The COVID-19 pandemic fundamentally disrupted the U.S. health care system, reshaping not only service delivery but also public perceptions and experiences of care. One revealing source of insight comes from online reviews—an increasingly common way for patients to share their experiences and influence others’ health care decisions.

A team of LDI Fellows—Neil K. R. Sehgal, Anish Agarwal, Raina Merchant, Sharath Guntuku, and colleagues—sought to investigate how patient sentiment toward health care facilities changed before and after the pandemic. Their study analyzed nearly 1.5 million Yelp reviews of 150,000 U.S. health care facilities from 2014 to 2023. It uncovered a striking trend: the proportion of positive reviews declined significantly post-COVID-19, with the sharpest drops occurring in rural areas. 

Using U.S. Census data and rural-urban commuting area codes, the researchers found that beyond rural communities, facilities in areas with higher percentages of Black or White residents also saw steeper declines in positive reviews. In contrast, areas with larger Hispanic populations experienced less of a negative shift in patient sentiment. 

In areas with more Black or Hispanic residents, complaints about insurance and billing issues increased, while in predominantly White areas, negative reviews more often cited longer wait times. These findings suggest that different demographic communities faced distinct challenges in accessing and navigating the health care system post-pandemic. 

According to Guntuku, this research highlights the potential for online reviews to act as an “early warning system” for systemic issues such as rising wait times, billing disputes, and provider shortages—which all intensified during the pandemic.

Read our full Q&A with Neil Sehgal, Anish Agarwal, and Sharath Guntuku below to learn more.

Why are online reviews useful for analyzing patient experiences during the COVID-19 pandemic?

Sehgal: Online reviews were a unique lens through which to see how patients felt during the significant upheaval of the pandemic. These online platforms, like Yelp, are able to capture raw, unfiltered feedback that traditional surveys may miss, like personal stories or frustrations. They give us a broader picture because they’re written voluntarily and often highlight unique aspects of care that surveys may not focus on. We wanted to understand the overall trends and how those experiences differed based on race and geography, which can help spotlight the gaps in care that might otherwise go unnoticed. 

Guntuku: Additionally, before seeking care at a hospital, patients often rely on social media reviews for decision-making.

What key findings on health disparities came from online patient reviews before and after COVID-19?

Sehgal: Our key takeaway is that patient satisfaction with health care facilities, as reflected in online reviews, dropped significantly after the pandemic began. Before COVID-19, most people had positive experiences with the health care system, at least based on Yelp reviews. However, after COVID-19, the majority of people are having negative experiences. This shift reflects how deeply the pandemic disrupted patient experiences with care delivery. 

Our study dug deeper into these trends, identifying highly impacted demographic groups and categorizing their concerns into overarching themes. For example, we found that areas with more Black individuals living in them had a higher proportion of negative reviews after COVID-19 and a much more significant increase in the number of reviews discussing billing issues. These insights can help health care system leaders and policymakers design targeted policy and practice changes.

How did you choose your research methods, and how is the study limited?

Guntuku: We wanted to study longitudinal trends in online reviews over a ten-year period (2014-2023), explicitly identifying pre- and post-pandemic comparisons of patient experiences. 

There are several limitations:

• This study relies on reviews from one platform, Yelp, whose users may not represent the general population.
• We analyzed aggregate ZIP code-level data, which may not always align with individual patient experiences.

Sehgal: Our group has conducted multiple studies using Yelp as a unique source of data to understand people’s health care experiences. However, until now, we had never examined such broad changes over time. Importantly, not everyone uses Yelp, so its user base might not reflect all patients. In addition, we couldn’t directly link reviews to individual patient demographics—just the neighborhoods they came from. Also, online reviews may overrepresent negative experiences, especially during periods of heightened frustration like the pandemic.

What policy changes could improve patient satisfaction and ease disparities?

Guntuku: Focused interventions are necessary to improve health care satisfaction. For instance:

• Wait times and staffing shortages were growing concerns in majority-White communities.
• Complaints about insurance issues increased in Black and Hispanic communities.
• Prescription and medication challenges became more prominent in rural areas.

These findings suggest the need for expanded billing assistance programs and reduced administrative complexities. Monitoring real-time feedback on online platforms can help health systems identify and address different population struggles.

Sehgal: To tackle the disparities we uncovered, policymakers should focus on making health care systems more patient-centered and equitable. For example, simplifying billing and insurance processes could reduce frustration, particularly in communities with higher proportions of Black residents, where billing issues were a recurring concern. Rural health care systems need investment to improve staffing, telehealth access, and infrastructure, as these areas also face some of the largest drops in satisfaction.

Where are you going next? Can other digital platforms help improve patient experiences?

Guntuku: AI-powered systems can help track real-time shifts in patient experiences. We are working on expanding this research to integrate multiple feedback sources to inform quality care.

Agarwal: It would be interesting to think about weaving all patient experiences together to develop a more comprehensive perspective to guide future care. Combining these online reviews with traditional patient experience surveys and with qualitative insights from patients will help drive forward care.

---

The study, “Disparities by Race and Urbanicity in Online Health Care Facility Reviews,” was published on November 22, 2024, in JAMA Network Open. Authors include Neil K. R. Sehgal, Anish K. Agarwal, Lauren Southwick, Arthur P. Pelullo, Lyle Ungar, Raina M. Merchant, and Sharath Chandra Guntuku.

---

Author

---