JMIR Publications

ABSTRACT

Background:

Patient experience (PE) is directly related to health outcomes and minority groups historically have had worse PE. Evaluation of patient experience (PE) among minority groups has been difficult due to lack of representation in traditional health care surveys.

Objective:

This study assesses the feasibility of Twitter to identify racial differences in PE and changes within the US over time.

Methods:

In total, 851,973 PE tweets from the United States with geographic location information were collected from 2013 to 2016. PE tweets discussed any exposure to health care such as care received in a hospital, urgent care, or any other health institution. Ordinary least squares multiple regression was used to model patient experience sentiment and racial groups across time.

Results:

There was a significant correlation between PE user counts per state and population estimates from the Census 2016 5-year survey (r2=.99; p<.001). The overall mean PE sentiment was highest for Whites, followed by Asians/Pacific Islanders, Hispanics/Latinos and American Indians/Alaska Natives. The change in PE sentiment increased for all racial groups from 2013-2016. American Indians/Alaska Natives experience the highest increase in patient experience sentiment from 2013 to 2016. The greatest reduction in PE disparities was seen with Hispanics/Latinos experience 1.5 times greater increase in PE sentiment compared to whites.

Conclusions:

These findings show the feasibility of using data captured from Twitter to evaluate broader PE in the United States for racial minority groups. PE Social media data can monitor a diverse set of experiences in health care for populations that are not traditional captured.