Predictive Analytics for Personalized Healthcare

I am fascinated by the way predictive analytics has revolutionized so many industries. Be it retail (Amazon), Online Search(Google), Social Media (Twitter/Facebook). Whenever I use these companies’ products, I get a feeling that they know me very well. Based on my previous interactions they predict my future behaviour and more often than not they are spot on. They seem to know what I want which makes me feel I am not just another consumer but a unique person with particular tastes, likes and dislikes. And this truly delights me. On the other hand my experience with my physician is hardly personalized. Appointments after appointments the saga remains the same. Is it possible for healthcare industry to adopt the predictive analytics toolset to delight its consumers (patients)? Can they delight the patients the way other industries are doing?

Doctors have long been predicting outcomes based on historical information. But lack of holistic data and sophisticated tools for analysis made them rely on their judgement to deliver care. This made healthcare delivery as much art as it is science. With access to new sources of data and capabilities of analytics tools, the subjective component can be significantly reduced. What’s more – doctors can now, with the help of predictive analytics, deliver personalized care which was not possible with the traditional practices.

Simply put, personalized healthcare (PHC) is right care, at the right time, and at the right place. With the advancements in genetic sequencing and advent of big data analytics, PHC is now a reality. If predictive models based on behavioural and genetic data can tell how an individual will metabolize coffee, imagine their potential to predict the right treatment for a particular patient. Here’s a video that demonstrates how powerful predictive analytics enables personalized care-

Hippocrates once said “It is far more important to know what person the disease has than to know what disease the person has”. Cannot agree more in the context of personalized healthcare. And if personalized healthcare is the end, predictive analytics is certainly one of the means to the end.

Advertisements

Healthcare Analytics to Check Skyrocketing Costs

Annual healthcare spending in the US is estimated to be around $3 Trillion, nearly 18% of the country’s GDP.  According to a report by the Institute of Medicine, the country lost $750 billion due to inefficiencies in the system. The report states that the primary areas of wasteful expenditure are unnecessary services, improper care delivery, excess administrative costs, inflated prices, prevention failures and fraud. While there are multiple measures being taken to control these inefficiencies, healthcare analytics is certainly one of the cogs in the wheel.

Healthcare industry so far has not utilized analytics the way other data heavy industries such as retail and banking have. This could partly be due to legacy systems, unavailability of data, data privacy issues and lack of incentives. Given that the data generated and collected through the introduction of electronic health records, health insurance exchanges, and social media portals is on a rise, the healthcare analytics sector is ripe for change. Moreover, the move from pay-for-service model to pay-for-performance, focus on wellness to prevention and universal coverage clearly makes a strong case to implement analytics in the healthcare space.

Some of the use cases where healthcare analytics can deliver value are as follows:

Payers:

  • Fraud Analysis: Large amounts of claims data can be analysed to using fraud detection models. Predictive models can detect suspicious claims by providers which can then be further investigated.
  • Incentive Design Analytics: Customer records can be analysed to predict health issues before hand. The payers then can promote preventive measures by linking premiums to the use of such preventive measures

Providers:

  • Evidence-Based Medicine: Doctors can use EMR, EHR, financial, operational, clinical, and genomic data. This will eventually decrease the cost by reducing readmission rates, wrong diagnoses and efficient care.

Pharma Companies:

  • Research & Development: R&D can benefit from analysing clinical data which can provide insights to accurately design trials which in turn would provide data points to efficiently design clinical trials, leaner R&D pipeline, improve time to market drugs hence reducing costs in the drug discovery process.

Although big data offers a very promising proposition to tackle the rising healthcare costs, there are certain challenges that need to be addressed.  One of the aspects that need special attention is patient data security and privacy. With a large number of regulations in place, data analytics applications would also be required to comply with the rules of the game. Additionally, all the stakeholders in the healthcare space will have to foster an open culture of trust to allow data sharing & data-integration to reap the full potential of analytics solutions.