Hidden and not so hidden bias in nutrition research

Groups related to this event

Centre for Health Informatics

Event date

Wednesday, 25 May 2016

Abstract

Public health guidelines are increasingly required to be based on rigorous evidence. If underlying studies are not sound, guidelines cannot be evidence-based or credible. The methods used in clinical practice guidelines for conducting systematic reviews, assessing bias in included studies, and rating recommendations must be adapted for the type of evidence relevant to public health guidelines.

Our group is conducting a series of studies aimed at strengthening the evidence base of public health recommendations relevant to nutrition by identifying and remediating research bias. I will describe how the methods we have used to study “bias”—the systematic error or deviation from the true results or inferences of a study—in tobacco, pharmaceutical and chemical research are being applied to nutrition research. Bias can be introduced throughout the entire research process (questions asked, design, conduct or publication). I also describe challenges to empirically demonstrating that corporate research sponsorship and investigator conflicts of interest introduce biases that influence research results.

Our work will contribute to policies to make data and funding more transparent, manage conflicts of interest and improve systematic review methods. The findings should also improve nutrition research and guideline development, and will be applicable to other public health areas where methods for guideline development are evolving.

Speaker profile

Professor Lisa Bero is a pharmacologist and researcher in evidence-based health care who is internationally renowned for her studies on the integrity of clinical and basic research evidence that is used to influence health policy, and the manner in which evidence is communicated to key groups such as physicians, policy-makers, journalists and the community. She leads the Bias and Research Integrity Node at the Charles Perkins Centre. Prior to joining the University of Sydney in 2014, she was Professor at the Department of Clinical Pharmacy and Institute for Health Policy Studies, University of California San Francisco (UCSF).

Professor Bero is recognised for her methodological studies on bias (including publication/reporting, design and funding biases) in the fields of clinical medicine (pharmaceuticals), tobacco control and environmental research, and on the use and implications of the evidence for prescribing decisions/policy. Her expertise lies in investigating hidden biases in the design, conduct and publication of research, and includes ground-breaking work that demonstrated the selective reporting of data for drugs approved by the FDA (PLoS Medicine 2008) and a paper (BMJ 2012) which showed that including unpublished outcomes of drug studies in meta-analyses changes the results of all relevant meta-analyses. At the University of Sydney, she directs a multidisciplinary team analysing research quality (initially in nutrition research).

Seminar details

Date: Wednesday 25 May 2016

Time: 12-1pm

Venue: Seminar Room, Level 1, 75 Talavera Road, Macquarie University

Chairperson: Dr Adam Dunn

To register for this seminar please click here