The ExpertLens– Exploring different pathways for different contexts

Original content from the Mapping Pathways blog team

Caroline Viola Fry

The ongoing International AIDS Conference (AIDS 2012) in Washington D.C. is an opportunity for Mapping Pathways team members to disseminate information and findings on the project. 

In this two-part series, we take an in-depth look at the ExpertLens, one of the cornerstones of the Mapping Pathways project, with colleagues Ohid Yaqub and Caroline Viola Fry from RAND Europe.

Ohid Yaqub

MP: Please introduce yourselves and tell us a bit about what you do. 

CVF: I work as an associate analyst at RAND Europe, a non-profit organisation that works to improve public policy and decision-making through objective research and analysis. I come with a global health background and work in the Innovation and Technology policy team, which explores the way governments or institutions support translation of research into new ideas or technologies that society can make use of.

OY: I’m an analyst with RAND Europe and work in the same team with Caroline. My motivation and interest in this project came through my Ph.D. in vaccine innovation. One of the cases I looked at was the vaccine development effort for HIV. The Mapping Pathways project interested me hugely because ARV drugs, which were previously thought of only as therapeutic options, were starting to take on properties and policy issues very similar to vaccines i.e. as preventatives.

MP: Please explain what ExpertLens is.

OY: ExpertLens is particular software, developed by RAND, which is a derivative on a technique that was actually developed back in the 1950’s, called the Delphi method.

While the Delphi method requires participants to be present in person, the ExpertLens is an online method of gathering qualitative information in a structured way.

                                   

You gather a set of experts into a panel and ask them a set of questions in round one. In round two, participants are presented back not only their own answers but also details about how the group answered, such as the median answer, the first quartile answer and last quartile answer.

In round two, participants take the results and discuss them in an online forum. Participants can view how their answers match with the group median. If their answer is well above the group median, they may stick with that answer and justify their high score or, after participating in discussions, may revise their answer in the next round.

In round three, participants get the opportunity to refine their answers by changing them to either move closer or further away from the group median or not changing them at all.

This three-round iteration process helps us get insight into what the members of the expert panel agree on, disagree on and even what they agree to disagree on.

MP: How is the ExpertLens different from a survey?

OY: The key difference from a survey is that the ExpertLens happens over multiple rounds, allowing respondents to refine their answers. This means that the questions need to be designed in a way that allows us to analyse and report back the answers to the group. For example, questions have to be tailored so respondents rank options into their preferred order or they rate a particular option from one to 10.

Responses to these types of questions allow us to return to the participants with information on how far their answer was from the group average and gives them food for thought for the next round. But a question like “what do you think about this?” will not return answers in a way that can be reported back to the group.

CVF: The key thing about the Delphi method is that it’s a consensus-seeking process. Where consensus is not achieved, the ExpertLens helps illuminate the reasons behind the lack of consensus. So it has in a lot more depth than surveys.

MP: How did the ExpertLens fit from a Mapping Pathways perspective?

OY: For the Expertlens from a Mapping Pathways perspective, we looked at a number of scenarios in which ARV drugs can be used. The strategies that we looked at were oral PrEP, topical PrEP, TLC+ and PEP.

For example, we’d ask a question like “Rank the following four care strategy in order of how cost effective you feel they are for your country.” In a hypothetical scenario, if the median for PEP were 2, we’d ask in the discussion forum whether it correlated with the score different participants gave for it. Participants might then discuss why they gave a particular care strategy a lower or higher score and would be given the chance to change their answer in the following round.

We reminded participants during the process that we were not trying to lobby or push people in any particular direction towards any particular strategy. What we were trying to do was generate an evidence base that would help decision makers make decisions on which strategies, if any, were appropriate for their contexts. Our goal, as with the Mapping Pathways project, is to explore which of these pathways may be right in which context.

Stay tuned to the blog as we bring you the second part of the series as well as slides and posters being presented by the Mapping Pathways team at AIDS 2012. All Mapping Pathways posters, presentations and materials will be archived here.

[Content that is linked from other sources is for informational purposes and should not construe a Mapping Pathways position. Please look for us on Facebook here www.facebook.com/MappingPathways and you can follow us on Twitter @mappingpathways as well.]