The field of medicine has always gyrated around probabilities. Whether a doctor is deciding what treatment path or drug to prescribe or diagnosing a patient, the decision-making process is always constructed around the most logical evidence available.
In the digital era, however, there happens to be a new doctor around: predictive analytics. The approach leverages various technologies to convert historical data into predictions.
According to a recent survey, close to 93% of healthcare providers and payers believe that predictive analytics is significant to the future of their business. Moreover, approximately 89% said they plan to use predictive analytics within the next 5 years.
Organizations are increasingly shifting toward predictive analytics to address problems regarding patient costs, satisfaction, readmissions, risks, diagnosis, profitability, clinical outcomes and mortality.
In this piece, we will look at five powerful use cases for how healthcare organizations are harnessing predictive capabilities to extract forward-looking, actionable insights from their growing data assets. But first, let us get to understand what predictive analytics truly is.
What Is Predictive Analytics?
Predictive analytics can be defined as a branch of advanced analytics that is put to use in the derivation of predictions about unknown future activities or events that lead to decisions.
It is a discipline that utilises varied techniques such as modelling, data mining, and statistics, as well as artificial intelligence (AI) to assess real-time and historical data; and makes predictions about future events. These predictions offer a distinctive opportunity to leap into the future and identify upcoming trends in patient care both at an individual level and at a macro scale.
Predictive analytics is based on the science that is outlined from theories developed by humans to fit a hypothesis (supervised learning). A set of processes and rules are evolved into a formula that undertakes calculations and is known as an algorithm.
Predictive analytics can also be based on unsupervised learning which does not have a guiding hypothesis and uses an algorithm to look for structure and patterns in data. It then clusters them into insights or groups. In unsupervised learning, the machine may not be aware of what it’s looking for but as it processes the data it begins to recognize complex patterns that a human may never have identified otherwise. This technique can add considerable value to researchers looking for something out of the ordinary.
Both these forms of predictive modelling are sound analytical tools to use in a well-rounded application of these technologies.
The health care sector, with its countless stakeholders, stands to be a key beneficiary of predictive analytics, with the advanced technology being recognised as a fundamental part of health care service delivery.
1. Bolstering patient engagement and satisfaction
Consumer relationship management has gone down to become a significant skill for both insurance companies and healthcare providers aiming to encourage holistic wellness and lower long-term spending – and predicting patient behaviors is a key component of developing effective adherence and communication procedures.
“We need to know what works and what doesn’t in our engagement programs, and how to anticipate and predict the best outcomes given very complex characteristics of our membership sub-populations, which span every single segment of the US population,” said Patrick McIntyre, Senior Vice President of Healthcare Analytics at Anthem.
“Our goal is to grow our consumer engagement skills, because we are shifting into much more of a service-oriented, consumer-oriented industry.”
Anthem is making use of its data analytics tools to create customer personas that enable the payer to improve customer retention, send tailored messages, and find out what approach has a greater probability of being impactful for each individual.
Healthcare providers are also using behavioral patterns to produce meaningful care plans and keep patients engaged with their clinical and financial responsibilities.
“Both payers and providers have a wealth of information that they can use to build models. Healthcare providers can also acquire some other sources, like the social determinants of health, for example, that will really help the strength and accuracy of their models,” said Lillian Dittrick, Fellow of the Society of Actuaries.
“When we use predictive models to look at all the variables, it helps us prioritize those patients who are really going to be receptive to changing something in their lifestyle, such as nutrition or exercise.”
Using predictive analytics to develop stronger, more dedicated relationships between patients and providers and to make intelligent care management decisions can improve long-term patient engagement and lower the risk factors associated with chronic diseases.
“We’re seeing more and more that automation and machine learning tools really help with sorting through and processing these very large amounts of data,” said Dittrick. “There is some kind of predictive modeling that could help improve processes in just about any facet of healthcare.”
2. Getting ahead of patient deterioration
While still in the hospital, patients face numerous potential threats to their wellbeing, including the acquisition of a hard-to-treat infection, the development of sepsis, or a sudden decline in health due to their existing clinical conditions.
Data analytics can help healthcare providers react at the earliest to sudden swerves in a patient’s vitals, and may be able to recognize an upcoming collapse before symptoms clearly manifest themselves to the plain sight.
Machine learning strategies are certainly well suited to predicting clinical events in the hospital, such as the development of sepsis or an acute kidney injury (AKI).
At the University of Pennsylvania, a predictive analytics tool leveraging EHR data and machine learning was developed that aided to identify patients on track for septic shock or severe sepsis 12 hours prior to the onset of the condition, a 2017 study explained.
“We have developed and validated the first machine-learning algorithm to predict severe sepsis and septic shock in a large academic multi-hospital healthcare system,” said lead author Heather Giannini, MD, of the Hospital of the University of Pennsylvania. “This is a breakthrough in the use of machine learning technology, and could change the paradigm in early intervention in sepsis.”
Another initiative at Huntsville Hospital in Alabama established that integrating clinical decision support (CDS) tools and predictive analytics could lower the rate of sepsis mortality by more than half. The analytics-driven strategy surpassed the precision of existing gold-standard tools.
3. Preventing patient self-harm and suicides
Early identification of individuals expected to cause harm to themselves can make certain that these patients receive the mental healthcare they are in need of to avoid serious events, including suicide.
