How Hospitals Can Counteract Inaccurate Crowdsourced Ratings

Crowdsourced ratings such as Facebook reviews reflect patient experience better than clinical quality, and hospitals can push back.

Crowdsourced ratings of the "best overall" hospitals produce scores similar to Hospital Compare's ratings, but crowdsourced ratings are less reliable as indicators of clinical quality and patient safety, recent research shows.

The research, which was published in the journal Health Services Research, examined hospital ratings on Facebook, Google Reviews, and Yelp. The findings show crowdsourced ratings reflect patient experience rather than other factors.

"For the most part, what we found is that the social media scores tell us about patient experience, but they don't tell us about the best and worst hospitals on the basis of clinical quality or patient safety," the lead author of the research, Victoria Perez, PhD, told HealthLeaders last week.

The study has significant implications for how patients should view crowdsource ratings, said Perez, who is an assistant professor at Indiana University in Bloomington. "We wish that people would understand that even if hospitals are not scoring well on Facebook in user reviews, they could have excellent clinical scores."

Addressing bad reviews
 

If a hospital has generated negative reviews on a crowdsourcing site, there are ways to counteract the negative publicity, Perez said. "Hospitals can advertise that they score well on Hospital Compare and establish marketing strategies to respond to social media scores."

Hospital leaders also need to be aware that crowdsourcing scores are based largely on patient experience, she said.

"If hospitals are worried that patients are just looking at social media scores, they need to realize the scores reflect patient experience rather than clinical quality and patient safety. Other than advertising they don't have a lot of control over this. The options are marketing and engaging on the social media platform."

There are ways to shift the online focus of patients toward clinical quality and patient safety, Perez said.

"Hospitals can share Hospital Compare clinical quality and patient safety scores on their homepage, on their Facebook page, and on Twitter. Many hospitals have a social media presence, so they can definitely share clinical quality and patient safety information, and they can encourage patients to look at Hospital Compare."

Research findings
 

The research, which examined data from nearly 3,000 acute care hospitals, has several key findings:

• For best-ranked hospitals on the crowdsourcing sites, 50% to 60% were ranked best in Hospital Compare's overall rating.
   
• For best-ranked hospitals on the crowdsourcing sites, 20% ranked worst in Hospital Compares overall rating.
   
• For clinical quality and patient safety, hospitals ranked best on crowdsourced sites were only ranked best on Hospital Compare about 30% of the time.

Perez said Hospital Compare, which combines as many as 57 metrics for patient experience and clinical quality, was used to gauge the accuracy of the crowdsourcing sites for several reasons.

"The clinical quality and patient safety measures are based on Medicare claims data, which means there is a lot of information about patients and they can do risk adjustment," she said of Hospital Compare.

Risk adjustment is crucial when comparing hospitals, Perez said. "Rather than being concerned that some hospitals are treating a sicker pool of patients and have worse outcomes as a result, the Hospital Compare data can be adjusted for the health of the patient mix."

The crowdsourcing sites are more prone to bias, she said. "A concern when you look at social media is that people only write reviews when they have really good or really bad patient outcomes."