Using virtual health to push equity in health care

Virtual health solutions must be designed and deployed with careful consideration of differing levels of broadband access and availability of technology, as well as mindfulness of appropriate environments for virtual care. Twenty-five million Americans cannot access the internet at home. Lower levels of access especially affect older Americans, Medicaid members, and those with lower incomes. In 19 states, households with a Medicaid enrollee were 10% less likely to have internet access than households without an enrollee.

Adults living in households with an annual income of less than $30,000 per year are far less likely to report using the internet than households with incomes of more than $75,000. Additionally, according to the Pew Research Center, 27% of US adults aged 65 and older reported they did not use the internet in 2019. Beyond internet access, Americans with lower levels of education, lower income levels, and who live in rural areas report lower rates of smartphone ownership. This may mean not all populations benefit equally from the rise of 5G connections, despite the technology’s enormous potential to connect the 46 million rural Americans and 13.6 million urban households without internet access.

Paths to equity:

• Partner with local resources: Work with schools, shelters, libraries, community health centers, and more to set up “Connectivity Zones” where those without reliable internet service at home can go to receive needed virtual care.

• Fund or provide technology directly: Devote resources to bringing virtual health equipment and/or reliable Wi-Fi to patients or members who need it. Such investments may transform how individuals successfully manage chronic conditions from home.

• Offer technical and culturally competent support: Extend care team composition or capabilities to assist in the setup and use of technology, and ensure staff is culturally competent. Such support can enable effective remote patient monitoring and hospital at-home services.

Populations leveraging virtual health solutions have varied levels of digital and health literacy, cultural and language barriers, accessibility needs, and self-advocacy and care team advocacy.

Lack of digital literacy, whether it stems from preference or lack of internet access, can impede the ability to find and use virtual care. Populations with lower digital literacy in the United States tend to be less educated and older, and are more likely to be Black, Hispanic, or born outside the United States. Furthermore, studies have shown that older, lower-income, and less educated populations access health information online less frequently or are less interested in online provider interaction compared to counterparts. These disparities may help explain why some studies have shown these same demographic groups having lower levels of telehealth usage or reporting lower comfort levels of telehealth utilization compared to counterparts.

Paths to equity:

• Tailor intuitive and accessible user experiences: Use simple language, straightforward design, and simplified navigation when designing apps and web pages—and be sure to design for all kinds of modalities and devices, from laptops to tablets to cell phones.

• Conduct targeted outreach: Reach out to segments of the patient population that are less digitally literate to provide assistance and education.
Provide accessible educational materials: Publish step-by-step instructional documents, visual workflows, and video tutorials that can help patients and staff smoothly navigate virtual encounters and educate consumers on online privacy and security.

• Sponsor or launch localized virtual health hubs: Connect patients to physical sites with the technological and staffing resources to assist with virtual visits.

• Leverage existing social supports: Use technology to seamlessly engage and connect social supports (e.g., family members, friends, libraries, and community organizations) in integrated platforms.

As we have noted previously, biases in health care advanced analytics are a serious barrier to a more equitable future. Bias can increase mistreatment, undertreatment, and overtreatment—just one recent example of this phenomenon includes a biased algorithm inappropriately assigning Black women lower chances of successful vaginal birth after caesarean section.

These issues, in turn, can lead to poor health outcomes for consumers and worsened financial performance for hospitals, health systems, health plans, health technology firms, and life sciences companies. Consumer sentiment acknowledges such dangers—57% of consumers believe artificial intelligence has the potential to do damage over the next 10 years due to misuse. Organizations can avoid these outcomes by building processes and procedures (as described below) to spot and combat biases before they infect their use of analytics technology.

Paths to equity:

• Demand robust governance: Establish data guidelines and thresholds and encourage teams to check each other’s assumptions and models for quality and sensitivity. Use robust governance to protect sensitive consumer data, and retrain virtual health technologies (e.g., virtual triage systems) that are discovered to be biased. Demand transparency and explainable information when collecting data sets and training models. Reference Deloitte’s Trustworthy AI framework to accelerate your understanding.
Build diverse teams: Design (or redesign) the organization’s operating model/structure to focus on team diversity. Train data scientists and developers to avoid common technology bias pitfalls within virtual health. In the absence of diverse teams, organizations that are intentional about activating equity can still risk perpetuating bias.

• Apply human-centered design: Leverage human-centered design and testing for accessibility and usability in virtual health applications and programs to help ensure equitable deployment across relevant populations. Involve multiple stakeholders—especially historically underrepresented patient populations—in advanced analytics design.

• Audit proactively: Engage subject-matter experts and tools to audit advanced analytics and underlying data sets used in virtual health programs. Commit to a regular cadence of external audits to detect bias and define mitigation tactics when the tool or model is launched. Recurring audits can help account for changes to consumer behaviors and as the algorithm matures.