Population health – it seems everyone in healthcare is talking about it, but few have actually done much about it. Population health can certainly seem overwhelming, and as a result, many organizations just freeze and do nothing about it. But like many things in life, with population health, it’s important to work on the little things first in order to make progress on the big initiatives.
So what exactly is population health? Here is a definition from the Institute for Healthcare Improvement (IHI):
“Population health is defined as the health outcomes of a group of individuals, including the distribution of such outcomes within the group. These groups are often geographic populations such as nations or communities, but can also be other groups such as employees, ethnic groups, disabled persons, prisoners, or any other defined group.”
Improving population health is one of the three goals of the IHI’s Triple Aim, which seeks to:
- Improve the patient experience of care
- Improve population health
- Reduce the per capita cost of healthcare
The IHI believes that all three of these must be done in conjunction with each other to optimize health system performance.
If you are looking to embark on a population health initiative, here are 4 small ways to get started that can serve as the initial steps for a larger initiative.
Have the conversation about population health
It’s a simple first step, but also a required one. The first step to improved population health is to talk about it. Meet with your peers to discuss what information you want to gather, what information you want linked, what you want to do with this data, and more. Sure, there will be disagreements. But nothing will get done without first having that conversation.
Use BI technology to merge patient records
It makes sense that if a patient is in a hospital system, there should be one record and one record only on that patient. But in practical use? That’s often not reality. For example, one of our customers is a large hospital system with many locations and linked physician practices. Part of its challenge was to make sure that all of its physician and hospital data were linked together. Using data integration, this system was able to link its patients to hospital data using medical record numbers, social security numbers, insurance ID numbers and date of birth. They then validated this process against actual data to gain confidence in their approach. This helped ensure that each patient is only in the system once, allowing the hospital system not only to better aggregate data, but also to enable physicians to make better patient decisions since they now have a more complete picture at their fingertips.
Flag certain types of patients in your system
Part of population health is being able to identify problems earlier in your patient populations. One easier step to take is to flag those patients with certain types of diseases or conditions. What those conditions are will come from the conversations you have in step #1. You don’t want to overdo it, or else everything will be an exception.
For example, one of our hospital system customers is currently flagging all patients who have diabetes. That way, no matter where those patients are in the system, they will have a flag on file indicating that they are diabetic. This can enable providers to more quickly spot complications of the disease, and hopefully down the road, this hospital system will be able to correlate data between these patients to better spot trends.
Make sure patients make follow-up appointments
Readmissions rates are widely thought to go down when patients make their follow-up appointments and follow their physicians’ post-admissions advice. For example, one study found that patients who saw their doctors for follow-up within 30 days and were better armed with medication and care information from their providers were 30% less likely to be readmitted or use emergency room services during the 30 days post-discharge.¹
That’s why one of our hospital customers makes sure that each patient is seen by a physician within 5 days of discharge. This allows physicians to better determine if there are complications and allows patients to ask questions about medication regimens, etc. This hospital is also building into its system reminders for patients to be seen by their primary care physicians annually.
Another large hospital system identified patients at risk for readmission on its daily census by using several criteria that are known to increase the risk of readmission. These criteria include the number of scheduled oral medications (greater than 5 increases risk) and the number of inpatient admissions in a rolling 12-month period (anyone greater than or equal to 4 is highlighted). Lastly, the hospital system incorporated the LACE index which was typically used retrospectively to identify patients that could have been at risk.
By incorporating these criteria to the entire census, the case managers were able to identify those patients at risk of readmission and provide intervention prior to discharge. This hospital system also has a very large pastoral community with a list of various congregations that assist with “home visits” to those patients living within their zip codes. With the use of our software, this organization was able to identify those patients at risk and also identify if they lived within a zip code that provided “free” home visits. If a patient was at increased risk and lived in one of the zip codes, case management would notify the Pastoral Care Team so members could meet with the patient and offer the home assistance (at no cost to the patient). Members of the patients’ congregations would offer to go to the pharmacy, make meals, take them to doctors’ appointments or just visit them.
These 4 steps are just scratching the surface when it comes to population health. But progress requires movement, even if it’s just baby steps at a time. One of the biggest issues hospitals face is what population to start with. It is best to start with what you know and expand as you succeed in your approach. Many organizations that have “self-owned” health plans have started there. Other organizations have started with disease populations in the hospital patients – such as diabetics, heart failure, obesity, etc. Population health casts a very wide net – starting small and testing approaches and processes will certainly lead to more success than trying to “boil the ocean”.
Read more of Nora’s articles on issues that healthcare providers are facing here on the Dimensional Insight blog.
¹ “A Reengineered Hospital Discharge Program to Decrease Rehospitalization: A Randomized Trial,” Annals of Internal Medicine, Feb. 3, 2009, http://annals.org/aim/article-abstract/744252/reengineered-hospital-discharge-program-decrease-rehospitalization-randomized-trial.
- 4 Ways BI Can Help Prevent Harm in Healthcare - September 29, 2016
- 3 Ways Business Intelligence Can Help Control Diabetes - September 7, 2016
- How to Assemble a Winning Healthcare Analytics Team - May 24, 2016