims and clinical data are generated, and when the data are available to use. The goal is that payers will have real-time access to claims data, and providers will have real-time access to clinical data, so Baystate and other organizations can avoid using old data to make current care decisions.

“We want a software app that is viewable to care managers and intuitive in presentation to address individual needs of patients and population subsets by presenting pre-analytical real-time clinical data,” Kudler says. With real-time data, providers and payers then can conduct analyses and learn if clinicians are giving the right tests and following the right medication protocols. Data also can help identify patients who aren’t taking medication because they can’t afford the co-pay or don’t have transportation to the pharmacy.

Baystate also is looking at smartphone apps for patients and providers, but early tests show that patients are not quite ready to use mobile communication with doctors, and the doctors are worried that using their smartphones to communicate with patients might run afoul of HIPAA rules.

Boosting revenue, outcomes

Coding is the base on which population health management starts at Village Family Practice, which has 30 providers at five sites in Houston.

To support the move to value-based contracting with insurers, Village Family Practice selected analytics firm Talix so it can analyze structured and unstructured patient information, such as physician notes, to identify missed coding opportunities prospectively, at the point of care and also retrospectively.

Most physicians don’t know their sickest patients, says Clive Fields, M.D., president of the practice. Talix supports use of natural language processing to glean information from physicians notes, scanned into the EHR, to risk stratify patients. The technology also can identify appropriate codes that can raise revenue because it can identify patients who need interventional treatment to improve health status and outcomes.

To improve physician use and appropriate selection of codes, Village Family Practice recently put claims data from insurers into the EHR. Using Talix analytics, customized reports can be generated to present coding opportunities by physician, condition and health plan. A physician, for instance, may code “diabetes” but not code the complications of diabetes. There is a green button on the EHR screen that, when selected, displays potential diagnoses and codes that physicians may not be aware of, as well as other information on patient needs.

Analyzing the claims data enables the practice to identify patients with chronic conditions who are not being seen often enough – that could cause a diabetic whose condition worsens to require amputation of a toe or a foot, for example. At the same time, data analytics has turned into a significant revenue generator for Village Family Practice because patients are getting more comprehensive treatment, thus improving outcomes. Because patients with multiple chronic conditions now are better identified, care coordinators know who needs to be monitored more closely.

“Any well-trained doctor can apply treatments to improve outcomes,” Fields says. “But they need to focus on the most chronic patients.” The practice has a panel of 2,000 patients – about 300 of those have diabetes, with about 100 of them considered to be high risk. Now, the practice better knows these patients, as well as others with risk factors such as high body mass index, hypertension or congestive heart failure.

Valley Family Practice needed about six months to adopt the new tools. “As risk stratification drives care coordination and clinicians learn of success stories, they will want to make sure their patients are getting that level of care,” Fields says.

Even playing field

Every organization engaged in population health management is doing it a little differently. In Falls City, Neb., 25-bed Community Medical Center with a five-provider clinic, also is relying on its EHR vendor, NextGen Healthcare Information Systems, as the anchor technology provider for population health management.

The hospital may be small, but it has an active patient outreach analytics initiative and is assessing how accurate Medicare is with its quality determinations, which affect reimbursement rates. It also is looking at getting health information exchange software from the Mirth unit of NextGen.

At Community Medical Center, Ryan Geiler, assistant manager and clinical analyst, gets reports from Medicare on how well providers are treating diabetics, and he says his own data shows the hospital is doing much better. He’d like to advance past canned reports and do deeper analyses, not only on quality measures to support a case for higher payments, but also on the “leakage” rate of providers, which reflects gaps in care.

He wants to know, for instance, how many patients have been diagnosed with diabetes but don’t have a primary care provider, or have a provider who isn’t managing the diabetes. “Where are these patients?” he asks. “Are we missing them or does the other practice in this area have them?” Geiler also wants to know where orthopedic patients are getting treated, if not at his hospital. Are the region’s schools sending injured football players to Omaha? If so, he needs to educate the schools on the expertise at their local hospital.

Community Medical also is marketing itself to the overall community, having aggregated patient lists and set up a rules-based hierarchy that governs an automated communications system. If a provider reaches out to a patient via the patient portal and the patient doesn’t respond, then an email is sent. If that doesn’t work, a postal letter is sent. Letters, so far, are turning out to the best way to get a response, but the hospital soon will add text messages.

In addition to using NextGen analytics to assess payer reimbursement, the hospital now is identifying patients with diabetes, high levels of lipids, high body mass index levels, hypertension or thyroid disease, and getting them in to see a doctor. In the first month, 28 new thyroid cases were found. A flu campaign resulted in 76 percent more patients coming in for a shot during the past year.

The population health program is working at Community Medical Center, but its start was rocky. Geiler learned an important lesson early—he was overwhelming the staff. “I’m a nurse, but mainly an IT guy and I thought getting data out of the system and contacting patients would be easy.”

Educating patients on why they are being contacted, and the importance of that contact, worked better than expected, but the staff wasn’t ready, Geiler says. But policies needed to be set, such as standing orders on what other risk factors and treatments these patients might need. Population health management means changing the mindset of clinicians to be more proactive in the care they provide.

Fewer readmissions

As a Medicare Pioneer ACO participant, reducing hospital readmissions is a priority for Montefiore Medical Center in New York City. Henry Chung, M.D., CMO of the care management program and medical director of the ACO, worked with the IT department to develop an algorithm that pulls EHR and claims data and runs it through the 3M Clinical Risk Grouping software to stratify high-risk patients being discharged.

Centralized discharge transitions were created so nurses could call patients for a couple weeks following treatment to discuss how they are doing, if they have their prescriptions and know how to take them, if home care services have arrived, and to confirm their next physician appointment. In these calls, nurses are learning that some patients don’t like the doctor they are seeing, so the care plan is reworked to involve another physician.

Montefiore also is using the 3M software to stratify all patients in the ACO into risk categories, to determine which patients need immediate care to prevent hospitalizations. Clinical risk grouping software along with claims data from large health plans in the Bronx also can track the performance of owned and independent medical practices and educate the practices, Chung says.

Now, Montefiore is rolling out new technology, using a mobile care management platform to engage high-risk patients through interactive care plans, starting with Medicaid beneficiaries. The vendor is SenseHealth, a start-up company that’s part of the portfolio of the New York Digital Health Accelerator. Montefiore has worked with SenseHealth for nearly three years as the product was developed, Chung says.

The SenseHealth platform enables clinicians to text messages to patients, and patients then can respond. The clinicians inquire about their progress and talk about the importance of medication adherence, upcoming appointments and other issues. The program also will target behavioral health by sending motivational messages to patients. Almost all high-risk patients, even the homeless, have a cell or smartphone, “so you’re coming right into their stream,” Chung says.

A lesson learned in the pilots was that most mobile programs require a smartphone and a data plan, so an organization needs to know which of its populations are more protective of data and not willing to share it, and also need to make sure that participants know how to load apps on their phones.

Now, Montefiore is looking at offering “televisits” through which patients can use Skype to communicate with a clinician. There is a place for that, Chung believes, but audio-visual consultations without vital signs have limited value. He’s also looking for a device to guide a patient through a remote examination, to take a picture of the throat, skin or ear, take their temperature, or take their respiration and heart rates.

Register or login for access to this item and much more

All Health Data Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access