Implicit Bias Training Betters Psychiatric Care at One Seattle Hospital

LONG BEACH, Calif. — Staff training in implicit bias curbed racial disparities in the use of mechanical restraints at one Seattle hospital, a researcher reported.

From 2015-2017, about 13% of ethnic minority patients were restrained in the psychiatric units at Harborview Medical Center versus just over 8% of white patients, explained Timothy Meeks, MN, RN, the clinical director at Harborview.

After training began in 2017, the 2018-2020 numbers were about 9% for both groups, he said in a presentation at the American Psychiatric Nurses Association (APNA) annual meeting.

“We’re very proud that we’ve been able to get more equity,” said Meeks, adding that Harborview began focusing on racial disparities in psychiatric care in 2016. The staff at the inpatient psychiatric unit is 43% white, 43% Black, and 7% Asian, he said.

He noted that, in general, “Black patients are diagnosed with schizophrenia more often than white patients with the same symptoms. Black individuals are less likely to receive newer antipsychotics than white patients. If a patient goes to the emergency room for a psychiatric evaluation, Black patients are more likely to be physically or chemically restrained than white patients.”

An analysis of Harborview data from 2012-2015 found that, among 4,506 total patients served in the psychiatric units, ethnic minorities were more likely to be:

• Mechanically restrained: 9% (n=241) for whites vs 13% for minorities (n=187, P<0.001)
• Placed into seclusion: 16% (n=415) vs 20% (n=286, P=0.001)

Also, the number of hours in seclusion was higher for minorities at a median of 8.16 versus 6 for whites (P=0.001), with Meeks pointing out that “BIPOC [Black, Indigenous, and people of color] patients were secluded for about 30% longer than our white patients were.”

There was an even larger gap for Black patients in particular, who spent a mean 9.25 hours in seclusion (P<0.001) even though Black patients had fewer mean seclusion episodes per patient (2.45 vs 3.07 for white patients, P<0.001), according to Meeks.

Post-training, he reported that seclusion rates did not shift as dramatically:

• 2015-2017: About 16% of whites were put in seclusion vs about 23% of minorities
• 2018-2020: About 18% vs about 23%

“We still have some work to do with seclusion. The next stage of this project is really to reduce seclusion altogether and make sure we’re using it in an equitable way,” Meeks explained.

He noted that “when we presented this information the first year, we got very strong reactions. It ranged anywhere from anger to denial to a lot of sadness. There were tears that first year. We told them that we didn’t know exactly what was causing this, but we suspected that it was probably implicit bias that was driving a lot of these disparities.”

He said the hospital is in the process of examining length of stay data for psychiatric patients, as there are indications that white patients may stay longer, possibly because staff worry that they’d be too vulnerable in shelters. “I do see a pattern there, and it’s something we’re going to keep looking at,” Meeks said.

An APNA attendee asked Meeks about trends in violence against staff members. Meeks reported that post-training, workplace violence in the psychiatric units fell about 14% from 2020-2021, although other interventions may have played into that, he acknowledged.

Meeks also addressed the fact that the data did not look at differences between the groups in terms of psychiatric conditions or presentations. He also was asked if all patients in the same minority or ethnic group should be managed the same way.

“I wouldn’t say that the goal is to treat patients the same,” Meeks told MedPage Today. “The goal is to eliminate healthcare disparities. It’s pretty obvious across the whole spectrum of data on implicit bias that disparities exist in mental health care. That’s what we found when we looked [at their institution], so we’re trying to fix it.”

Training consists of identifying levels of implicit and explicit bias, understanding confirmation bias, use of the Implicit Association Test (IAT), and small group discussion sessions.

“Our hope is that we can get people to slow down so that implicit bias doesn’t creep into their decision making, particularly around seclusion and restraints,” Meeks said.