Behind every datum is a real person’s life: moving medicine toward equity

The fiction that technology can, in and of itself, cure societal problems has been proven again and again to be nonsense. But it arises (like the phoenix or a zombie, depending on how charitable you care to be) with each new technological innovation. AI, for instance, was going to cure the imbalance of human prejudice in the justice system. Instead, we uncovered the widespread pervasiveness of this problem in the “objective” data and algorithms themselves, making the cure worse than the disease

Now, triggered by COVID-19, another injustice has come to the fore. Disparities in care quality and health outcomes across patient groups from different socioeconomic backgrounds have become impossible to ignore, particularly in the United States.

These disparities include geographical access to healthcare providers; access to health insurance; cost, trust, and bias issues; and a well-documented refusal by the healthcare industry to listen to minority patients regarding issues including pain, and social determinants of health that shape patient wellness, such as access to healthful food options.

So instead of yet again cheerleading the imaginary magic wand of technology, let’s examine and describe the problem with a steady hand and see if the wise use of data might help patients and providers turn the tide against the growing problem of inequality in healthcare. 

By the way, we use the term “health equity” instead of health equality. Equity is each person or group getting what they need versus equality, which is each person or group getting the same thing. 

The current state of health equity

“Before the pandemic hit, there was public scrutiny on health equity but it didn’t have the same ramifications as it does now,” says Christina Jimenez, Senior Product Marketing Manager at Snowflake. “When COVID hit, the strain on the healthcare system caused providers to rethink their timelines for using data and analytics to drastically improve care delivery without shortchanging quality. It became obvious that the healthcare system is capable of doing that in record time. We have a responsibility to take that same data-driven approach and ensure that certain patient populations aren’t shortchanged in the care that they receive.”

Widespread disparities in patient care outcomes are well documented in the U.S. One example is when patients are faced with language or cultural barriers, which can complicate care coordination and follow-up.

“Imagine a patient who arrives at a hospital and speaks no Spanish, but speaks an indigenous regional dialect,” says Jimenez. “They face an incredibly challenging journey through the healthcare system. As a care team, what data do you have on hand to ensure that patients like this one receive the same level of care as any other patient in your unit? How easy is it for this patient to access additional services they might need, such as a medical interpreter or transportation assistance to make their follow-up appointments? These are the types of questions we need to solve for, and COVID has shown us that we have the analytical means to do it.”

The inequity of research

After the initial contact with those who need help, inequity has a deeper, wider effect. 

Saurabh Gombar, Co-Founder and Chief Medical Officer at Atropos Health, a company that provides real world evidence to clinicians to improve care decisions, notes the widespread inequality in the research that produces treatments which propagate across the healthcare system. 

“There is something in medicine called ‘the evidence gap,’” he says. “If you look at randomized, controlled trials and prospective studies over the last few decades, about 80% of them will not include patients who have multiple common comorbid conditions (for example, arthritis and insomnia), or who are from classes of patients that are routinely understudied.” 

Pregnant and breastfeeding women are frequently excluded from large trials. This includes the early COVID vaccine trials. So are the frail and those experiencing housing insecurity. 

“These routinely understudied classes have a higher proportion of minorities,” says Gombar. “All of those patients are understudied.”

The understudied are the underserved. These groups, as Gombar indicates, are not only ethnic minorities, though they are intensely underserved. It is also those experiencing poverty, or who can’t afford or don’t have access to adequate health insurance. 

“How do you as a health system ensure that patients without robust insurance coverage still receive the same level of outreach for their next appointment?” asks Jimenez. “Answering that question starts with having fresh data, deeper patient insights, and the willingness to dig into how your economically disadvantaged patients receive care at your facility.” 

To address inequity, we need to ensure that people can actually get care and that they’re supported by the system before and after the treatment itself. 

How we can use data as a tool for health equity

For Gombar, if data is not the whole answer, it is a major part of the solution. 

“One of the ways you can solve healthcare inequity is through greater data sharing   across healthcare organizations” he says. “To accomplish this, we will need to look at all patients who routinely get care. For example, in the U.S. we need to look at all 300 million individuals in the country.”

Jimenez concurs. “Patient privacy and data security concerns are just a few factors that are complicating data sharing, along with the massive technical debt that the industry is struggling with,” she says. “The healthcare community at large has so much work to do from a data sharing and interoperability standpoint. But it’s necessary for us to move forward in terms of improving equity of care.”

Another obstacle to data

Let’s say you go to a general practitioner. She refers you to a specialist. At each stop, you have to provide your information. It’s never just one-and-done. 

“There is a lack of access to patient information throughout the system,” says Jimenez. “It’s not a new issue, and some of those roadblocks to data access exist for a good reason. It’s ultimately to protect patients’ privacy. But without a unified view of the patient across the care continuum, there’s a good chance that you’ll miss the bigger picture, which means that care quality and equity can be compromised.” 

We have elsewhere examined the challenge and promise of a “blue button” function that would bring together any data needed by a patient, practitioner, pharmacist, or researcher without compromising on our privacy and safety. 

Data is no abstraction

Too much of the discussion surrounding health equity is feel-good, a kind of “nice day for some nice data” story. The reality is that healthcare is a fundamental human right, and when true health equity is achieved, everyone benefits. 

“There’s a lot more pushing we need to do on this issue. We have the data and we have the technology to make it happen. What’s really lacking is the determination to actually solve for it,” says Jimenez.

None of this—not the needs, not the challenges, not the possible solutions—are abstractions. Because as Jimenez notes, “Behind every data point is a real person’s life and we don’t have time to waste.” 

Links for additional reading

Snowflake conversations

Papers and reports