Last week our live learning event, Unconscious Bias in Medical-Decision Making, was attended by 225 organizations from 40 states in the U.S., Canada, and Switzerland. In this blog post, Dr. Joseph Betancourt responds to seven questions we didn't have time to answer during the event.
Q1: Is institutional (larger systems) bias connected to provider bias? And how do we change this phenomenon on a wider level?
If our institutions have certain practices, or have certain “unspoken” rules, we are absolutely susceptible to them at the individual level. For example, in many hospitals, Medicaid patients and others with limited financial resources are seen in settings primarily staffed by medical residents or medical students. This policy, which may emerge from reimbursement issues, can be seen as biased against Medicaid or poor patients, given that their care is primarily delivered by trainees—something that isn’t routinely done for patients with commercial insurance.
This institutional bias can then create the bias (or stereotype) in providers that “these” patients are less worthy of seeing attending physicians, and that commercial patients shouldn’t be seen by trainees. This may certainly have an impact on quality care—consciously or subconsciously. It could then be further exacerbated if a large percentage of an institution's low income/Medicaid patients are also people of color—where now the stereotype takes on another dimension.
Q2: Does unconscious bias training change people's actions?
Q3: How can you measure impact of unconscious bias training?
I'll answer these two questions together. This field is early in its evolution, and we have mixed evidence regarding the impact of unconscious bias training. Given that unconscious bias is a very real problem within healthcare, our efforts cannot be held back by lack of definitive evidence. In fact, we engage in lots of educational activities within healthcare that are not held to this standard.
For example, since the early 2000s there has been an incredible focus on patient safety. Comprehensive initiatives have been designed to prevent medical errors, with components that include informational campaigns, robust tracking systems, standardization of practices, and lots of education and training for the healthcare team. Today we have seen a reduction in medical errors over the years due to the impact of all of these efforts—combined. However, we have not separated out patient safety training to study its independent impact on the overall improvements.
The point I’m trying to make is that education is important, and just one part of a broad portfolio of systemic interventions and strategies—including job aides, systems for accountability, and incentives—that lead to behavior change and improved quality. We will get there, slowly and steadily, evaluating step by step. Along those lines, here is a paper I wrote that provides guidance on how we should measure the impact of cultural competence training that can be helpful as we try to measure the impact of training on stereotyping and unconscious bias.
Q4: Long after trainings and discussions like this, how can hospitals and clinics sustain the conversation and encourage continued positive behavior?
Q5: How do you gain prioritization for this issue without becoming a target or a trouble-maker?
This must be done in a smart, strategic way, aiming for messaging that promotes engagement, not disengagement. The words we use will matter, and we should test several out in our environment to see which are more effective, and bring more people along. We have great science to back all of this up now, great studies on impact of stereotyping on decision-making—these should be our starting points. The evidence on disparities, and the fact that unconscious bias and stereotyping have been identified as one of the root causes, can become a place to anchor this work.
We should also position the importance of caring, and putting patients first, as our main goals. An institution can sustain this work by building a campaign that aims for improvement, and equitable, high-quality care for all. Few will disagree with such an effort. We must assume that health systems, and healthcare providers, want to do the right things. Having blind spots that we aren’t aware of, or can’t manage, preclude us from assuring everyone we care for gets the best we have to offer.
If your organization is able to identify a few disparities locally, or within your system, that serves as great substrate for an initiative like this, and urgency—so long as this training isn’t the only thing that is done, and it is coupled with other strategies and interventions.
Q7: What are the specific studies that demonstrate an improvement in patient satisfaction and/or outcomes when cultural competency training is completed?
At Massachusetts General Hospital where I work, we took the time to survey minority patients to get at their experiences of care, above and beyond the standard methods and questions used in HCAHPS. We conducted a dedicated survey and in 2004, for example, it demonstrated that 21% of African American patients, and 25% of Latino patients felt like the white patient next to them was getting better care than they were.
From 2005 to today we have engaged in an entire portfolio of cultural competence training for doctors, nurses, and front line staff using our tools at Quality Interactions. In 2012, when we re-surveyed, our numbers improved significantly. Just about 9% of African Americans, and 6% of Latinos felt like the white patient next to them was getting better care than they were. This was our proof of concept that cultural competence training led to improvements in patient experience—and this was just one of a broad series of questions we asked where we saw improvement. More information can be found at www.mghdisparitiessolutions.org.