Big Data Can Save Lives – We Just Need to Use It

Featured article by Sarah Landrum, freelance writer and blogger

Big data is something most of us associate with business servers and storage, but it is becoming a very important driver of almost every field. It doesn’t matter if you’re in business, manufacturing, research and development, or any field in between, big data is likely part of your strategy.

Big data is changing the face of how we do business, and the introduction of these data methods could do more than grow your company. It could potentially save lives.

There is so much medical data out there right now that it’s hard to imagine going through it all. The biggest reason is that most of the data is sitting sequestered in non-networked servers, not being used.

The capabilities exist — big data can be applied to benefit patients and people all over the world. But it’s not being used, at least not to its fullest potential.

By collecting this data, networking it and applying the concept of predictive analytics to it, researchers could potentially find lifesaving connections and answers to the questions that we’ve been trying to answer, as well as to those we didn’t even know we needed to ask yet.

We could make better decisions, build new tools, and save over $149 billion with improvements to efficiency.

Clinical Trials and Drug Development

The applications of big data could have tremendous effects on clinical trials and drug testing programs. A database of genetic information could be used to help researchers identify the best candidates based on their genetic markers or other related demographic information.

Trial matching, or choosing the best candidates for a human drug or treatment test, is an essential part of the drug development process. Researchers won’t get the results that they need by testing on the wrong subset of subjects. You wouldn’t want to test a new pediatric oncology treatment on a patient that doesn’t meet all of the eligibility characteristics. While computers allow for some pre-screening to be completed automatically, the majority of participant selection still has to be done by hand.

By utilizing big data and a comprehensive pool of potential trial subject data, professionals can both select their subject pool more accurately and make changes to that same subject pool quickly in response to new information — raising drug trial efficiency.

The more that’s known about an individual genetically, the more personal their healthcare can be. Information from cancer patients and even epidemics also contribute to understanding diseases and infections, and how and where to best treat them.

How Predictive Analytics Affects Diagnostics and Pricing

Predictive analytics — or the study of data points to find common information — could also be used to improve the accuracy of diagnoses, predict insurance product costs and create a better overall outcome for patients in facilities across the country. By collecting patient information into one easily accessible data pool, doctors can look for trends and quickly make informed decisions about individual patient care.

As the data pool grows, it may even be possible to use predictive analytics to predict who is at risk for certain diseases, such as diabetes or heart disease.

On the insurance side of things, this same data pool can help providers manage their bottom line by identifying trends, if not specific individual diagnoses. This gives the insurance industry a leg up, allowing them to easily assess the impact of illnesses across the country, or the world.

When it comes to healthcare, the data utilized in predictive analytics is like a fine wine in that it gets better with age. The longer a data pool is allowed to grow, the more insights you will be able to obtain from that pool. More information means more accurate and useful predictions.

The Implementation of Electronic Health Records

Gone are the days of patient files being stored in manila folders in the back room of the doctor’s office. Instead, many medical professionals have opted to use electronic health records (EHRs). EHRs have quickly become the most common implementation of big data in the medical industry.

As many as 94% of hospitals have currently adopted some form of electronic health records in their facilities. If a patient is traveling and encounters a medical emergency, their comprehensive medical history is just a couple of clicks away.

These EHR systems are good for more than just storing patient information. The SHARP (Strategic Healthcare IT Advanced Research Project) program, currently sponsored by the Mayo Clinic, is looking for new ways to utilize EHR data, including:

– Phenotyping: Phenotypes, or traits that can be physically observed, are unique to a person and his or her genetic code. EHR data can be used to identify individuals who posses one or more phenotype that might be important to a particular study or drug trial.

– Coordinated Care: A unified EHR system allows medical professionals across all fields to have access to the same information. If, for example, a patient needs to see a cardiologist who works in one facility and a pulmonologist who works in another, both professionals would have access to the same information. This makes it easier for both professionals to coordinate their care of the patient.

An Ounce of Prevention…

An ounce of prevention is worth a pound of cure, or so the old saying goes, and that holds especially true today when considering the world of big data and healthcare. Think of your colleagues or friends — at least one of them probably wears a Fitbit, Apple Watch or Samsung Gear. You may be wearing one yourself as you read this.

These handy little pieces of wearable tech allow you to track your activity, monitor your active heart rate, monitor the length and quality of your sleep and count how far you’ve walked, run or biked on any given day.

When you sync your device with your phone or computer, you can keep track of your progress. Sometime soon, though, you may be able to upload that data to a collective cloud that will add to the massive pool of health-related data. This anonymous data could be used to keep track of the general state of public health. Even if you are a healthy individual yourself, every contribution to a data pool like this makes the predictions more accurate and thus potentially more useful.

Making Medicine More Personal

While going to the doctor is always a personal experience, the medical industry as a whole can be anything but personal. Drugs or treatments that work for one patient, or even a hundred patients, inexplicably don’t work for patient number 101. Or one patient out of a thousand suffers a previously unheard of allergic reaction to a drug.

Many of these problems or complications are due to an individual’s genetic makeup. By utilizing big data and the massive amount of patient information that is already available, it is possible to make medicine a more personal experience.

If professionals encounter a patient that isn’t responding to an established treatment or any other variable, big data could potentially provide the tools to find the treatments that will succeed. By searching for patients with similar phenotypes, genomes or symptoms, doctors could find the best way to personalize a patient’s treatment.

Risks and Rewards

As with any networked information, there is always the risk of the system being compromised and information being either copied or lost. Security measures can help prevent this, and should be a part of the overall dialogue.

While there are risks involved in networking data, the potential rewards both in questions answered and potential lives saved outweigh the risks. If the medical industry navigates the horizon of big data well, the risks will seem trivial in retrospect.

Big data is slowly but surely changing the medical and pharmaceutical industries. As long as we are vigilant in the protection of these networks, there is no limit to the advances that the amazing men and women in medicine are capable of. We may have already found all the pieces to the puzzle, now we just need the computing power to put them together.