It was a few years ago when clinical and information technology leaders at the Hartford, Conn.-based Saint Francis Care committed to upgrading its technology platform, specifically its electronic health record (EHR), first on the inpatient side of the medical center, and then a year later on the outpatient side. It was also during this time when senior leadership at Saint Francis Care, now part of Trinity Health - New England, an integrated healthcare delivery system, posed to itself the core question of, “How can we better support our hospital mission, which is to [provide] the best care for a lifetime at Saint Francis Care?” says Danyal Ibrahim, M.D., chief data and analytics officer, emergency department physician.
And according to Ibrahim, the other key question to consider was that as the general healthcare landscape moves from volume to value, what were the organization’s capabilities on the analytics side? To this end, Ibrahim says that Saint Francis Care's assessment around analytics first involved identifying a data strategy, which was again, broadly speaking, to transform data in the organization to support better care, better health, and lower costs for a lifetime.
Drilling down, Ibrahim, whose specialty is toxicology/poisoning and says half of his time is spent in clinical with patients, while the other half is overseeing the analytics in the organization, notes that the analytics assessment was composed of direct interviews with very basic questions in a survey format that was sent to 100 individuals across the system in clinical, financial, IT, administrative, and leadership departments. The questions that were asked applied to the following various domains, Ibrahim reports: patient safety, quality, efficiency of the care delivered, patient engagement, patient experience, care coordination, care redesign, revenue and market share, population health, and workforce engagement.
“In each of these domains, I started off by asking if we have the data and analytics,” Ibrahim says. “So for the metrics, measurement, monitoring, and improvement efforts, do we really have that data? Second, do we have all the components of the data? Can you tell a comprehensive and meaningful patient story from when they come until they leave? Third, if we have the data and it’s integrated, do we apply sound analytical methods to it to derive meaningful conclusions? Are we presenting it in timely and intuitive manner? That speaks to the buy-in and the validity,” he says.
Danyal Ibrahim, M.D.
Ibrahim feels that in healthcare, there is often mistrust about the data which leads to drawing conclusions about incomplete data. So that led to the last question of the assessment—as ultimately the purpose of analytics is to aid and make better decisions at the point of care, can it identify the what, the who, and the where? “What’s going on, who is involved, and where is it happening? Ultimately, why is it happening? That will help apply the right intervention to move towards improvement,” he says.
Ibrahim says the assessment revealed that Saint Francis Care was doing well on the domains of revenue and cost, meaning it had a good sense about what its utilizations were and the cost around that. But when applied to the other domains such as quality and safety, efficiency, and patient and workforce engagement, gaps were identified. “The data was incomplete or the pieces weren’t integrated in these domains; we were not presenting it in a timely or simple way,” he says. “So ultimately the value of it was not intended, meaning we were not able to use it in a way to push the initiative, through identifying the opportunity and applying the intervention.”
The next stage for the organization was using the results of this assessment to create a plan of action. “We realized through the interviews and surveys that our data stewards and data domains are siloed,” says Ibrahim. Some of data is in the finance department, some of it is in IT, and some of it is in the quality department. So number one, we wanted to break those silos, so we can work together towards our common goal.”
As such, Ibrahim and his staff wanted to build a solid and organized team that incorporated all the different skillsets from multiple areas of the patient care system. “If you look at how analytics teams are siloed in our industry, you have financial analysts, IT [people], BI [business intelligence] analysts, and then clinical quality analysts. So we wanted to come up with multi-disciplinary team for all these fields,” says Ibrahim.
After tackling the people aspect, the next step was to tackle the technology. “We needed to be nimble in how to build the data platform that will accommodate our various data sources,” Ibrahim says. “We largely have one EHR [from Epic Systems], but we wanted to build a platform that allowed us to bring in data from other sources, too. We wanted to be able to bring in the data, clean it, normalize it, aggregate it, and present it in way that was meaningful—and then adopt meaningful measurement metrics,” he says. To this end, Saint Francis Care engaged in a partnership with RelayHealth, a McKesson subsidiary, which offered a vendor agnostic solution to solve these big data challenges. Ibrahim says that the organization had a history of using McKesson’s tools, so it was a natural fit.
Now, he adds, interactive analytics are presented in a visual and appealing way so stakeholders can “slice and dice it themselves,” which is more efficient than the traditional reporting approach of one request at a time every time the clinical team has a data need. This improvement helped get providers at Saint Francis Care more on board and engaged, which can be a common problem, Ibrahim notes. “I am a physician, and among physicians, there is a lack of trust around healthcare data. So I use my analytics and platform to engage the clinicians, by showing them a dashboard with metrics, I invite feedback, and I take it all in. I like to say that there are five stages of data: denial, anger, acceptance, working together, and then real transformation. I understand that; it’s natural. So I use this as a real tool for engagement,” he says.
Moving the Needle
At the end of the day, Ibrahim understands that analytics need to help clinicians make better decisions to care for their patients. In other words, no matter how great the assessments and process might seem, it’s all for naught if the data can’t be actionable at the point of care.
As such, Ibrahim points to Saint Francis Care’s nursing units, which each have standardized dashboards, which he says “speaks to how we are delivering care to our patient, and speaks to metrics around quality and safety.” He explains that on these dashboards, one can quickly see how many patients have been seen in the last month as well as an index of sickness, and how long on average the patients stayed in that unit. Regarding safety, one can see how many patients in that unit were injured, how many patients developed pressure ulcers while in unit, for example, and how many had an adverse event.
Then on quality, the dashboards track 30-day readmissions, which are drilled down by condition. And for patient experience, surveys are sent out to see how the organization stands in terms of delivering care. Patients are asked to give ratings on how courteous the staff was, if patients’ medications’ adverse effects were explained properly, and how response times were if patients needed something. These metrics were presented on a regular basis for every nursing unit, explains Ibrahim.
What’s more, as there is a large focus on patient complications and hospital-associated conditions, analytics are also provided to design initiatives that target improvement efforts, and also to track improvement. And according to Ibrahim, “That has been extremely successful, as we have been able to reduce those events significantly.” He also notes that for sepsis, a condition with a very high death rate and that has a large national focus, care teams designed interventions and leveraged tools in the EHR to support those plans and track progress. The result was a “dramatic improvement in reducing the death rate after the intervention,” he says.
Another area where the analytics have helped is with efficiency. Ibrahim says the following data is tracked: How long are patients waiting in the ER, and do bad things happen while they are waiting? How long does it take before they get moved to a bed? Is Saint Francis Care doing everything it can to keep them in the hospital only as long as they need to be? He says there have been great results here as well; patients have been pushed through the ED in a timelier manner with reduced lengths of time to get a bed, and hospitals stays have been shorter.
Moving forward, Ibrahim says he would like to start tracking metrics from the patient’s point of view. “So when patients come in for a stay, it’s not just about having them leave the hospital without an infection, but have them be able to attend a loved one’s wedding, or something that is meaningful to them. I would love to see more of these patient-monitored outcomes that are monitored on our side, but important to patients,” he says.
Ibrahim also adds that while he largely operates only in the hospital acute setting, he would like to see the value from analytics be gained across the whole continuum, meaning patients in their community, in their home, in the nursing home, and in home health facilities. “The hospital journey is only one piece of it. Let’s make what we have accomplished in the acute care setting possible across the care continuum,” he says.