The Impact of Implicit Bias on Prescribing Habits in Type 1 Diabetes
Do you remember first learning about insulin pumps or continuous glucose monitoring? Maybe your doctor or someone on your diabetes care team brought it up as a suggestion to help you or your child with diabetes management. You decided to check with your employer-paid health insurance to see if these technologies would be covered and then chose which devices would be best for you or your child. After the learning curve of figuring out the technologies, you did have better outcomes just as study after study shows happens when you use insulin pumps, CGMs, or closed loop systems.
But what happens when your diabetes team does not bring up this option to you? What if you did not hear about insulin pumps as an option until you met other people with diabetes in person or online, and then you ask yourself, why wasn’t this an option for me? Was I not “in control” or “compliant” enough to deserve the option of using things to make diabetes management easier? The truth is that everyone with diabetes deserves to have access to the tools and technologies that can keep you healthy and improve your quality of life. The reality is that there are many things that get in the way of peoples’ access.
One of the ways that access can be limited is through implicit bias in the health care provider. Whether we like it or not, we, as humans, carry implicit biases with us –by definition unknowingly. This means that everyone, including nurses, doctors, pharmacists, and the like all carry their own biases that could impact the care of their patients. Sometimes these biases are related to physical attributes such as race/ethnicity, gender or sexuality; other times they can be more nuanced such as biases against people of lower socio-economic status (SES).
Dr. Ananta Addala and her team at Stanford University conducted a study about implicit bias among a multi-disciplinary team of diabetes health care providers to determine if it affected prescribing habits.1 Specifically, they were interested in whether the providers’ implicit bias against youth with public insurance influenced their recommending diabetes technologies. The implication is that having public insurance means the person is from a lower socioeconomic status. They titled it, “The Gatekeeper Study,” since, in essence, the health care provider is the gatekeeper between patients’ and their medical treatment options.
They were able to recruit 39 health care providers from all over the U.S. to participate. The participants then completed a new tool named “the diabetes provider implicit bias tool,” which includes two parts. The first is a case example where two identical cases are presented in a randomized fashion, where the only difference is the insurance type – public or private. The second part of the tool has a ranking exercise of patient factors that are considered important for the recommendation of diabetes technology. They were completed separately for insulin pumps and CGMs.
Most of the participants were female (89.7%), non-Hispanic White (79.5%) physicians (48.7%). Notably, 30% of the participants had a personal history of T1D, and a majority worked in urban and academic institutions. Of these 39 participants, a bias against public insurance was present in 33, meaning 84.6% of the participants. The providers who were newer to their practice had less bias present but was not dependent on the age of the provider. The authors discussed the continued challenges in maintaining coverage of public insurances for diabetes technologies and the increased requirements for work on both the patient/family side and the health care provider side as possible contributing factors into this provider bias.
This was one of the first studies of its kind and, hopefully, the start of more research examining ways in which health inequities can be reduced and health outcomes improved. The authors suggested that the challenges associated with changing coverage policies and higher standards for youth with public insurance to get access to diabetes technologies may contribute to the bias seen in the participants. Furthermore, they surmised that the solution to this issue will not be simple and will require changes for both providers and insurers to help reduce the inequities for pediatric type 1 diabetes.
If you’re interested in seeing what your own implicit biases are, Harvard University has a great resource titled “Project Implicit” where you can complete tests for a variety of common biases.
Written and clinically reviewed by Marissa Town, RN, BSN, CDCES