In a 2018 study conducted by KP in association with the Mental Health Research Network, the combination of a standard depression questionnaire and EHR data precisely identified individuals who were at a higher risk of a suicide attempt.
Using a predictive algorithm, the team found that suicide attempts and achieving success in commiting them were 200 times more likely among the top 1% of patients flagged.
The most accurate predictors of a self-harm attempt included the use of psychiatric medications, previous suicide attempts, mental health or substance abuse diagnoses, and high scores on the depression questionnaire taken up by each one of these patients.
“We demonstrated that we can use electronic health record data in combination with other tools to accurately identify people at high risk for suicide attempt or suicide death,” said first author Gregory E. Simon, MD, MPH, senior investigator at Kaiser Permanente Washington Health Research Institute.
The growing public health crisis and its impact on the veteran population is also a matter of dire concern in the US. In September of 2019, the Veteran Affairs Department released the 2019 National Veteran Suicide Prevention Annual Report, which suggests that of the 45,390 American adults who died from suicide in 2017, 6,139 were U.S. veterans. Results also indicate that more than 6,000 veterans died by suicide in each year between 2008 to 2017 and that, in the most recent year, the suicide rate for veterans was 1.5 times the rate for nonveteran adults.
Using a predictive model that’s continuously updated, the US Department of Veterans Affairs developed REACH VET—or Recovery Engagement and Coordination for Health-Veterans Enhanced Treatment.
This program analyzes data within veterans’ health records to identify patients “at a statistically elevated risk for suicide, hospitalization, illness or other adverse outcomes”- Deputy Director for Innovation and Program Development, Aaron Eagan, told Nextgov. If an individual is flagged in the system, their health care provider is notified, who then decides to offer pre-emptive care and support, in certain cases even before a veteran begins to feel suicidal urges.
4. Developing new therapies and precision medicine
As genomics and precision medicine gain momentum, researchers and providers are turning to analytics to supplement drug discovery techniques and traditional clinical trials. “In silico” testing is a hopeful way to reduce the need to recruit patients for compound and expensive clinical trials while speedening the assessment of new therapies.
“FDA’s Center for Drug Evaluation and Research (CDER) is currently using modeling and simulation to predict clinical outcomes, inform clinical trial designs, support evidence of effectiveness, optimize dosing, predict product safety, and evaluate potential adverse event mechanisms,” said FDA Commissioner Scott Gottlieb, MD, after the passage of the 21st Century Cures Act.
“To take just one example of the benefits of these approaches, as we enter an era of drug individualization, modeling and simulation that incorporates aspects of individual physiology and genetics in drug metabolizing enzymes is being used to identify patient subgroups that need dose adjustments.”
“In silico models are being put to use in creating control groups for trials related to degenerative conditions such as Huntington’s disease, Parkinson’s disease, and Alzheimer’s”, the FDA added.
“We’re at the beginning of a transformative era in science and medical technology,” Gottlieb added further.
Predictive analytics is playing a vital role in turning new drugs into precision therapies. CDS systems are beginning to foretell a patient’s response to a predetermined course of treatment by matching the results from previous patient cohorts with genetic information, enabling providers to select the therapy with the greatest chance of achieving success.
Doing so may allow researchers to better comprehend the relationships between the effectiveness of particular therapies and genetic variants; and eventually improve outcomes.
5. Ensuring powerful data security
Predictive analytics and artificial intelligence are also expected to play a major role in cybersecurity, particularly as the sophistication of attacks is on the rise.
Using analytics tools to assess patterns in data sharing, access, and utilization can display an early warning for organizations when something is unusual – especially when those changes indicate an intruder may have pierced through the network.
“Predictive tools and machine learning techniques can calculate real-time risk scores for specific transactions or requests and respond differently depending on how the event has been scored”, explained David McNeely, a Fellow at the Institute for Critical Infrastructure Technology (ICIT).
“Once the risk score has been determined in real-time, the system can use this during a login event to either grant the access for a low-risk event or to challenge for Multi Factor Authentication (MFA) or possibly block the access for high-risk events,” he said.
“In this way, the system enables IT to apply MFA more liberally across infrastructure and applications since the machine learning system will make decisions of risk which determine if MFA will actually be applied or not.”
“This strategy could be particularly effective for preventing ransomware from affecting a healthcare organization”, added ICIT Senior Fellow James Scott.
“Early adoption of sophisticated algorithmic defenses such as machine learning or artificial intelligence solutions will transform healthcare cyber defenses beyond the capabilities of average attackers.”
Predictive analytics is most useful when both the predictor and the care intervention are integrated with the same systems. It is easier to identify trends and obtain best results. Data can be compared to analyse the outcomes for all the diseases that a healthcare provider could encounter. Physicians and insurance companies rely on predictive analytics for extensive research and statistical analysis that can range from improvement of patient’s health, post medication, to their readmission rates.
Predictive analytics is playing an important role in transforming the world today and all sectors are reaping benefits from what it has to offer. The health care sector has also profited remarkably from predictive analytics, and it can be asserted that this technology is the central aspect of the future of health care delivery and medicine in general.
Billions of people around the world await the benefits of its adoption, with patients able to rejoice an ameliorated service delivery that both- predicts future challenges and addresses them proactively.
Diagnosis would be more precise, as would the care that follows. Providers would also benefit, given how easy it would be to retrieve useful information and take necessary steps toward seeing the health of their patients get better.
The benefits associated with intelligently designed and implemented predictive analytics in the health care sector are unfathomable.