More than a decade ago, two economists conducted a simple studyopens in new window that found striking evidence of the pervasive reach of racism. They sent more than 5,000 resumes to companies in Chicago and Boston, all identical except for the name on top. Some of the resumes had the stereotypically white names Emily or Greg, and some had the stereotypically African-American names Lakisha or Jamal. The researchers found that the "white" names received 50 percent more calls from hiring managers than the "black" names did.
Now, two new papers from two separate research labs suggest that psychotherapists, like hiring managers, are not immune to bias. Using similar methods, these researchers found that psychotherapists are less likely to offer appointments to African-American and, in one study, working-class potential clients than to white, middle-class ones.
Research has already shown that African-Americans don't access mental health care at the same rate as white Americans. A recent reportopens in new window from the federal Substance Abuse and Mental Health Services Administration, for example, found that between 2008 and 2012, 16.6 percent of white Americans used at least one kind of mental health service, but only 8.6 percent of black Americans did.
The reasons for this disparity are complex. Researchers have explored how structural and cultural barriers like poverty and stigma could contribute, as well as mental health providers' implicit biases—those that operate outside of conscious awareness—and their lack of knowledge about how to work with African-American and other minority patients. The new studies suggest that—at least for those seeking psychotherapy—some disparity may start at the moment when patients begin to seek care.
The studies
In the first study, published in June in the Journal of Health and Social Behavioropens in new window, Princeton sociology graduate student Heather Kugelmass used voice actors to record phone messages for 320 New York City-based psychotherapists, each asking for a new patient appointment.
The psychotherapists, all of whom had doctoral degrees and were selected from the directory of a large HMO, each received calls from one black middle-class and one white middle-class caller, or from one black working-class and one white working-class caller. Each caller mentioned symptoms of depression, mentioned that they were covered by the HMO health insurance plan, requested a weekday evening appointment, and asked the therapist to call back with possible appointment times. Kugelmass varied the names, wording and accents of the callers to indicate race and class, and pre-tested each voicemail with a group of listeners on Mechanical Turk to make sure that it conveyed the race and class information she intended it to.
Overall, she found that 44 percent of the voice messages were returned, and that 15 percent elicited a clear appointment offer from the therapist (some therapists left messages indicating that they weren't accepting new patients, and others simply asked the patient to call back without mentioning availability). Importantly, offer rates varied by race and class—the therapists offered appointments to 28 percent of white middle-class callers but only 17 percent of black middle-class ones. Among working-class callers, only 8 percent of both black and white appointment-seekers received offers.
In the second study, which will be published in November in The Counseling Psychologist, Richard Q. Shin, PhD, of the University of Maryland, and his colleagues used a similar phone-message method, leaving voice mails requesting an appointment for 371 licensed counselors and psychologists in the mid-Atlantic area, whom they selected randomly from an online therapist referral database. Unlike Kugelmass, they looked only at the effect of appointment-seekers' race, and they indicated that race only by changing the caller's name from "Lakisha" to "Allison." They kept everything else in the message, including the callers' wording and accent, exactly the same.
This one seemingly small change made a difference: "Allison" received appointment offers 63 percent of the time, while "Lakisha" received them only 51 percent of the time—a smaller effect than Kugelmass found, but still a statistically significant one.
A broader problem
The two strikingly similar studies were conducted independently—Shin and Kugelmass didn't even know of one another's work until Kugelmass's study was published in June. The overlap might seem unusual, but some experts say the area was ripe for research.
"I think this work is really important, and I'm actually a little surprised that it hadn't been done sooner," says Monnica Williams, PhD, a psychology professor at the University of Louisville and director of the Center for Mental Health Disparities there, who was not involved in either of the studies. "We as mental health professionals haven't focused the lens of scrutiny on ourselves."
That lens has focused in many other directions before now. Over the years, other real-world "audit" studies such as these have found that in addition to hiring managers, landlords are less likely to offer housing to African-American tenantsopens in new window and professors are less likely to meet with prospective studentsopens in new window who are women or minorities.
Fewer audit studies have looked specifically at health-care providers. But other types of research have found that health-care providers are not immune to bias. One recent literature review, for example, found that health-care professionals have low-to-moderate levels of racial/ethnic bias, similar to levels seen in the general population (American Journal of Public Healthopens in new window, 2015).
And psychotherapists, Kugelmass points out, have more opportunity than many other health-care providers to let their biases affect their choice of patients, since many are solo practitioners and thus can choose who to serve.
In their papers, both Shin and Kugelmass suggest that implicit biases may be behind the effects they found. But neither study can prove that connection directly.
"That's a limitation of these studies, at most we can just speculate about what's triggering [the disparities]," Shin says. Presumably, other factors could also be at work, such as therapists' explicit biases or a self-assessment that they're not qualified to work with a patient.
Devising a study to explore the factors at work is tricky, Shin says. He and his colleagues have considered running qualitative, small-group interviews with practitioners, to discuss the results of the study and get their feedback. But of course, the practitioners might not be aware of their own and others' implicit biases.
"It would help us improve our training if we knew more about what was happening," Shin says. "But it's difficult to think of a way to facilitate informative conversations with these participants."
Solutions
So what can psychologists, and psychology, do to address the problem? Simple awareness is a first step, says Lynn Bufka, PhD, assistant executive director for practice research and policy at APA.
"Fundamentally, any therapist first and foremost is human, and humans have biases," she says. "This is something psychologists need to be attentive to, so we can do our best to ensure that our practices are welcoming and accessible to all."
Bufka also notes that the overall callback rates in both studies were low—in Kugelmass's study for example, 56 percent of all callers did not receive a return call. "The fact that over half didn't get a callback was very concerning," she says. "We need to improve that."
But the racial and class-based disparities meant that African-American and working-class callers were hit hardest. To address the root of that problem, graduate programs must invest in teaching psychologists how to serve a diverse, multicultural clientele, according to Williams.
"We have to decide as a field that we're going to prioritize nonwhite issues," she says. Programs need to not only require multicultural competence and diversity training as part of their coursework, she says, but also make sure they hire faculty with the expertise to teach it effectively.
If implicit bias is at work, then there is another angle to explore. Trainings and seminars intended to reduce implicit bias have been gaining popularity in the last few years, and have been adopted by employers including Facebook, Google and the U.S. Department of Justice. But the research on whether these brief trainings work is mixed. Psychologists need to lead the way in developing more effective methods to reduce implicit bias, says Shin: "That's a critical next step."
For a look at one psychologist who is working to increase multicultural competence, read about Dr. Erlanger Turner in our "Rising Stars" profiles.
