Healthcare consumers' rapidly growing use of social media platforms has transformed the way hospitals and their communities interact.

According to a new hypothesis-generating HIT Lab study published online in the American Journal of Medical Quality, abundant data from social media activity may serve as tool for public health researchers to evaluate traditional measures of hospital quality.

Patients turn to select hospitals’ profiles on Facebook, Twitter or various other platforms for insight into the organization, a connection with social media-savvy clinicians, and/or free and credible information about medical topics. Conversely, consumers shy away from engaging with other hospitals via social media (and some providers have little to no social presence at all).

“As scholars of the connection between public health and healthcare information communications technology (ICT) we wonder what an individual's choice to opt in to a hospital’s social media presence – specifically, to “Like” a hospital’s Facebook page – says about the hospital itself,” said Alex Timian, lead author of the study, “Do Patients ‘Like’ Good Care?Measuring hospital quality via Facebook.”

Together with HITLAB co-authors Sonia Rupcic, Stan Kachnowski and Paloma Luisi, Timian explores the connections between the number of “Likes” on a hospital’s Facebook page with the quality of care it provides and its patient satisfaction levels.

The paper is based on an exploratory quantitative cluster analysis of Facebook data, 30-day mortality rates and patient recommendation rates for 40 hospitals in the New York, NY metropolitan area. After controlling for bed count, teaching hospital status and the activity (number of admin posts) on each Facebook page, the authors found that:

·         Facebook “likes” have a strong negative association with 30-day mortality rates. Based on the dataset, for every 93 additional Facebook “Likes,” there is a corresponding 1 percentage point decrease in 30-day mortality.

·         The number of “likes” are also positively associated with patient recommendation, although not as strongly as they correspond with 30-day mortality.

These preliminary findings should not be used as determinants of hospital quality or patient satisfaction, the authors noted. Instead, they are hypothesis generating in nature and intended as a platform for future research incorporating other social media tools and larger hospital datasets, analyzing specific hospital services, or looking at hospitals in other regions of the nation or globe.

“We’re encouraged that the correlations support the idea that free, widely accessible data made available via social media will continue to find a place in academic assessment of hospital quality,” Timian said. “Our hope is that this exploratory work will serve as a stepping stone for other public health and ICT researchers to build on by analyzing data from Facebook and other social media tools against various traditional measures.”

Let us know your thoughts on the study. Have you explored associations between social media and traditional indicators of hospitals, clinics, healthcare providers or other stakeholders in the industry?