For the past 10 years, I have worked on closing the communication gaps that keep older adults at arm’s length from research that could improve their lives.

I worked with Detroiters to bridge the digital divide by developing tools that make it easier for older adults to get online, allowing them to connect to health information and learn about benefits they’re eligible for. I have also codesigned projects with members of the community to help improve older residents’ access to services.

My overriding goal is to help older minority adults connect with research so they are not left out of the very studies meant to reduce health disparities in aging. My work has focused on older adults in Detroit, a majority minority city, to help improve health for all residents.

Despite my best intentions, I recently had an experience where my work created unintended harm for vulnerable people.

I want to share my experience as a cautionary example of how researchers can fail to understand the government benefits that low-income older adults rely on, especially when it comes to research stipends.

Detroit seniors, unplugged

Recently, I completed a project that aimed to bridge both the digital divide and the divide between Detroit residents and researchers.

This project was inspired by the COVID-19 pandemic, when it became clear how hard it was going to be to connect with Detroit residents. Delivering environmental education and outreach is part of my work at the Center for Urban Responses to Environmental Stressors, also known as CURES. In response to the disconnect caused by the lockdown, our team was fortunate to get funding to deliver computers to 1,700 older Detroiters. Our community partners and advisory board members helped us distribute them.

But we soon learned many of the recipients didn’t know how to use the computers, and some couldn’t get online. At the time, Detroit had one of the lowest internet connectivity rates in the nation. Rates were as low as 40% in 2020.

Detroit has used some creative approaches to getting its residents online.

Poor connectivity has been called a “super-determinant” of health. Not being able to access the internet harms individuals because they are cut off from resources that could make them healthier – such as telehealth appointments, for example. It also creates health inequities for groups of people when research lacks a diversity of perspectives.

Naturally, our next step was to develop tools to help the people who received computers get plugged in. We applied for funding, won it – and soon I was working alongside community health and aging advocates, researchers, service providers and housing administrators to develop and refine a technology tool kit.

Once the tool kit was ready, we distributed it widely. The tool kit is designed to cover the basics for older adults who are new computer users. For example, we included directions for connecting to Wi-Fi and creating an email account. We made this resource available for anyone who is interested in using it.

Intro to Research 101

Our community partners next gathered a cohort of 10 Detroiters who were 65 or older so they could learn how to get involved in virtual research. We developed an online research readiness curriculum to introduce them to the basics of how research is done.

Remembering the challenges of COVID-19, we set a goal of engaging the older adults entirely online. Our early meetings started with 45 minutes spent troubleshooting cameras and microphone connections. A few months later, we were all camera-ready in less than nine minutes.

Because I value their time, I budgeted to pay everyone involved in the project. It is difficult for people to take part in programs when they can’t afford to cover basic expenses, and payment can help relieve these financial pressures. What I didn’t realize is that these modest amounts of money could be treated by HUD as income and trigger increases in rent.

But that is exactly what happened.

The older adults involved in our project lived in HUD housing, and their rental costs are based on their income.

We paid residents $120 monthly. This stipend increased their incomes, which in turn led to increases in their rent, sometimes by the same amount as the stipend. Having higher housing costs left them in worse shape than before they joined our project. The stipends were designed to phase out after 10 months, but it was unclear when their rents would be adjusted again. By being involved, their finances became more precarious.

In my opinion, this illustrates how research involvement, even when designed to be fair and respectful, can create an unintended financial strain for people whose budgets leave no room for error.

My privilege was a blind spot.

Problem-solving through partnership

I would likely never have known about the problem if the housing administrator, who was one of my project partners, had not spoken up on behalf of the residents involved in the project. The residents did not come to me to report the issue. They went to a person they already knew and trusted to talk about the unexpected burden.

Some residents stopped accepting payments for their participation, but they still faced months of higher rental costs while we worked to get their money back.

That relief eventually came, thanks to a vigilant HUD administrator, weeks of calls and emails, and late nights spent reviewing HUD’s policies.

In the end, HUD emailed to say it had agreed to exempt the stipends from the residents’ income because we argued that the payments were “temporary, sporadic and nonrecurring.” In response, the HUD site administrator immediately made adjustments, and the overpayments were returned to the residents.

Everyone involved was hugely relieved.

Learning from my mistake

And that may have been the end of the story if one of my community partners, Zachary Rowe, hadn’t encouraged me to write about what happened so that others could learn from our experience.

In my view, this cautionary tale reveals a critical gap in how researchers engage and support people who are underrepresented in studies, including those who rely on housing assistance and other safety net programs. Without attention to these details, efforts to broaden participation in studies can unintentionally exclude or burden the very people researchers are working to include. Experiences like this reinforce that institutions must adapt their policies so paying people for their time never jeopardizes their basic needs.

Researchers, university research review boards and community partners could all benefit from plain‑language guidance about how earnings interact with safety net programs, benefits and income rules. Projects should start with collaborative efforts to anticipate the real-world implications of engagement.

This kind of persistent troubleshooting supports ethical practices and helps build the kind of trust that makes long‑term research partnerships possible.

I view the additional effort and advocacy required to take these precautions as part of the work of shaping who gets represented in research at all. If engaging people with complex lives and constrained resources were easy, our study samples would already be diversified.

Sharing these difficult experiences can be uncomfortable, but it can also help researchers, institutions and partners do better for those who might otherwise be harmed along the way.