How can Big Data be used to answer public health research questions?

Research output: Contribution to journalArticlepeer-review

Abstract

Background: The use of existing clinical/administrative Big Data (BD) when conducting research has become common. In the absence of a single comprehensive source of data, data linkage of multiple sources has proved to be of value in midwifery and public health research.
Aim: The purpose of this paper is to describe the process of accessing BD, linking sources and validating the outcomes to answer specific public health questions.
Method: This article describes and discusses the experience and challenges of using BD whilst undertaking a study to answer the research question of what is the prevalence and risk factors of congenital heart defect (CHD) in Northern Ireland (NI)?
Data from the Northern Ireland Maternity System (NIMATS) (2010-2014), Enhanced Prescribing Database (EPD) (2010-2014), and the HeartSuite Database (HSD) (2005-2014) were accessed, linked and analysed using Stata version 13. Ethical approval was obtained from the National Health Service (NHS) - Research Ethics Committee (REC) (No. 17/SC/0103, 27 February 2017).

Permission to access data was obtained from the Honest Broker Service (HBS) acting as a safe haven for the conduct of BD research. Training in the use of the HBS was undertaken to ensure confidentiality, anonymity and data protection. There was a fee to be paid to HBS to cover data preparation cost.
Findings:
Successfully linking routinely collected BD sets from multiple sources (HSD, NIMATS, EPD) facilitates the creation a robust and more complete data set from which conclusions can be drawn that may not previously have been possible. The experience of working in the HBS was a huge learning curve in acquiring an in-depth understanding of the data extraction process, data verification and data analysis.
Conclusions and implications: Data anonymization and controlling access to the data are important as they enabled ethical access of clinical/administrative BD. The quality of the data was variable in the databases but the linkage process enhanced the value of the outputs.
Research of this type has value for midwives, other healthcare practitioners and policy makers in public health settings as it provides factual data that can impact public health. However, as in all types of research, using and linking clinical/administrative BD for public health research has its strengths and limitations which should always be acknowledged.
Original languageEnglish
Pages (from-to)19-31
Number of pages13
JournalEvidence Based Midwifery
Volume19
Issue number1
Publication statusPublished - 1 Mar 2021

Keywords

  • big data
  • Honest Broker Service
  • clinical data
  • administrative data
  • public health research
  • HeartSuite database
  • Enhanced Prescribing Database
  • Northern Ireland Maternity System
  • evidence based midwifery

Fingerprint

Dive into the research topics of 'How can Big Data be used to answer public health research questions?'. Together they form a unique fingerprint.

Cite this