Prove It: Racial Bias in Prehospital Analgesia
Does a patient's skin color affect how EMS manages pain?
Medic 17 and Engine 23 respond to a report of an injured person at a construction site. The assessment reveals a conscious and well-oriented male who suffered a mid-shaft right femur fracture when his leg became crushed in some machinery. The patient's vital signs are slightly elevated although given the injury, Paramedics Davis and Martin and not too concerned. Communication between the patient and the paramedics is complicated by the fact that the patient only speaks broken English.
As the paramedics splint the leg and immobilize in preparation for transport, the patient winces with every movement. A construction supervisor asks if Paramedic Davis could administer something to help ease the patient's pain. Davis informs the supervisor that pain medication might actually compromise further assessment and complicate the patient's recovery. With the supervisor in the cab of the ambulance, Medic 17 transports the patient without incident, although the patient continued to complain throughout the twenty-minute trip.
During the initial assessment in the emergency department, the attending physician instructs the staff to administer analgesic medication as soon as possible. Before Medic 17 leaves the hospital, the patient is resting comfortably.
Three weeks later, Paramedic Davis and his partner are called to a meeting with the EMS Chief and the Medical Director. The patient has filed a formal complaint alleging that the EMS field treatment protocols allow analgesic administration given the patient presentation, but the medics did not administer the drugs because of racial bias. The Medical Director wants to know whether the patient's race influenced the decision to withhold the medication.
Research Review: Effect of patient race on prehospital administration of pain medication
Researchers in California examined the effect of patient race on prehospital administration of pain medication (Young, Hern, Alter, Barger, & Vahidnia, 2013). For the study, researchers selected an urban/suburban county with a population of just over one million people.
A single private EMS agency responded to the majority of 911 responses in the county. Paramedics entered the patient care information into an electronic patient care record (EPCR). During a 10-month period in 2009, researchers programmed the EPCR software to force medics to select the patient's race from a drop-down menu before being able to close the record.
Within that EMS agency, morphine was the only analgesic available for field administration. Medical protocols encouraged the paramedics to assess and record the findings of a pain scale and to treat pain with morphine, unless contraindications existed.
Researchers selected all EPCRs where the paramedic chose blunt trauma as the primary impression. From that sample, the team excluded patients under the age of 18 years and patients with a systolic blood pressure less than 85 mm Hg.
This generated a sample of 6561 patients who met the inclusion criteria. From that, the researchers excluded 163 cases because of incomplete records leaving a final sample of 6398 cases.
Of those, 516 patients (8%) received morphine before arriving at the hospital. Statisticians used multivariate analysis to examine the individual impact that a variety of factors had on whether the patients in the sample group received morphine. Neither the sex nor age of the patient appeared to influence the administration of morphine.
As one might expect, the odds of receiving morphine increased as the pain score increased (indicating more severe pain) and with an increased period before arrival at the hospital. However, this study was really about the influence of patient race on analgesic administration.
Statisticians use a technique called regression analysis to create a mathematical formula that shows the effects that each predictor variable has on outcome. Using that formula, the statistician can mathematically adjust all predictor variables except for the variable of interest, in this case race.
This would be analogous to saying, "If all other variables are equal, what effect does race play in morphine administration.
After controlling for the effects of sex, age, time spent with the patient and heart rate, the odds of receiving morphine following blunt trauma decreased if the patient was African-American when compared to being Caucasian, regardless of whether the paramedic recorded a pain score (OR = 0.55, 95% CI [0.33, 0.91]) or not (OR = 0.15, 95% CI [0.05, 0.52]).
The odds of a Hispanic patient receiving morphine was not different from a Caucasian patient if the paramedic recorded a pain score (OR = 0.57, 95% CI [0.33, 1.00]). However, if the paramedic did not record a pain score, the odd of a Hispanic patient receiving morphine significantly declined compared to Caucasian patients (OR = 0.29, 95% CI [0.10, 0.80]).
I hope that all the numbers and symbols in the previous paragraph are not making your head spin. An odds ratio represents the probability that something happens divided by the probability that it does not happen.
An odds ratio of one (OR = 1.00) means that the patient is just as likely to receive morphine as not receive morphine, sort of like a 50-50 chance. When the odds ratio is less than one (OR = 0.55), the patient is less likely to receive morphine; in this case the odds of an African American receiving morphine are almost half (0.55) the odds of a Caucasian receiving morphine (1.00).
In addition, you must always look at the series of symbols and numbers that follow the odds ratio. Bear in mind that regardless of what instrument a person uses to measure something, there is always some amount of measurement error.
Statistics attempts the explain probabilities given certain measurement errors. In the previous example, the notation 95% CI [0.33, 0.91]) means that if you were to conduct this investigation again, you might not get exactly the same odds ratio (OR = 0.55). Instead, you can be 95% confident that your new result would be within the interval 0.33 to 0.91, which is still below the value of one needed for a 50-50 chance.
