PROVIDENCE, R.I. [Brown University] — A lot of evidence shows that a patients' race or ethnicity is associated with differences in how health care providers communicate with them, the health care they receive, and their health outcomes. In HIV care, a key to those outcomes is whether people take their medications as prescribed. A new study of the doctor-patient dialogue about HIV drug adherence found several specific differences in those conversations depending on patients' race and ethnicity.
The study, which appears online Jan. 25 in the journal AIDS and Behavior, is based on a highly structured analysis of recorded office visits between 45 health care providers and 404 patients, 245 of whom were black and 59 of whom were Hispanic. Researchers used the Generalized Medical Interaction Analysis, which divides the dialogue into units called "utterances" based on speech acts such as asking questions, giving information, giving instructions, making requests, or expressing desires. It also classifies the topic or subject matter of each utterance.
Lead author M. Barton Laws, assistant professor (research) of health services policy and practice in the Brown University School of Public Health, and his colleagues produced three main findings:
Different speech patterns: Because black patients spoke less to their providers than either white or Hispanic patients, they experienced significantly greater provider dominance in their discussions. Black patients and their providers exchanged fewer expressions of goals or values than did Hispanics and whites. Meanwhile, providers asked Hispanic patients fewer open-ended questions and their discussions involved less humor.
More dialogue about HIV drug adherence with minorities: Providers and either black or Hispanic patients exchanged a higher volume of dialogue — more utterances — about adherence than did providers and white patients. This difference occurred regardless of how adherent patients were to their medication regimens or whether lab tests showed that their disease was under control.
Directives, not problem solving: In those more extensive dialogues between black and Hispanic patients with their providers, there were not any more utterances about problem-solving than for whites. Instead there were more provider directives (e.g. "Take your medicine or you'll get sick") said to the minority patients than to whites.
Laws said that the research does not reveal why providers – 34 of whom were doctors and 30 of whom were white, with most of the rest of Asian ancestry — talked more with minority patients than with whites about adherence. One hypothesis – yet to be tested – is that the providers may be attempting to compensate for what they read in studies about lower adherence among minorities.
"The possibility that seems most compelling to me is the doctors don't trust their black and Hispanic patients as much to be adherent," Laws said. "It has been epidemiologically observed that they do tend to be less adherent, but it's not because they are black or Hispanic."
Other factors likely cause a lack of adherence, he said, but how the providers spoke to minority patients didn't correlate with whether they were adherent. Doctors also didn't spend much time trying to find out whether patients faced barriers to staying on their medications and helping find solutions.
In previous research analyzing office visits, Laws and colleagues found that physicians generally tended to use more directives than engage in problem solving when addressing HIV adherence. But research in behavioral counseling shows that simply exhorting people again and again does not make them more likely to engage in a desired behavior.
The emphasis on directives, rather than problem solving, is likely a product of how providers relate to all patients, he said, regardless of race. However, the other differences in the overall visits are suggestive of ways in which differences in culture and background may affect communication.
Laws cautioned that the current study does not examine whether the racial differences in the dialogue lead to different clinical outcomes, but that question can now be investigated.
"We've found there is something going on and it would be good to understand it better," Laws said. Ongoing work with collaborators is assessing whether provider communication styles can be changed, and how communication patterns are associated with outcomes.
In addition to Laws, other authors on the paper are Yoojin Lee and Dr. Ira Wilson at Brown, William Rogers at Tufts Medical Center, Mary Catherine Beach and Richard Moore at Johns Hopkins University, Somnath Saha and P. Todd Korthuis at Oregon Health and Science University, Victoria Sharp at St. Lukes-Roosevelt in New York, and Jonathan Cohn at Wayne State University.
The research was supported by the National Institute of Mental Health (grants R34MH089279 and RO1MH083595) and the Agency for Healthcare Research and Quality (grant: 290-01-0012).
AIDS and Behavior