The Future Of Predictive Analytics in Healthcare

5 min readJan 10, 2023

Yatin Jaisingh (Lead Consultant, Thoughtworks)

What is Predictive Analytics?

• The process of learning from historical data and using advanced analytic techniques in order to make predictions about future outcomes & trends [1].

• A variety of techniques drive predictive analytics, it could be machine learning algorithms, data mining, Artificial intelligence etc.

• The advancements of intuitive tools, new predictive techniques and hybrid cloud deployment models has paved a way making it more accessible than ever before.

Predictive Analytics Process [2]:

Predictive Analytics in Healthcare:

In the past & present, the healthcare professionals’ focus has always been about reducing patient risk, anticipating and detecting early signs of patient deterioration. Advanced predictive analytics in healthcare paves a way to make better informed decisions which otherwise leaves clinicians with making the hard decisions without absolute certainty. It can alert caregivers and clinicians about the likelihood oh health complications even before they occur[3] . Such analytics of foreseeing the future events, may help shift the focus to prevent rather than cure the health issues when they occur.

The image above depicts that the prediction certainty is high when asked about things that occurred in the past like: What did I eat today? But the prediction certainty decreases when asked about what is going to happen in future like: Can I get a cardiac failure? It is not that the traditional healthcare practices are not about predictions. The only difference is those are driven by individual minds, experience & knowledge. The idea behind predictive analytics in healthcare is about going above and beyond individual experiences and widen the knowledge base and training data set in order to provide better care & treatment to the patients [4] . With a large amount of data, the analytic algorithms can be trained better to provide more accurate forecasts or predictions of what health issues may occur in future and can also recommend ways to prevent the likelihood of any such adverse event

Themes of Predictive Analytics [5]:

Understanding how the 3 themes of Predictive Analytics in Healthcare are a Gamechanger:

Benefits & Risks:

Predictive analytics is advancing at a very rapid pace but every advancement in technology has two sides of the coin- Benefits & Risks. But the amount of risk Vs the amount of benefit that any technology can bring in varies widely. So, it is of utmost importance that we balance the risks and benefits in a way that the benefits outweigh the negative consequences of any technological advancement. Any investment in new technological advancement has to be coupled with good governance. Though it may be argued that too much regulation could stifle or prevent the innovation but when it comes to healthcare, the argument may not hold very strong

Some of the use cases of Predictive Analytics in Healthcare [6]:

Predictive analytics can help to answer some of the very difficult questions in healthcare. For instance:

1. What is the best course of treatment for a given patient?

2. What would be the effect on a patient post giving any procedure? Is there a likelihood of any adverse effect?

Various researchers across the globe have leveraged predictive analytics and have successfully proven how it can be a game changer in the healthcare industry at various points in the patients’ treatment journey:

1. Diagnosis: As a prediction into malignant mesothelioma diagnosis in a patient cohort, researches have proven that using predictive analytics, patients can be diagnosed at an early stage. Thus, a timely treatment increases the chances of survival [7].

2. Prognosis: At the UCLA hospital, the researchers used predictive analytics to predict which patients were at a greatest readmission risk following a hospital stay using the physiological data from patients with congestive heart failure (CHF). Thus, this enabled physicians could intervene early and prevent the readmissions [8].

3. Treatment: In another instance, researchers used the ML based predictive analytics model to determine the treatment for patients with chronic pain [9].

A few more examples where predictive analytics in healthcare has come handy [10]:

Global trends of Analytics in Healthcare:

As a percentage of GDP, the expenditure incurred on healthcare is on an upward trend globally. Not only the developed countries like United States or the European countries, developing countries are also witnessing a rapid upward trend in healthcare expenditures. Amidst this upward trend, the major factors that instil confidence in the growth of predictive analytics in healthcare are the emergence of personalized and evidence-based medicine. Also, there is a growing need for more efficient healthcare sector and an increasing demand to bring down the healthcare expenditures by cutting down on unnecessary medical costs. Predictive analytics is proving instrumental in reducing hospital readmissions and reducing unnecessary diagnostic tests which are the major components of healthcare costs.

“In developed countries like the United States, health care spending was 17.8% of the GDP in 2015, increasing from 13% in 2000. In the majority of European countries, the healthcare expenditure share has already reached double-digit numbers and continues to increase[11] .”

Conclusion:

The future of Predictive analytics looks promising and gives an assurance to bring in better patient care, lower medical costs and early detections of diseases. But like any technological advancement this too has some adoption pathways and associated challenges [12].