I believe missing data is the common challenge we will face in handling the big data as prioritized in the paper. Even with a small application or tool designed to record day to day data, there are chances of not completing all the fields especially if it is not mandatory. That can be due to the data entry focal not being aware of the linkage between the original dataset and further analysis and useful information. Sometimes, a client chooses not to answer questions about his or her rights and some privacy concerns. To cope with this, it is learned that depending on the portion of the missing data, we can choose two kinds of solutions. If the missing data is low, we can substitute a relevant guess as Ching To Chung mentioned. If the missing data is at random, it is suggested to remove the entire data point so that there is no more bias for that data.
Sometimes, it is possible that the study dataset does not represent the population intended to be analyzed. This can lead to results that are not strong enough to validate. Choosing the sample randomly may be a way of reducing the biased sample and this in turn improves the result.
Data Analysis and Training
Handling big data is the biggest challenge we are facing today since it is no longer efficient with previous analysis tools. Advanced technology like machine learning can identify patterns and its prediction can help the big data in changing the valuable insights and outcomes. Together with the advanced tools, there is a need to train the relevant focal to promote their ability to analyze and use the big health data effectively.
Interpretation and Translational Applicability of Results
If we cannot interpret correctly the findings in real world setting, it’s a waste use of technology. We will surely need the help of service providers and professionals from health sectors to understand the practical use of findings and interpretations so that it is a more usable dataset.
Privacy and Ethical Issues
A thorough ethical review and informed consent from clients are a must to prevent any ethical related issues. Ethical guidelines and regulations to follow should be in place for any data collection and usage. We should also minimize the risk of identifying personal information by making the data anonymizing procedure.