The Meaningful Involvement of People Living with HIV/AIDS (MIPA) in big data research: The Eng/aging project.

This article is co-authored by Wayne Campbell, Peggy Frank, Patience Magagula, Valerie Nicholson, Miriam Muirhead, Kathleen Inglis and Catherine Worthington. Wayne, Peggy, Patience, and Valerie are long-term survivors living with HIV.

The trend toward “big data” in research has largely excluded community engagement. This absence is acutely felt in the field of HIV research where 40 years of community activism and engagement has been crucial. The Meaningful Involvement of People Living with HIV/AIDS (MIPA) is a principal about ensuring that People Living with HIV are central to the decision-making processes that affect their lives, in terms of research, services, funding, and policies.

Our project entails two nested research elements:

1) Community-led administrative data research on recurrent cardiovascular events among People Living with HIV. (Represented in Image 1)

Peer researchers led the process of identifying a research topic relevant to and useful for People Living with HIV that could be examined using COAST data. Cardiovascular disease (CVD) is an age-associated disease and a leading cause of death among People Living with HIV over the age of 55, with increased incidence rates compared to people living without HIV. Our research question takes a strengths-based approach focusing on survival after a first CVD event and examines the recurrence rates of CVD events among People Living with HIV compared to people without HIV. 

2) Critical reflexive analysis of the research process, which we call “Gathering Wisdom.” (Represented in Image 2)

While doing the COAST research, we continually analyze our process to determine how to engage community as leaders in administrative data research.

Artwork by Peggy Frank, adapted from Joe Average's original piece

Here are just a few of our lessons learned so far: 

Lesson Learned: Need continual and emergent training 

At our first in-person meeting, we recognized that not everyone in the group was at the same level of understanding of administrative data research. We worked together to ensure that all participating group members were at the same level of understanding.

Our five peer researchers determined that the team needed to have an extended time period that was distraction free. We organized a weekend retreat, away from Vancouver. Over the weekend we explored the complexities of COAST/administrative data research, and the sources (databases) we are drawing our information from. We needed to adjust to the impersonal character of the numeric data we would be working with. We also began discussions about the limitations of the COAST information. For instance, data is not collected on ethnicity or gender (just biological sex at birth). And undocumented qualitative information, as well as situations when there was no encounter with the health care system, are invisible to us, but nonetheless realities. For example, COAST data cannot capture social isolation and loneliness, intergenerational trauma, and stress. Since many of our team members had been involved in qualitative research, these limitations were often disappointing.

We also spent time developing our research question(s) or focus. 

Once we had consensus about our research direction, we developed an amateur draft directed acyclic graph (DAG), drawn by an artist-peer researcher.  

We accomplished a lot over the weekend, however more technical training was needed. Special trainings by our data scientists and epidemiologists were organised both in-person and over Zoom to review the ‘how and why’s’ we should incorporate a DAG into our project.  Peer researchers saw how the variables/confounders interacted with each other and influenced other factors.

Having clear research questions and a completed and reworked DAG, the next step was to check with COAST statisticians to confirm that our sample sizes would be large enough. This led to adjustment of certain parameters to increase sample size.

All research has its own unique methods for collecting and using data. Administrative data research is highly technical. Peer researchers needed extra time and training to use their knowledge and collected wisdom to effectively guide the project. In administrative data research, this is necessary for equitable learning. 

In addition, administrative data has unique considerations. Many of the HIV+ peer researchers and qualitative researchers needed to understand that it remains at arms-length from the communities we traditionally work with. The team had to keep reminding ourselves that obtaining data directly from our community was not possible in this project.

Lesson learned: Challenge deficit-based thinking 

Notions commonly used in epidemiology and administrative data research – mortality rates, incidence rates, poor health outcomes, predictors of disease, risk factors -- were activating to team members. For the wellbeing of the peer researchers and community participants in the COAST data, it was essential to actively frame the research on strength. Unsettling the traditional or normal ways of doing administrative data research is feasible and important.

Lesson learned: Need diversity and respect 

The Eng/aging project is comprehensive because of our collaborative team. Team members are dedicated HIV researchers, physicians and clinicians, men and women living with HIV, professors, Elders, students, statisticians and mathematicians. We come from different communities in Canada and different parts of the globe. We have worked with respect and ceremony to gather and unfold the information we needed to address the topics we selected as a team. We acknowledge with gratitude each team member. When we didn’t know details for communities that are not represented by our team, we remained curious and actively reached out to people who could assist us. In this way our team reaches beyond the diversity that we took care to include. 

While these processes represent long-established principles in community-based research, we felt the burden of their weight (or an urgency to address them) in administrative data research which presents unique limitations, as mentioned above. Our team’s diversity and care enabled us to “see” what was missing in the data, why it matters for CVD research (and in general), and how to handle it research-wise, if we can. It also refines our understanding about how to call for better data quality as a future project goal. 

Call to Action 

The ability of people living positively to overcome obstacles and challenges is long documented. People with lived/living experience, in general, are in an ideal position to imagine how best to make use of administrative data research even where it fails to capture certain realities. By involving and supporting peer researchers in the area of big data, research can better reflect the needs of communities affected and have a beneficial impact on their lives.   

We urge big data researchers to engage people with lived/living experience – it makes for better and more effective science. 

If you have any questions regarding the project, please contact Kathleen Inglish at kathleeninglis@uvic.ca.