What This research Means for You
Pain relief is a common reason for people to seek emergency medical care. Researchers using data from the National Hospital Ambulatory Medical Care Survey report that one in five patients transported to the hospital have a complaint of moderate or severe pain (McLean, Maio, & Domeier, 2002).
Researchers in New Zealand found that 54% of patients had a complaint of pain upon arrival at the emergency department (Chambers & Guly, 1993).
The Care Quality Commission (2008) found that many patients experiencing pain while en route to the hospital felt the ambulance crew did not do everything they could to ease their discomfort. This study seems to support the Commission's findings as only 8% of the patients received prehospital analgesia.
Racial bias in analgesia administration is not unique to the prehospital environment. Physicians may under estimate pain severity in three-fourths of African American patients and over half of Hispanic patients (Anderson et al., 2000).
Many studies indicate that patient race and ethnicity affect analgesic administration in the emergency department (Heins et al., 2006; Pletcher, Kertesz, Kohn, & Gonzales, 2008; Tamayo-Sarver, Hinze, Cydulka, & Baker, 2003, Todd, Deaton, D'Adamo, & Goe, 2000; Todd, Samaroo, & Hoffman, 1993) although some researchers have not been able to demonstrate that such bias exists (Fuentes, Kohn, & Neighbor, 2002).
A variety of factors may contribute to analgesic administration disparities among the races. Although two studies failed to demonstrate racial or ethnic differences in the ability to perceive pain (Todd, Lee, & Hoffman, 1994) (Zatzick & Dimsdale, 1990), racial, ethnic or cultural factors can influence the way that patients verbally express pain (Greenwald, 1991) and healthcare providers generally perceive that African Americans feel less pain than do Caucasians (Trawalter, Hoffman, & Waytz, 2012).
These factors may affect health care provider's perceptions of the discomfort level experienced by the patient (Wandner, Scipio, Hirsh, Torres, & Robinson, 2012).
It is also possible that language barriers prohibit effective communication of pain levels, thereby contributing to less frequent analgesic administration to individuals who may not speak fluent English.
Over 80% of paid EMTs and paramedics in the United States are classified as "White not Hispanic" (National Highway Transportation Safety Administration, 2008) and it is reasonable to assume that English is their primary language. At the time of the study, the majority of both the paramedics working in Contra Costa County (78%) and the patient sample (62%) were Caucasian.
Since being Caucasian increased the odds of receiving analgesia in this study, one must consider that paramedics are more likely to administer analgesic medication to patients with similar racial and ethnic characteristics.
Although it is tempting to conclude that simply assessing pain on a standardized scale and recording the results increases awareness and compliance with prehospital analgesia, it is difficult to know whether the paramedics recorded the scores because they performed a more comprehensive examination or whether they proactively decided to give the medication and wanted an objective measure of whether the medication worked.
However, from the patient's perspective, the reason probably matters little.
Paramedics chose the race and ethnicity of the patient. There was no standardized definitions provided and the paramedics were free to choose any race they desired. One must consider that patient self-selection of those variables could have altered the racial composition of the sample, which would produce a different outcome.
Another limitation is in pain perception. There is clear evidence that racial and ethnic groups report pain differently and it is reasonable to assume that difference exists within the same racial and ethnic category even with similar injuries.
The researchers in this study attempted to control for that difference by using a standard pain scale, however there is no way to know that a score reported by one patient means the same thing when selected by a different patient.
The authors did not attempt to control for injury severity. They used the imprecise term "blunt trauma" to select the sample. Blunt trauma can range from relatively minor contusion to life threatening abdominal or thoracic trauma.
It is reasonable to suspect that paramedics would be more likely to administer pain medication to patients with more severe injuries. It is not reasonable to expect that one racial or ethnic subgroup would be more likely to experience blunt traumatic events at a different rate.
Paramedics in this study who spent more time with patients were more likely to administer analgesic medications. The authors do not explain whether the increased out-of-hospital intervals resulted from increased distances or from other factors such as entrapment and extrication delays or a longer and more thorough assessment.
Every patient has a fundamental right to receive effective pain relief from healthcare providers, including those who work in the prehospital environment. Unfortunately, racial disparities in analgesic administration exist, although the reasons for those disparities are not clear. EMS personnel must guard against allowing racial and ethnic factors to influence the decision to administer pain medication.
Anderson, K. O., Mendoza, T. R., Valero, V., Richman, S. P., Russell, C., Hurley, J., DeLeon, C., Washington, P., Palos, G., Payne, R., & Cleeland, C. S. (2000). Minority cancer patients and their providers. Cancer, 88(8), 1929-1938. doi:10.1002/(SICI)1097-0142(20000415)88:8<1929::AID-CNCR23>3.0.CO;2-2
Care Quality Commission. (2008). Briefing note: Issues highlighted by the 2008 category C ambulance service user survey. Retrieved from http://www.cqc.org.uk/_db/_documents/Briefing_note_final.pdf
Chambers, J. A., & Guly, H. R. (1993). The need for better pre-hospital analgesia. Archives of Emergency Medicine, 10(3), 187-192.
