by Kory M. Johnson, Emma Fixen, David J. Elpern, Douglas W. Johnson
- Mount Sinai Health System, New York, NY
- John A. Burns School of Medicine, PGY1, Honolulu, HI
- David Elpern MD: The Skin Clinic, Williamstown, MA
- University of Hawaii School of Medicine, Honolulu, HI
Keywords: implicit bias, explicit bias, immigrant, indigent, psoriasis, lymphoma, healthcare
A 46-year-old Micronesian woman with history of severe disabling psoriasis and psoriatic arthritis had been our patient for many years (Figures 1 and 2). Her psoriasis started as an adolescent and had been treated with numerous therapies including topical steroids, phototherapy (sunlight exposure in Micronesia), methotrexate, cyclosporine, acitretin, apremilast, etanercept, infliximab, ustekinumab, secukinumab, ixekizumab, IL-23 blocker, and adalimumab. All of these therapies failed to significantly adequately control her psoriasis, but she felt that adalimumab helped the most with her psoriatic arthritis. Her treatment was complicated by the development of multiple squamous cell carcinomas. Overall, she was on adalimumab for the longest of all these therapies, approximately 10 years in total. At the time of this case report she was on guselkumab and had received two injections.
She presented to dermatology clinic in January 2018 for psoriasis follow-up and 3rd IL-23 blocker injection. She said her psoriasis had not improved and complained of groin pain and fever for three days. New-onset erythema, tenderness and induration of the right proximal thigh was noted. Preliminary diagnosis was abscess and incision and drainage was attempted in office and was unsuccessful. Systemic psoriasis treatment was withheld. She was sent to the emergency department where a repeat attempt was made to incise and drain the right groin mass but failed to drain. She was then admitted for intravenous antibiotics and a CT was obtained that showed a 5.5 cm right thigh mass concerning for necrotic lymph node versus abscess as well as widespread lymphadenopathy. A left groin lymph node FNA was performed three days after admission and pathology showed peripheral T cell lymphoma. She was seen once by oncology as an inpatient, discharged and referred for outpatient lymphoma treatment.
Eight months later she returned to dermatology clinic having stopped her psoriasis treatments. When asked about treatment for her lymphoma, she described negative interactions on her first and only visit with the outpatient oncology staff and oncologist, stating that “those people weren’t nice to me” and that she didn’t want to go back to see them. She stated that she planned to take “medications from home” and use alternative herbal therapies. An attempt was made to refer her to other oncologists in the community; but they refused to see her and suggested that she go back and see her first oncologist.
Thirteen months after her lymphoma diagnosis she again presented to dermatology clinic with severe psoriasis, extensive lymphadenopathy, weight loss, bilateral severe leg edema, difficulty walking, and increasing shortness of breath (Figures 3 and 4). She was wheelchaired to the emergency department, evaluated and sent home at this time because her PO2 was satisfactory. Six days later she re-presented to the emergency department with severe abdominal pain and was found to be in septic shock with hydronephrosis from tumor-ureteral obstruction. She died seventeen days later from multi-organ failure secondary to proteus mirabilis urosepsis. She had never received treatment for her lymphoma.
We wish to examine the factors that contributed to the failure of this patient to receive appropriate treatment for her lymphoma. We believe this patient’s medical care and ultimate death was affected by a combination of implicit biases, including bias against her race (Micronesian), insurance (Medicaid), and disease (severe psoriasis). Implicit bias may have played a role in her inability to obtain timely diagnosis and oncologic treatment during both admissions, in the three-day delay before an FNA was performed during her first hospitalization, and in her failure to be admitted during her second visit to the emergency department. She was sent home from her first hospitalization without receiving any treatment for her lymphoma, and received a cold reception at the oncology office when she tried to follow-up as an outpatient. She was later sent home from the emergency department in a debilitated state, only to be readmitted six days later in septic shock, and ultimately died without treatment for her lymphoma.
Implicit biases are unconscious associations between a group attribute, such as race, and a negative evaluation or stereotype . These involuntary biases can affect judgment and interpersonal interactions with patients, despite a conscious desire to treat everyone equally. Compare this to explicit biases, which are held consciously, believed to be correct in some way, and acted upon by the biased person . Implicit biases against certain patient populations can negatively affect communication between patient and provider and the care that these patients receive. This has been shown to impact patient confidence in oncologic therapies. In one study, looking at interactions between non-black oncologists and black patients, oncologists with higher pro-white/anti-black racial implicit bias (as measured by the implicit association test) had shorter patient interactions and their communication was rated by patients and observers as less patient-centered and supportive . Higher oncologist implicit bias was also associated with increased patient difficulty remembering contents of the patient-provider interaction, and indirectly with less patient confidence in recommended treatments and more perceived difficulty in completing them.
In our case, we believe implicit bias against the patient’s race may have played a significant role in her medical care and in the failure to receive treatment for her lymphoma. Unfortunately, there exist significant implicit and explicit biases against the Micronesian community in Hawaii. A recent study from the University of Hawaii found that one in four Micronesians surveyed had experienced bias against them at work due to being Micronesian. One in ten reported discrimination by medical and social services . Another study conducted in Hawaii found that Micronesians tended to be hospitalized at a younger age compared to White or Japanese patients, and that they had higher severity of illness at time of hospitalization across a variety of categories (cardiac illness, infectious disease, cancer, endocrine) . Micronesians were also significantly more likely than the other groups studied (White, Japanese, or Native Hawaiian) to be either Medicaid or self-pay patients.
