Predicting HCAHPS scores from hospitals' social media pages: A sentiment analysis

Social media is an important communication channel that can help hospitals and consumers obtain feedback about quality of care. However, despite the potential value of insight from consumers who post comments about hospital care on social media, there has been little empirical research on the relationship between patients' anecdotal feedback and formal measures of patient experience.

The aim of the study was to test the association between informal feedback posted in the Reviews section of hospitals' Facebook pages and scores on two global items from the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey, Overall Hospital Rating and Willingness to Recommend the Hospital.

We retrieved star ratings and anecdotal comments posted in Reviews sections of 131 hospitals' Facebook pages. Using a machine learning algorithm, we analyzed 57,985 comments to measure consumers' sentiment about the hospitals. We used regression analysis to determine whether consumers' quantitative and qualitative postings would predict global measures from the HCAHPS survey.

Both number of stars and the number of positive comments posted on hospitals' Facebook Reviews sections were associated with higher overall ratings and willingness to recommend the hospital. The findings suggest that patients' informal comments help predict a hospital's formal measures of patient experience.

Consistent with crowd wisdom, ordinary consumers may have valid insights that can help others to assess patient experience at a hospital. Given that some people will judge hospital quality based on opinions voiced in social media, further research should continue to explore associations between anecdotal commentary and a variety of quality indicators.

Administrators can tap into the wealth of commentary on social media as the forum continues to expand its influence in health care. Comments on social media may also serve as an early snapshot of patient-reported experiences, alerting administrators to problems that may appear in subsequent HCAHPS survey results.