Cleeland, C. S., Gonin, R., Baez, L., Loehrer, P., & Pandya, K. J. (1997). Pain and treatment of pain in minority patients with cancer. The Eastern Cooperative Oncology Group Minority Outpatient Pain Study. Annals of Internal Medicine, 127(9), 813-816. doi:10.7326/0003-4819-127-9-199711010-00006
Fuentes, E. F., Kohn, M. A., & Neighbor, M. L. (2002). Lack of association between patient ethnicity or race and fracture analgesia. Academic Emergency Med, 9(9), 910–915. doi:10.1197/aemj.9.9.910
Greenwald, H. P. (1991). Interethnic differences in pain perception. Pain, 44(2), 157–163. doi:10.1016/0304-3959(91)90130-P
Heins, J. K., Heins, A., Grammas, M., Costello, M., Huang, K., & Mishra, S. (2006). Disparities in analgesia and opioid prescribing practices for patients with musculoskeletal pain in the emergency department. Journal of Emergency Nursing, 32(3), 219–224. doi:10.1016/j.jen.2006.01.010
McEachin, C. C., McDermott, J. T., & Swor, R. (2002). Few emergency medical services patients with lower-extremity fractures receive prehospital analgesia. Prehospital Emergency Care, 6(4), 406–410.
McLean, S. A., Maio, R. F. & Domeier, R. M. (2002). The epidemiology of pain in the prehospital setting. Prehospital Emergency Care, 6(4), 402–405.
National Highway Transportation Safety Administration. (2008). EMS workforce for the 21st century: A national assessment. Final report. Washington, DC: NHTSA.
Pletcher, M. J., Kertesz, S. G., Kohn, M. A., & Gonzales, R. (2008). Trends in opioid prescribing by race/ethnicity for patients seeking care in US emergency departments. Journal of the American Medical Association, 299(1), 70–78. doi: 10.1001/jama.2007.64.
Tamayo-Sarver, J. H., Hinze, S. W., Cydulka, R. K., & Baker, D. W. (2003). Racial and ethnic disparities in emergency department analgesic prescription. American Journal of Public Health, 93(12), 2067–2073.
Todd, K. H., Deaton, C., D'Adamo, A. P., & Goe, L. (2000). Ethnicity and analgesic practice. Annals of Emergency Medicine, 35(1), 11–16. doi:10.1016/S0196-0644%2800%2970099-0
Todd, K. H., Lee, T., & Hoffman, J. R. (1994). The effect of ethnicity on physician estimates of pain severity in patients with isolated extremity trauma. Journal of the American Medical Association, 271(12), 925–928. doi:10.1001/jama.1994.03510360051035
Todd, K. H., Samaroo, N., & Hoffman, J. R. (1993). Ethnicity as a risk factor for inadequate emergency department analgesia. Journal of the American Medical Association, 269(12), 1537–1539. doi:10.1001/jama.1993.03500120075029
Trawalter, S., Hoffman, K. M., & Waytz, A. (2012). Racial bias in perceptions of others' pain. PLoS One, 7(11), e48546. doi:10.1371/journal.pone.0048546
Young, M. F., Hern, H. G., Alter, H. J., Barger, J., & Vahidnia, F. (2013). Racial differences in receiving morphine among prehospital patients with blunt trauma. Journal of Emergency Medicine, Article in Press. doi:10.1016/j.jemermed.2012.07.088
Wandner, L. D., Scipio, C. D., Hirsh, A. T., Torres, C. A, & Robinson, M. E. (2012).The perception of pain in others: how gender, race, and age influence pain expectations. Journal of Pain, 13(3), 220-227. doi:10.1016/j.jpain.2011.10.014
White, L. J., Cooper, J. D., Chambers, R. M., & Gradisek, R. E. (2000). Prehospital use of analgesia for suspected extremity fractures. Prehospital Emergency Care, 4(3), 205–208.
Zatzick, D. F., & Dimsdale, J. E. (1990). Cultural variations in response to painful stimuli. Psychosomatic Medicine, 52(5), 544–557.
The author has no financial interest, arrangement, or direct affiliation with any corporation that has a direct interest in the subject matter of this presentation, including manufacturer(s) of any products or provider(s) of services mentioned.
Send correspondence concerning this article to Kenneth W. Navarro, The University of Texas Southwestern School of Health Professions, 5323 Harry Hines Blvd, MC 9134, Dallas, Texas 75390-9134. E-mail: email@example.com