Remember the days when you had to use two different platforms for phone calls and for calendaring and task management? In today’s iPhone/Android laden world, such a separation seems heretical. Yet in healthcare’s hot new niche of population health management, we see that same heresy in the artificial, and suboptimal divide between care coordination/management software and analytic software.
Back around the turn of the century, personal digital assistants (PDAs) were widely used to organize our lives through calendaring, list maintenance, note taking, and other similarly handy functions. The PDA was, in some respects, the successor to the planner/datebook that had become indispensable to many people in that era. Devices that also allowed access to email and the Internet, like the iconic Blackberry, further made the PDA almost a necessity for the modern worker, if not the average citizen. The rise of the PDA coincided with the boom in the cell phone market, which untethered people from their landlines. Cell phones, which made telephony mobile, tried to incorporate lightweight versions of PDA-style functionality, with hard-to-use calendars and haltingly unusable email clients, but if you wanted to do serious organizational tasks, you had two separate devices for talking and management of your life. Early pioneers, like Blackberry and Handspring with its “Treo,” realized that it would be cool to combine the PDA and the cellphone to avoid carrying two devices. The category grew for the all-in-one device, and the “smartphone” was born, although often clumsily implemented until Apple came along with its revolutionary iPhone. The rest, as they say, is history.
The smart thing about the smartphone wasn’t just that it combined two devices into one, but that it recognized the synergies between telephony and PDA activities, and that combining them into one platform enabled a tremendously beneficial linking of the two. The new “whole” was greater than the sum of its parts. For example, instead of getting a reminder on your PDA to call Fred, then whipping out the phone to find Fred’s contact and then dial him, the smartphone could queue up Fred’s number in the reminder itself so you could directly place the call while acknowledging the reminder. The innovative linkages of these once-separate concepts since those early days have fundamentally changed both organization and communication.
Similarly for healthcare, many care coordination offerings sold by case management vendors, EHR vendors, and others are all but divorced from analytics, or bolt on reporting as an afterthought. Much like the cell phone vendors of old, they believe their main functions of documentation and tracking for individual patient care are also the main things needed for population care, so cursory treatment of analytics is justified. On the other hand, there are powerful stand-alone analytic packages that can turn mountains of data into insights on what is or isn’t working with care delivery, but have no direct way to translate those insights into something useful and specific that leverages the chosen platform for care coordination. But the “smart” platform in this context brings together powerful analytics and real care coordination tools and exploits their synergies to perform real population health management.
When analytics and care management are tightly coupled, they become mutually reinforcing in ways that make population health management come alive and go beyond documentation, checklists, and graphical trends. For example, let’s consider a scenario where a community-based care management program that has been in operation for some time has a collection of interventions that are routinely employed, including following up with patients after they have been discharged from the hospital. In addition to the interesting analytics that can be done, such as comparing the rate of positive versus negative outcomes for those in the program versus those who aren’t, looking at the relative severity of same, etc., a number of special care management actions can be done when this information is linked to the care coordination functionality employed to support care teams.
For example, the cohort of discharged patients that have not received follow-up (a gap in care) can receive a special “audit” intervention automatically inserted into their care plans so the care team is prompted to baseline the status of those patients and appropriate care can resume, based on the analytic results. Assigned care team members, like care managers or PCPs for those cases, can automatically receive additional reminder alerts in their work queues to prevent the emergence of those gaps in care in other cases, with chronic offenders triggering the same alerts to the care team’s supervisors as well. The identified cohort of patients can be monitored more highly, and subject to additional or alternative intervention activities, like special assessments, case conferencing that highlights the gap-in-care analysis, or automatic patient and care-team notifications that are designed to combat the predicted consequences of the identified gap in care. The interplay between analysis and care management itself is substantial, and benefits greatly by the bidirectional free-flow of information between the tools that facilitate analysis and those that facilitate care management.
There is a lot of talk in the industry these days about making information “actionable,” but this remains only rhetoric as long as analytic information cannot be readily transformed into the lingua franca for action, which is embodied in the care management tools used by the care teams for directly delivering care. Likewise, when care management tools are armed with indicators of care gaps, they can do a better job highlighting those patients during the care process, and feeding those extraordinary care activities to analytics appropriately tagged with metadata or other enhanced information to enrich further analysis.
Standalone analytic packages may make us smarter, and stand-alone care management systems may make us more coordinated, but when they are isolated and are not co-resident in a single cohesive technology infrastructure, neither necessarily makes us better.