The use of health management programs to help people live healthier lives has increased rapidly during the last decade.
A holistic population health management approach takes into consideration the various determinants of health, including public health interventions, the adequacy of medical care, social and physical environments, individual behavior and genetics.
The need of the hour is for healthcare organizations to leverage analytics to categorize their patients on the basis of potential health risk or disease burden. This also enables these organizations to identify patients who will generate most of the health costs in the near future.
By trying to ensure that healthy people get preventive care and that people with moderate chronic diseases keep them in check, healthcare organizations can meaningfully manage their rosters, impact overall quality, and also reduce the total cost of care in doing so.
In this piece, we will be looking at how population health analytics can improve annual visits and ameliorate overall health outcomes. But first, let us get to understand:
Role of Data in Health Analytics?
Healthcare organizations, both small and big, have been implementing descriptive health analytics to medical data. Using queries; reporting tools and technologies, healthcare professionals normally collect data on the performance of different healthcare services.
This fresh idea is based chiefly on the accessibility and availability of data and information pooled through the integration and interoperability of an array of tools and systems such as electronic medical records, hospital information systems, clinical decision support systems, and other specialized medical equipment.
The main objectives of health analytics is to locate performance gaps and to propose best possible strategies as well as recommendations to fill these gaps.
Medical and health data may include information on health management program engagement, health risks that are collected via the use of survey-based health assessments, as well as health insurance claims and membership files.
Information on the physical or social environment comes from external data sources that describe characteristics of home life, neighborhood, and both the local supply as well as quality of healthcare at large.
Some of the most common means through which healthcare organizations can congregate data for population health analytics include:
Wearables store large troves of health information. These devices can be used to collect data on a user’s health including heart rate, calories burnt during the day, blood pressure, release of certain biochemicals, seizures, sleep tracking, physical strain, etc.
Today, there is a growing interest to use wearables not only for individual self-tracking, but also within corporate health and wellness programs. This can come across as an added advantage, especially when trying to study patterns within population health.
If a healthcare facility ties up with a corporate office, it can easily gain access to all the wearable data its employees generate.
- How Big Data Works with In-home Healthcare to Improve Outcomes
- This is how Technology is Transforming the Patient Care Landscape
- 5 Exceptional Use Cases for Predictive Analytics in Healthcare
Electronic Health Records (EHR)
The average patient will have multiple encounters with their primary care physician, nurses, specialists, lab technicians, and other clinicians. Each encounter will result in critical information, which, when considered as a whole, furnishes a complete picture of that patient’s overall health.
When all care providers have access to this same set of data, they can avoid prescribing contraindicated medications, ordering duplicate tests, and ameliorate health outcomes.
For this very reason, the data collected by EHR systems poses a lot of potential in health analytics.
Big data and Predictive Analytics
With advances in computer technology and software now enabling small-to-medium businesses to get involved with big data more easily than ever before, it only makes sense for healthcare organizations to consider the benefits of analyzing this information on a regular basis.
Big data and analytics is reshaping the wearable device market today. Demographic, usage, and consumer expectation data is rapidly pouring in from wearables. This data is being analyzed by analytic-based systems to get the customer what he or she wants.
There is a treasure trove of information being stockpiled by your organization that you can start using more intensively to support business and medical decisions.
By integrating big data and predictive analytics with your EHR system data, you can manage your medical organization more efficiently, which leads to better health care and boosts your bottom line.
Apart from the ones mentioned above, there are many other ways in which healthcare organizations can collect data for population health analytics.
Now moving to the most important part of this article:
How can Population Health Analytics Improve Annual Visits?
Population health analytics can help healthcare practices in a number of ways. For one, they can considerably improve annual visits and scheduling for any healthcare facility, small or big.
By utilizing the statistical information gathered by this model, practices can effectively group patients on the basis of ones that will be needing checkups at regular intervals (every 15 days or on a monthly basis) and ones that can go without routine checkups for a longer period of time (say, 3 months or more).
This grouping can be done on different levels depending on the different factors or patterns observed within the data collected from patients. For example, a patient who smokes can be called to the facility for routine checkups every few months to better track their health and mitigate the risk of cancer at an early stage.
On the other hand, a patient who is in good shape and whose family history doesn’t exhibit signs of any chronic illness can be called to the facility every three months or so.
Not only will this systematic approach help improve the healthcare facility’s annual visits, it will also leave more time with the doctors and staff to treat emergency cases without having to juggle tasks or suffering at the hands of a burnout.
This, in turn, will streamline care delivery and ameliorate patient outcomes as a whole.
Here’s a supporting case study to help you better comprehend the benefits that come alongside the adoption of this model:
Case Study: King Faisal Specialist Hospital and Research Center
In January 2015, the King Faisal Specialist Hospital and Research Center in Jeddah planned a performance improvement project by leveraging population health analytics in two phases.
The first phase was to perform a retrospective analysis of all the available ER data, which was conducted earlier that month. The study data was retrieved from the data warehouse system of the hospital including all data elements of all emergency encounters of the last year; 2014. A total of 26,948 encounters with valid data were retrieved.
Analytics techniques were used in the form of identifying and calculating different ER data variables and testing for any relationship between those variables and the admission status probability of the patient.
The second phase of the study started in mid-January. This phase included the implementation of two suggested recommendations; a Fast-Track for lower acuity level ER patients; dedicating 20% of the ER bed capacity, in addition to an added internal waiting area for those patients who can stay vertical instead of occupying an ER bed.
The main objective was to assign ER physicians only to patient cases with higher acuity levels; CTAS Levels 1–3, and in the same time to reduce the demand for other resources by less acute patients, without any change in the manpower, working hours or in the total number of the ER beds or capacity.
According to the findings of this study, eight principle variables could be identified for evaluation using health analytics:
1) Patient Gender ,
2) Age group,
3) Patient Acuity Level,
4) Patient mode of arrival,
5) Patient discharge destination,
6) Day of ER encounter, and
7) Session of ER encounter.
Three variables only had statistically significant influence on the admission rates of ER patients to the hospital inpatient departments and services; those were Patient Acuity Level, Patient Mode of Arrival and Patient Age Group.
Other variables did not have any significant effect on the rate of admission, where the most influential variable among these three was the Patient Acuity Level.
The acuity level of all ER patients during 2014 were analyzed and categorized, counting total patients visiting the ER in each acuity category and number of patients admitted from ER to inpatient departments and services in each category and the percentage of admission in each of those categories.
As the acuity level goes down; become less severe, the percentage of admission becomes less, which is logical.
What Lies Ahead
Population Health Analytics is a relatively new field, and the rules of the game are changing as we speak. This is an industry in flux, a work in progress within the healthcare sector.
At this stage in the development of health analytics, it’s not likely that one healthcare organization will do everything and do it well.
Healthcare providers should plan their strategy in such a way that they are using more than one system to cover the variety of tasks — for example, one system for data aggregation and risk stratification, and a different system for care management. They should plan to adjust as the rules develop and mature.
Most importantly, organizations should start by building a data foundation that is not only robust, but also comprehensive. If the underlying database is inconsistent, or even incomplete, it will be impossible to deliver valid analytics and drive better care for an entire population.