BD:Blog:big-data-makes-big-play-use-of-predictive-analytics-in-health-care
BD:Blog
BD:Health
Budget Direct

Big Data Makes Big Play: Use of Predictive Analytics in Health Care

Looking for smarter
Health Insurance?

Get a Quote

Big Data Makes Big Play: Use of Predictive Analytics in Health Care

Just what is “Big Data” anyway?

In modern English we use the term to refer to a collection of observations, measurements or facts. Scientists generate a lot of data as they do their research, as do businesses and governments. All of us generate significant amounts of data as we go about our daily lives.

All these bits of data can be collected and compiled and built upon with more data over time. This brings us to the term big data, which refers to high-volume, high-complexity and high-speed data.

We hear so much about big data these days. As our computing technology advances at a rapid pace, we have reached a point where we can keep vast amounts of data in large data sets. In other words, the computing technology has become powerful and fast enough to handle a large volume.

Big Data Means Big Possibilities

With every clinical visit a patient leaves behind a small data footprint. Right now, the majority of these records are stored away in silos — big and small — in clinics and hospitals. However, we now have powerful technologies that allow gathering it all together to create large health-care-related data sets.

The current health-care sector is highly fragmented, inefficient and expensive. It is designed to deal with people in crisis and is based on short-term thinking.

Big data can help us harmonise and improve the health care decision-support systems to provide better patient outcomes, increase efficiency and decrease costs.

BigData3

Predictive analysis is an approach we can use to help us understand the patterns of how we use health care. It allows us to develop mathematical models that can test new designs of health care delivery — both on a local and global scale. Mathematical modelling of this kind can reduce the financial, social and personal burden associated with the current trial-and-error approach to health-care management.

Predictive analysis is an approach we can use to help us understand the patterns of how we use health care.

If we can integrate the clinical, laboratory, lifestyle, behavioural, environmental and other data points about patients we can generate predictive analytics, which help target interventions to the right patients.

Sharing personal health data can be a divided issue. But 60 percent of Budget Direct survey participants said they’d happily share their info to contribute to greater discoveries later, so long as specifics remained anonymous.

Hospitals can use big data technology to monitor the specific patient-related characteristics likely to cause future hospital re-admissions.

Targeted interventions can then be developed to address the needs of those patients to improve patient outcomes. This trickles down to reduce the need for re-admissions and keep costs down.

There is a particularly exciting example of real-time clinical collaborations made possible through the use of big data technologies. Hospitals can conduct continuous real-time clinical monitoring of critically ill patients while sharing the data with external experts and institutions. This could maximise the use of available specialist resources and improve patient outcomes.

Big Data on a Small Scale

The human body is made of chemical molecules — simple and complex. The biochemical interactions between the human body’s molecules and the pharmaceutical drug molecules are the basis of drug therapy.

BigData6

We are very lucky to live in the age of “omics” technology — the field of research capable of analysing the human body to the finest resolution. Omics technology allows us to build detailed biomedical maps of the human body to further understand health and disease.

By matching the drug data to a patient’s personal biomedical data, such as genetic and metabolic profile, we can predict how a particular person will react to a particular drug.

With omics technology, we can get a glance at the key potential molecular targets for the development of new medicines. We can then overlay this knowledge with the data we know about various chemicals. This allows us to identify many potential candidate drug molecules in a very short time and cut the time and money required to develop new medicines.

With the drug discovery process typically taking upwards of 15 years and several billion of dollars, this new technology can help save lives and dollars.

Omics technology leads us to the development of personalised medicine. By matching the drug data to a patient’s personal biomedical data, such as genetic and metabolic profile, we can predict how a particular person will react to a particular drug. We can then find the best fit for the person’s personal therapy.

Read our article called Can We Live Forever? Medical Science Says ‘Perhaps’ that talks about some of the advancements in medical science already.

Big Barriers to Big Data

There are a few hurdles we need to clear before we can truly enjoy the benefits that big data can bring.

The data coming from different sources needs to be standardised. As the data comes in all sorts of formats, from doctors’ notes to digital images; from health insurance and Medicare claims to hospital admissions data, it needs to be converted so that it can be analysed together.

Converting it all into analysable formats is no easy task, but the technologies capable of making it happen, such as Semantic Web, Data Ontologies and Data Linkage, are being developed.

Data doesn’t tell a story, but we can create storylines using the data if we use the right analytical tools.

Data doesn’t tell a story, but we can create storylines using the data if we use the right analytical tools. Big data needs to be analysed in new ways. The accepted scientific approach to data analysis, based on hypothesis testing, is cumbersome and unproductive if applied to large data sets.

The new big data predictive analytics technology uses an exploratory approach called data mining. This includes novel techniques like inductive reasoning, topic modelling, machine learning and graph analytics. This is a major paradigm shift for health and medical sciences, and embracing it will certainly take an adjustment.

BigData5

The biggest challenge: changing the administrator’s and policy maker’s mindset. There are needs for greater data sharing with appropriate privacy protection and security arrangements.

Those in charge of making these large-scale decisions should be open to the progressive ways of thinking rooted in the new knowledge-based technologies.

Big Hopes for Big Data

Albert Einstein said, “We cannot solve the problems with the same thinking we used to create them.” The big data revolution is giving us a chance to look at the old problems with new eyes. It is giving us a chance to find better solutions.

Deciding with data. How data driven innovation is fuelling Australia’s economic growth. www.digitalinnovation.pwc.cam.au

Wood WA, et al. (in press) Emerging uses of patient generated health data in clinical research, Molecular Oncology. 2014. http://dx.doi.org/10.1016/j.molonc.2014.08.006

Using big data to transform care. Health Affairs. July 2014;33(7). http://content.healthaffairs.org/content/33/7

IBM. Big Data at the speed of business. http://www-01.ibm.com/software/data/bigdata/industry-healthcare.html

First Data Bank. http://www.fdbhealth.com/

Biome Podcast: Personalised Medicine. http://biome.biomedcentral.com/podcast-transcript-personalized-medicine/

Merelli I, et al. Managing, analysing, and integrating big data in medical bioinformatics: open problems and future perspectives. BioMed Research International. 2014. Article ID 134023 http://dx.doi.org/10.1155/2014/134023

Survey Stats: Survey was conducted by Budget Direct in the month of April 2015 with a random selection of 1,000 people.

Kateryna Babina

Kateryna Babina

Dr. Kateryna Babina is a medical scientist with more than 15 years of experience in clinical, research, higher education and regulatory settings. Kateryna has a medical degree and a PhD. Kateryna has worked in the fields of public health medicine, epidemiology, human toxicology, biomarker development, human health risk assessment and clinical microbiology. Kateryna’s business, Scripta Medical Communications, is based in Adelaide, South Australia.