CONTENTS:

Speakers have kindly provided responses to questions submitted by conference participants during the Discussion session that did not get an opportunity to be discussed.

 

1. Does the panel think that the “slowness” of behavioural research is due to the nature of our science, or is it a more culturally ingrained issue? For example, we tend to get sticky about the fine grain details of funding applications and lose sight of the bigger picture issues.

Some of the slowness is the nature of behavioral research -e.g., mostly human subjects research, interventions that last weeks, etc. – but some is more related to our scientific culture – e.g., why does it take so long for the NIH to review and approve funding for grants?

2. To what extent do you think the COVID-19 pandemic will prompt funding agencies to ‘streamline’ their review processes? Does the NIH have any specific plans to move to a new, more rapid/streamlined process and how can you help encourage international agencies to do the same (e.g. CIHR in Canada).

The NIH also streamlined processes during ARRA (2008 economic stimulus) but mostly returned to standard procedures afterwards.  We’re hopeful that we can maintain some of the streamlined procedures following the pandemic.

3. Technically speaking, are there ways by which we can remove the effects of subjectivity on self-report measures (mainly physical activity & dietary indices)? As they are basically originating from a self-report and recall nature…

Reducing the retrospective nature (e.g., using EMA) at least reduces the recall biases.  And more modern psychometric approaches (e.g., IRT) can improve precision and accuracy.  Within clinical trials that are not blinded, useful to test for the perception of equipoise among participants and to ask for self-report at times when the intervention assigned is not so salient – e.g., during the same visit in which they received the intervention.

4. Has any research been done to ask funded researchers why they have not published in a timely fashion at all? Are there any common factors?

Very little.  We know some risk factors that tend to result in longer lags to publication – e.g., human research, clinical trials, longitudinal studies, but those explain only some of the variance.

5. Should we be endeavouring to do more collaborative, large-scale research including harmonizing data collection methods and measures? How do you overcome challenges related to cross-cultural applicability (and non-applicability) for many measures that were developed in the West (often the US) and which may lack face validity in other contexts?

Yes – we conduct too many underpowered studies that inadequately represent the breadth of potential patients.  Important to harmonize, but excellent point about different cultural contexts.  Developing IRT item banks and co-calibrating items across groups and cultures is one way to address these measurement differences.

6. Where much funding comes for applications of behavioural medicine, how do we ensure that the fundamental scientific questions about human behaviour are asked and answered?

NIH funds considerable basic behavioral and social sciences research.  Actually, about half of NIH funding for BSSR is basic research.  OBSSR is focused primarily on the translation of basic to applied and back.

7. Do you have any thoughts to share on accounting for the differences between health systems from country to country within international behavioural trials?

Clearly an important contextual factor to consider when it is reasonable to assume that the health systems impact the behavior or outcome of interest.  However, it would take a quite large trans-country design to have enough of the common types of health systems to assess this effect with any degree of certainty.  It may be useful to at least consider assessing the mechanisms by which these different health systems have their effect (e.g., access to specific care).