As the exponential growth and availability of Big Data progresses, new opportunities arise to enhance efficiency and productivity in various industries. Crucial information known as ‘Big Data’ is named this not only because of the indisputable volume but also because of the complexities, depth and relevance it provides.

Gathering and analyzing such data means developing more accurate analyses, more confident decision-making, greater efficiencies, cost reductions and reduced risks. Big Data is becoming a powerful tool that can positively impact our global community, especially when used in Healthcare.
“Big Data is being used to predict epidemics, cure disease, improve quality of life and avoid preventable deaths,” reports Barnard Marr of Forbes Magazine. Healthcare is doing an incredible job at gathering and analyzing Big Data, merging it with technology and finding answers to critical questions that impact world health.

Over the past decade, groundbreaking technology has provided Healthcare with invaluable information about patients. From the digitization of health records to smart-phones that double as activity trackers, water intake logs and diet planners, one cannot deny that patient data is becoming readily useable, searchable and available. However, while it is one thing to access the data, it is another to conduct analytics properly.

5 Things to keep in mind before conducting a retrospective database study:

• Make sure you have a thoughtful, well-defined research question, that is grounded in the literature and is clinically relevant or important;
• Consider which databases you will have access to, which you will need, and whether any linkages are required;
• Ensure you have used an appropriate case definition and have explicit eligibility criteria to identify the population of interest;
• Your analytical plan should be developed a priori (and this is often required as part of the Research Ethics Board approval process), and consult with an experienced statistician as needed;
• Finally, make sure that your analytical plan is focused on accessing clinically meaningful results, and that the results are presented and/or published in a way that is clinically relevant.

Adapted from:

• Smith AK, Ayanian JZ, Covinsky KE, Landon BE, McCarthy EP, Wee CC, et al. Conducting high-value secondary dataset analysis: an introductory guide and resources. J Gen Intern Med (2011) 26(8):920-929
• Motheral B, Brooks J, Clark MA, et al. A checklist for retrospective database studies – Report of the ISPOR Task Force on Retrospective Databases. Value Health (2003) 6:90-7

Relevant Reading:

• Smith AK, Ayanian JZ, Covinsky KE, Landon BE, McCarthy EP, Wee CC, et al. Conducting high-value secondary dataset analysis: an introductory guide and resources. J Gen Intern Med (2011) 26(8):920-929
• Motheral B, Brooks J, Clark MA, et al. A checklist for retrospective database studies – Report of the ISPOR Task Force on Retrospective Databases. Value Health (2003) 6:90-7
• Esposito D, ed. Reliability and Validity of Data Sources for Outcomes Research & Disease and Health Management Programs. Lawrenceville, NJ: ISPOR, 2013.