Like many Micronesian patients in the studies referenced above, our patient was on Medicaid. Both implicit and explicit bias and discrimination against patients based on insurance status can affect patient access to care. For example, many physicians may not accept Medicaid patients due to low reimbursement rates, increased paperwork requirements (prior authorizations, limitations on prescriptions), and high illness burden and psychosocial demands within this patient population . A study using data from the 2013 Minnesota Health Access Survey found that reports of insurance discrimination were higher among uninsured and publicly insured adults compared to privately insured individuals. Adults reporting insurance-based discrimination in this study were also more likely to face barriers to accessing care including lack of a usual source of care, lack of confidence in getting needed care, foregoing needed care due to cost, and provider-level barriers such as refusal of insurance or refusal to accept new patients . Another study looking at 2007 data from Florida Medicaid enrollees found that 14% of respondents reported discrimination based on Medicaid insurance, while 9% reported discrimination based on race/ethnicity, and 6% reported both types of discrimination . Patients who reported discrimination based on either race/ethnicity or insurance status also had worse patient experiences (looking at variables such as getting needed care, timeliness of care, provider communication) than those who did not perceive discrimination.
Bias against individuals with severe cutaneous disease may have also played a role in this case. This includes historical discrimination against people with leprosy (seen during the 19th and 20th century in Hawaii when people diagnosed with or suspected of having leprosy were exiled to Molokai) as well as present-day discrimination. Specifically, patients with psoriasis, such as our patient, may face discrimination. A study from 2018 found that laypersons reported social avoidance of people with visible psoriasis, and belief in myths such as psoriasis being contagious or that it only affects the skin and is not a serious disease . Medical students were also surveyed and reported significantly fewer stigmatizing attitudes than laypersons; however, some of them still had stigmatizing views. Knowing someone with psoriasis or having heard of the disease previously predicted fewer stigmatizing attitudes. Discrimination may also contribute to the psychiatric co-morbidities known to be associated with severe psoriasis. For example, a 2017 systemic review and meta-analysis found that patients with psoriasis had significantly higher likelihood of suicidal ideation, suicide attempts, and completed suicides, and that those patients who were younger and whose psoriasis was more severe were at higher risk for suicidality .
In conclusion, the sad fate of our patient highlights the importance of recognition of implicit bias in ourselves and our colleagues. Implicit biases are insidious and can negatively impact patient care precisely because they are unconscious. Hence, raising awareness of these biases is the first step in addressing them. Discrimination against individuals with severe cutaneous disease, both implicit and explicit, can have significant negative effects on patients’ wellbeing and health. Furthermore, biases against patients based on race, ethnicity, or insurance status can significantly affect access to care as well as the quality of care that they receive. How might this patient’s story have been different if she was a white practicing attorney or wife of a staff physician in our community? All physicians and especially dermatologists should be aware of this, so that proactive action can be taken to mitigate these effects.
- FitzGerald, C. and S. Hurst, Implicit bias in healthcare professionals: a systematic review. BMC Med Ethics, 2017. 18(1): p. 19.
- Blair, I.V., J.F. Steiner, and E.P. Havranek, Unconscious (implicit) bias and health disparities: where do we go from here? Perm J, 2011. 15(2): p. 71-8.
- Penner, L.A., et al., The Effects of Oncologist Implicit Racial Bias in Racially Discordant Oncology Interactions. J Clin Oncol, 2016. 34(24): p. 2874-80.
- Rebecca Stotzer, P., Bias Against Micronesians in Hawai’i. 2019, University of Hawai’i at Mānoa Myron B. Thompson School of Social Work. p. 2.
- Hagiwara, M.K., et al., Younger and Sicker: Comparing Micronesians to Other Ethnicities in Hawaii. Am J Public Health, 2016. 106(3): p. 485-91.
- Sommers, A.S., J. Paradise, and C. Miller, Physician willingness and resources to serve more Medicaid patients: perspectives from primary care physicians. Medicare Medicaid Res Rev, 2011. 1(2).
- Han, X., et al., Reports of insurance-based discrimination in health care and its association with access to care. Am J Public Health, 2015. 105 Suppl 3: p. S517-25.
- Weech-Maldonado, R., et al., The relationship between perceived discrimination and patient experiences with health care. Med Care, 2012. 50(9 Suppl 2): p. S62-8.
- Buckingham, V.L.J., Histories of Leprosy: Subjectivities, Community and Pacific Worlds. he Journal of Pacific History, 2017. 52(3): p. 265-286.
- Pearl, R.L., et al., Stigmatizing attitudes toward persons with psoriasis among laypersons and medical students. J Am Acad Dermatol, 2019. 80(6): p. 1556-1563.
- Singh, S., et al., Psoriasis and suicidality: A systematic review and meta-analysis. J Am Acad Dermatol, 2017. 77(3): p. 425-440 e2.
Figure 1: Example of this patient’s baseline cutaneous disease and deforming psoriatic arthritis
Figure 2: Severe psoriasis affecting both legs
Figure 3: Prominent axillary lymphadenopathy
Figure 4: Prominent cervical lymphadenopathy