How Using Big Data Improves the Public Health
Big data refers to complex, large data sets beyond the management capabilities of traditional hardware and software. While you might think your database of 500 results massive, big data means sets with thousand or tens of thousands of data. Once the purview of NASA, big data has moved into many industries and healthcare is one of the ones benefiting the most.
Big Data and Healthcare
In all aspects of health care, from insurance to hospitals to research facilities, there exists a requirement to improve quality while reducing costs. The analysis of big data helps accomplish this by providing a way to combine data sets to mine them for details and developments that prove beneficial to clinical decision making, disease surveillance, and population health management.
The data sets used aren’t limited to only survey results or hospital records. Researchers can assimilate data from numerous sources that provide insight into diseases, care methods, communications, even drug marketing. Researchers draw data from:
- administrative data,
- blogs,
- CPOE,
- electronic patient records (EPRs),
- emergency care data,
- insurance,
- laboratory,
- machine generated/sensor data,
- medical imaging,
- medical journal articles,
- news feeds,
- physician’s written notes and prescriptions,
- pharmacy,
- social media posts,
- web pages.
Big data analytics provide four “V’s” that keep it tops with researchers today: variety, velocity, veracity, and volume. You see the variety represented above in the numerous sources of data. The velocity stems from the rapid pace of real-time data collection. Most of it gets stored in the cloud. Its veracity refers to its data assurance – its credible and error-free nature. The volume of data grows at a remarkable rate since it gets stored and sometimes analyzed, as soon as it is created. Currently, researchers are applying big data analysis to a wide variety of areas ranging from autism to cancer to standardize knee joint-replacement surgery.
Help in Autism Research
Some of the biggest breakthroughs have occurred in autism research. Approaching it from genetics, big data has allowed researchers to analyze the genetics of 20,000 people with varying levels and types of autism. The results have mapped numerous genes related to the development of the disease.
The first break occurred in 2007, when Michael Wigler and partner on the research project, Jonathan Sebat, found that de novo mutations —mutations spontaneously arising rather than inherited — happen more often in the autistic than in typical people. The identified mutations appeared as a form of ‘copy number variants’ (CNVs). A CNV means a deletion or duplication of long DNA stretches. This occurs in other diseases, including cancer, Wigler’s prior area of research.
This discovery prompted more research by Wigler and two other research teams which led to the identification of a number of genes contributing to autism. Big data let them study the two percent of the human genome that encodes proteins, the exome. The analyzed only data of simplex families, a family including a single child diagnosed with autism also comprised of unaffected parents and siblings. Exome comparison of typical genomes related to the atypical (autistic) exomes exposed various de novo mutations. Using a data set of 600 families, the teams uncovered potential hundreds of genes, but six top candidates.
Follow up research in 2014 linked with high confidence 50 genes relating to autism. Using an even larger set of big data of 20,000 people and studying simplex families, the researchers identified 27 genes with rare de novomutations. That year, two separate studies identified the same ten genes. A more recent study found 28 candidates researchers refer to with “near certainty” as autism genes. Among the genes that appeared throughout each study after Wigler’s initial one are CHD8, DYRK1A, SCN2A.
Identifying genes helps move research forward. It provides opportunities for the development of gene therapies. Although there is a long trail to a cure, the big data sets have provided help in some ways already.
Finding Support Mechanisms
While waiting on a cure, genetic research has provided families with a member diagnosed with autism immediate help by helping them find support from others experiencing the same thing. Autism varies widely. There is no one type of autism. The experience of a child who tests low on the spectrum varies vastly from one who tests high. Big data hones the patient list with a specific diagnosis and enables patients and their families with the same diagnosis to form support groups.
It also helps non-profits that conduct autism research and education initiatives. They can combine big data with marketing initiatives combining the two using ClickFunnels techniques to leverage marketing insights and extend reach. Big data helps them hone the audience and better analyze marketing reach.
Big data helps researchers in numerous fields to conduct desk studies that can identify significant findings. In healthcare and public health research, its applications to a wide range of areas continue to drive developments in treatments for a plethora of diseases.