Abstract
This cross-sectional study assesses medical records for evidence of racial bias in clinician documentation among patients admitted for breast cancer, sickle cell disease with crisis, status epilepticus, hypertensive emergency, pneumonia, or motor vehicle collision.
Black patients frequently fare worse than White patients with respect to numerous health outcomes even after controlling for socioeconomic factors.1 In clinical situations where pathophysiology does not vary by race, patient race may predict differences in treatment and ultimate outcome.2,3,4 This suggests that racial bias is present and may be associated with variations in patient care. Our study looked for evidence of explicit racial bias in clinician documentation.
Methods
We conducted a retrospective cross-sectional study of patients admitted to an urban academic medical center. The study was approved by Virginia Commonwealth University’s Institutional Review Board. Consent was not obtained per institutional policy for retrospective research. Six diagnoses were chosen based on the presence or absence of racial differences in epidemiology, pathophysiology, or treatment strategy: breast cancer, sickle cell disease with crisis, status epilepticus, hypertensive emergency, pneumonia, and motor vehicle collision. The first consecutive 200 adult patients with each diagnosis were identified by querying discharge diagnoses in the electronic medical record from January 1, 2018, through December 31, 2019.
Patient medical records were examined for race and ethnicity as documented in the electronic medical record, routinely assigned on check-in by registration clerks. Admission notes were screened for the mention of race in the first line of the history of present illness. Clinician race, gender, and level of training were determined using publicly available information and photos.
Bivariate analysis to determine unadjusted association with race documentation was performed using the χ2 or Fisher exact test for categorical variables and t test for continuous variables. Multiple logistic regression was used to assess adjusted associations between race documentation and characteristics of patients and clinicians. Analyses were performed with R, version 3.5.3 (R Foundation for Statistical Computing). All significance tests were 2-sided, and P < .05 was considered significant.
Results
Our analysis included 1200 patients. Patient and clinician characteristics are summarized in the Table. Race was documented in the history of present illness in 323 patients (26.9%), including 257 of 777 Black individuals (33.1%) and 63 of 384 White individuals (16.4%) (P < .001). Univariate correlations with race documentation are shown in the Figure. After adjusting for other factors, Black patients had higher odds of having race documented (adjusted odds ratio [aOR], 1.57; 95% CI, 1.11-2.25, compared with White patients), and Black clinicians were less likely than White clinicians to document the patient’s race (aOR, 0.42; 95% CI, 0.20-0.80). Attending physicians were more likely to document race than residents (aOR, 2.37; 95% CI, 1.73-3.27). The patient’s age and gender and the clinician’s gender were not statistically significant predictors after adjusting for other factors. Hispanic ethnicity accounted for a small percentage of patients (6 [0.5%]) and clinicians (68 [5.7%]), and categorizing these individuals separately did not qualitatively change our results.
Table. Patient and Clinician Characteristics.
| Characteristic | No. (%) | ||||||
|---|---|---|---|---|---|---|---|
| Overall (n = 1200) | Hypertensive emergency (n = 200) | Sickle cell crisis (n = 200) | Status epilepticus (n = 200) | Motor vehicle collision (n = 200) | Breast cancer (n = 200) | Pneumonia (n = 200) | |
| Patient characteristics | |||||||
| Age, mean (SD), y | 52.2 (18.2) | 56.9 (14.7) | 34.7 (12.8) | 55.9 (16.6) | 46.8 (20.3) | 60.0 (13.0) | 58.9 (16.6) |
| Female | 663 (55.2) | 88 (44.0) | 120 (60.0) | 93 (46.5) | 78 (39.0) | 198 (99.0) | 86 (43.0) |
| Race | |||||||
| Black | 777 (64.8) | 166 (83.0) | 197 (98.5) | 115 (57.5) | 73 (36.5) | 101 (50.5) | 125 (62.5) |
