The changing landscape of healthcare is creating a huge demand for health data analytics for Population Health Management (PHM). PHM involves using healthcare analytics to help establish best practices related to the treatment, prevention and self-care of patients throughout a set population. As medical recordkeeping moves from paper to digital, the ability to share information critical to patient care and overall health outcomes is broadening.
So, too, is the ability for healthcare professionals to realize the goals of PHM courtesy of more readily available access to big data related to patients, populations and their care. The hope is that someday soon all healthcare organizations will have access to aggregated data for analysis that can improve patient outcomes while lowering costs across populations and subpopulations of people.
Understanding Population Health Management
PHM isn’t a new concept, but it’s one that’s gaining more ground as electronic health records (EHR) make mining data to improve patient care more feasible. Analytics may also be used to reduce healthcare costs while also cutting fraud, abuse and waste in the $2.7 trillion healthcare industry in an effort to maximize investments and improve results.
PHM has a few distinct goals that are achieved through the use of data mining. The overall goal is to use analytics to determine what works to improve patient outcomes and/or reduce costs while identifying what changes need to be made. The benefits of the in-depth analysis associated with PHM are valid for not only patients, but also providers, policymakers, employers and insurance companies.
Though effective PHM has long been a goal, realization has been hampered by an inability to mine all data related to populations and population subsets. The practice has been limited to public health clinicians simply because of the limiting factors associated with healthcare information sharing prior to the implementation of EHRs. For example, though a hospital might have a large data pool of information about diabetics in a certain geographical area, the picture of their medical status isn’t fully complete without information also gleaned from local urgent care centers, private medical practices, pharmacies and other medical stakeholders. EHRs are making the collaboration and information sharing required by PHM more feasible.
On the individual practice level, Practice-Based Population Health is feasible. This branch of PHM allows analytics to be used to help individual practices:
- Identify patient populations that might require targeted care
- Examine characteristics of the subpopulations
- Create reminders for patients that are actionable
- Track performance
- Make data more accessible for improved analysis
Where Big Data Comes into Play
Big data involves the implementation of EHRs and other forms of scalable, streaming data that can be used to help with real-time decision-making related to patient care and PHM. As more medical practices move to EHRs, the ability to mine data outside of a hospital’s boundaries, for example, is becoming feasible. That means analysis of more complete, integrated data is also becoming possible.
Using a diabetic population in a particular region as an example, simply tracking compliance levels with regular A1C testing used to be quite problematic. While a hospital might have some testing records, other patients’ records might have been held in a silo within a medical practice or testing facility. EHRs are making it possible to glean a better overall picture of patient status while also enabling real-time, point-of-care information that prevents duplication of services and promotes improved quality of care.
The ultimate goal is to be able to use data across patient populations to improve care and lower costs, but widespread use of big data is not yet available. As more practices move onto EHRs, however, the ability to aggregate data, filter it and analyze it is becoming a real possibility that could have a very positive impact on patient outcomes as PHM becomes more attainable.