HLTH 2110 Week 5 Evidence Critique Example

What 222 Medical Students and Residents Can Carry: Reading Hoffman and Colleagues' Study of False Beliefs and Pain From the Methods Section Outward

Student Name

Walden University

HLTH 2110: Behavioral and Cultural Issues in Healthcare

Instructor Name

Month Day, Year

What 222 Medical Students and Residents Can Carry: Reading Hoffman and Colleagues' Study of False Beliefs and Pain From the Methods Section Outward

The Source and Why It Is Cited

Hoffman et al. (2016) is among the most widely cited studies on racial bias in pain care. Its headline finding is that roughly half of the white trainees surveyed rated at least one untrue statement about Black and white bodies as possibly or more likely true, such as that Black people's skin is thicker, and that those who endorsed more of these beliefs judged a Black patient's pain as milder and recommended treatment less accurately. The study is often cited as evidence for why Black patients are undertreated for pain, a disparity documented across many settings (Anderson et al., 2009). This critique reads the study from its methods outward, because the finding means only what its sample and measures allow.

Recruitment

The paper reports two studies. Study 1 recruited 121 adults through Amazon's Mechanical Turk, an online platform where people complete tasks for small payments. Study 2, the one about medical trainees, recruited first-, second- and third-year medical students at one large public university, who completed the survey online during class sessions, and residents from multiple sites, who completed it online at their convenience (Hoffman et al., 2016).

Two features of this recruitment matter. First, all the medical students came from a single university. Their beliefs may reflect that school's curriculum, region and student body, and nothing in the design allows the finding to be extended to medical students elsewhere. Second, the two subgroups were recruited differently: students during class time, residents on their own time. Students in a classroom are close to a captive audience, while residents who chose to respond may differ from those who did not.

Response Rate

The paper does not report a response rate. The authors state that they had no set sample size and collected data from as many participants as they could. Without knowing how many students were enrolled in the class sessions or how many residents were invited, a reader cannot judge whether those who took part resemble those who did not. For the residents especially, a low response rate could mean the sample overrepresents people with particular views on race or pain.

Exclusions and the Sample Within Each Category

Of 418 medical students and residents who took part, 222 were analyzed. The authors excluded participants who were not native English speakers or not American, and they excluded all participants who were not white, explaining that the study concerned white people's beliefs in the historical context of Black and white relations in medicine. They report that including the non-American and non-native-English participants did not change the pattern.

The exclusion of nonwhite participants defines the population: the finding concerns white trainees only. That is a legitimate design choice, but citations that describe the result as applying to "medical students" or "doctors" overstate it. The category sizes matter even more. Only 28 residents were analyzed. Residents are the group closest to independent practice, yet any statement about residents' beliefs rests on 28 people recruited from unnamed sites with an unknown response rate. The sample can support statements about white medical students at one university more firmly than statements about residents.

Table 1

Study 2 Sample Before and After Exclusions

Figures from Hoffman et al. (2016); shares calculated for this critique.

The Wording of the Questions

Participants rated 15 statements about biological differences on a six-point scale from "definitely untrue" to "definitely true." Eleven were false statements; the authors averaged those 11 into the measure of false beliefs. The figure that about half of trainees endorsed false beliefs counts anyone who rated at least one of the 11 items as "possibly true," "probably true" or "definitely true."

This wording choice shapes the headline. "Possibly true" is the mildest form of agreement on the scale, and a trainee unsure about one item out of 11 is counted the same way, for the headline figure, as one who is certain of several. The authors also report the average: participants endorsed about 11.6% of the false beliefs. Both numbers are accurate. They describe different things: the share of people with any doubt about any item, and the typical level of endorsement. Citing only the first makes the finding sound stronger than the second suggests.

The Outcome Measures

Pain ratings and treatment recommendations were made for two mock medical cases, one describing a Black patient and one a white patient. Participants did not treat real patients. The paper itself names this as a limitation: survey methods were used for practical reasons, and future work would need to test whether the beliefs affect treatment in real medical settings. The effect sizes the authors report were small (partial eta squared of 0.03 and 0.04), though they argue the practical importance is meaningful, since those endorsing more false beliefs rated the Black patient's pain about half a point lower on the scale.

What the Sample Can Carry

Read from the methods outward, the study supports a narrower claim than its usual citation. Among white medical students at one public university and a small group of white residents from several sites, some false beliefs about racial biology were held, usually weakly, and those who held more of them rated a Black patient's pain in a written case lower than a white patient's. That is an important finding, and it is consistent with broader evidence of bias among health care professionals (Hall et al., 2015). It cannot carry a claim about how many physicians nationally hold these beliefs, a claim about residents specifically or a claim that the beliefs cause undertreatment of real patients. Those claims require samples and measures this study did not have.

Conclusion

Hoffman et al. (2016) designed a careful study within practical limits and reported those limits openly. The problems lie less in the study than in how it is often cited. Its sample is white trainees, mostly students at one university, with no response rate reported; its headline figure counts mild uncertainty as endorsement; and its outcomes are judgments about written cases. A writer who cites it should say so, and should match the claim to what 222 people answering a survey can show.

References

Anderson, K. O., Green, C. R., & Payne, R. (2009). Racial and ethnic disparities in pain: Causes and consequences of unequal care. The Journal of Pain, 10(12), 1187-1204. https://doi.org/10.1016/j.jpain.2009.10.002

Hall, W. J., Chapman, M. V., Lee, K. M., Merino, Y. M., Thomas, T. W., Payne, B. K., Eng, E., Day, S. H., & Coyne-Beasley, T. (2015). Implicit racial/ethnic bias among health care professionals and its influence on health care outcomes: A systematic review. American Journal of Public Health, 105(12), e60-e76. https://doi.org/10.2105/AJPH.2015.302903

Hoffman, K. M., Trawalter, S., Axt, J. R., & Oliver, M. N. (2016). Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites. Proceedings of the National Academy of Sciences, 113(16), 4296-4301. https://doi.org/10.1073/pnas.1516047113