Qualitative Improvements in Medical Research on Common Man - Methods to reduce bias in research

 

Impact of medical research bias on a common man: a qualitative study exploring its methods of improvement

In the context of biomedical or medical research, bias is any parameter that leads to solutions that are different from the truth. It affects efficient scientific discovery and efforts to achieve improved benefits for the common man. Bias distorts study findings. Bias is universal in every sector. It can be caused by the errors in study design, while collecting data and while analyzing. It cannot be changed by sample size. The existence of medical research bias only refers to the validity of the study. Bias cannot be accounted for with statistics. 

Some of the most prominent biases are Selection bias, detection bias, observer bias, recall bias, response bias, publication bias, regression to mean, Hawthorne effect and treatment selection bias.

Measures to resolve bias in clinical research include mandatory registration of interests for all people and institutions who conduct public health research, all journals and funders to uptake of registered reports and publicly funded research should be pre-registered and should be published on the World Health Organization-affiliated research registry.

Biomedical Informatics (data informatics) and Artificial Intelligence (AI) can be the new age tools to resolve the problems of bias in medical research.

Artificial Intelligence is said to be the new nervous system of the healthcare domain. Application of AI and data informatics in medical research includes training databases for research data, m-health, medical research data exchange and clinical decision support system. With more health data availability, and the recent developments of efficient and improved machine learning algorithms, there is huge scope of interest for AI in medical research. In the 21^st century, AI in medical research has gradually evolved towards essentially data-driven approaches.

Some of the ways in which AI and data informatics can resolve the issue of bias in medical research are via implementation of data governance strategies, ethical standards for privacy and unbiased datasets. Each stakeholder has a distinct role to play in order to successfully implement this technique in medical research. Biomedical Scientists, investors, research & development bodies and government bodies are some of the stakeholders.

Regardless of all the benefits of AI and data informatics, common middle-income families still have reservation in sharing their data with the medical researchers and biomedical scientists. This can be resolved by releasing control of who gets access to study information and when. Some families might want more incentives to be offered. Researchers also need to convince patients about the benefits of data sharing.

This article provided a little awareness into the application of data informatics and AI in medical research, however extensive research needs to be done. Furthermore, analysis should be done to understand the related menace involving these new age technologies in the healthcare sector.

Article written by,

Mufaiz Ul Zaman - A student and researcher at Department of Biomedical Engineering , Government College of Engineering and Technology (GCET) , Jammu & Kashmir, India. 

Click here 👉to see his LinkedIn profile.