| White | 384 (32.0) | 32 (16.0) | 1 (0.5) | 78 (39.0) | 108 (54.0) | 96 (48.0) | 69 (34.5) |
| Other | 39 (3.2) | 2 (1.0) | 2 (1.0) | 7 (3.5) | 19 (9.5) | 3 (1.5) | 6 (3.0) |
| Clinician characteristics | |||||||
| Female | 484 (40.3) | 75 (37.5) | 108 (54.0) | 66 (33.0) | 57 (28.5) | 82 (41.0) | 96 (48.0) |
| Race | |||||||
| Black | 66 (5.5) | 18 (9.0) | 12 (6.0) | 13 (6.5) | 0 | 10 (5.0) | 13 (6.5) |
| White | 776 (64.7) | 125 (62.5) | 114 (57.0) | 135 (67.5) | 151 (75.5) | 139 (69.5) | 112 (56.0) |
| Asian | 356 (29.7) | 57 (28.5) | 74 (37.0) | 52 (26.0) | 48 (24.0) | 50 (25.0) | 75 (37.5) |
| Other | 1 (0.1) | 0 | 0 | 0 | 0 | 1 (0.5) | 0 |
| Level of training | |||||||
| Resident | 612 (51.0) | 135 (67.5) | 38 (19.0) | 170 (85.0) | 32 (16.0) | 117 (58.5) | 120 (60.0) |
| Attending | 563 (46.9) | 62 (31.0) | 160 (80.0) | 27 (13.5) | 168 (84.0) | 71 (35.5) | 75 (37.5) |
| Advanced practice clinician | 25 (2.1) | 3 (1.5) | 2 (1.0) | 3 (1.5) | 0 | 12 (6.0) | 5 (2.5) |
Figure. Race Documentation by Patient Characteristics and by Note Writer Characteristics.
A, Race documentation by patient characteristics; B, Race documentation by note writer characteristics. APC indicates advanced practice clinician; BC, breast cancer; HTN, hypertensive emergency; MVC, motor vehicle collision; PNA, pneumonia; SCC, sickle cell crisis; SE, status epilepticus.
Discussion
We found significant differences in documentation of race in admission notes based on the race of the patient and the race and training level of the clinician. Our study is limited by being conducted at a single urban center. The use of visual inspection to identify clinician race is not optimal; however, this approach has been shown to be very accurate when categorizing White and Black individuals.5
The variation based on clinician suggests that younger physicians may be less inclined to consider a patient’s race when providing medical care, which may reflect successful efforts to raise cultural awareness in medical schools. It could also indicate that they are less aware of health disparities and do not find the designation helpful in any way. The variation in documentation by diagnosis may also offer hope. The trauma surgeons treating motor vehicle collisions rarely documented race, which may suggest that small-scale efforts within a department can be effective. In contrast, Black women are screened less frequently for breast cancer and present with more advanced disease, so race may be more relevant to that diagnosis.
Further study is needed to examine whether differential documentation of race is associated with subsequent care of the patient. If so, it would provide a simple way to identify those clinicians who might benefit from additional training in racial awareness. Removing race from the history of present illness altogether may further our goal of providing exceptional care to all patients regardless of race.
References
- 1.Nelson A. Unequal treatment: confronting racial and ethnic disparities in health care. J Natl Med Assoc. 2002;94(8):666-668. [PMC free article] [PubMed] [Google Scholar]
- 2.Vaccarino V, Rathore SS, Wenger NK, et al. ; National Registry of Myocardial Infarction Investigators . Sex and racial differences in the management of acute myocardial infarction, 1994 through 2002. N Engl J Med. 2005;353(7):671-682. doi: 10.1056/NEJMsa032214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hausmann LR, Ibrahim SA, Mehrotra A, et al. Racial and ethnic disparities in pneumonia treatment and mortality. Med Care. 2009;47(9):1009-1017. doi: 10.1097/MLR.0b013e3181a80fdc [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Whitman S, Ansell D, Orsi J, Francois T. The racial disparity in breast cancer mortality. J Community Health. 2011;36(4):588-596. doi: 10.1007/s10900-010-9346-2 [DOI] [PubMed] [Google Scholar]
- 5.Gomez SL, Kelsey JL, Glaser SL, Lee MM, Sidney S. Inconsistencies between self-reported ethnicity and ethnicity recorded in a health maintenance organization. Ann Epidemiol. 2005;15(1):71-79. doi: 10.1016/j.annepidem.2004.03.002 [DOI] [PubMed] [Google Scholar]
