Haas
JS, Weissman JS, Cleary PD, Goldberg J, Gatsonis C, Seage GR 3rd, Fowler
FJ Jr, Massagli MP, Makadon HJ, Epstein AM.
Discussion of preferences for life-sustaining care by persons with AIDS.
Predictors of failure in patient-physician communication.
Arch Intern Med 1993;153(10):1241-8.
The two purposes of this study were 1.) to assess whether certain characteristics
of patients with AIDS and of their physicians were associated with the
likelihood of discussing life sustaining treatment and 2.) to determine
if differences in patient-physician discussions regarding this care were
related to patients’ desires to communicate with their physician
about this issue.
The study participants were recruited from three sites in Boston, Massachusetts
that provided care to more than 20% of AIDS patient in surrounding areas.
The participants were interviewed and information was collected on “demographic
and socioeconomic characteristics, illness-related factors, use of health
care services, health status, preferences for resuscitation, and whether
preferences for resuscitation had been discussed with their physician.”
The physician “most responsible for the care of each patient”
was also interviewed (or a respective physician informant) at each site.
The investigators hypothesized that “patients treated by physicians
with demographic characteristics similar to those of the patients would
be more likely to discuss their preferences for care.”
Only 38% of patients had discussed their preferences for life-sustaining
care with their physician, yet 72% reported wanting to have this discussion
with their provider. Further, those patients who were less likely to have
discussed this preference with their physician were nonwhite rather than
white (26% vs. 42%), had lower pre-illness income (25% vs. 41%), were
heterosexual (25% vs, 43%), had poorer physical functioning, had an AIDS
diagnosis for no longer than 1 year, or were taking zidovudine.
After adjustment for possible confounders, only race, previous hospitalization
and affiliation with an HMO healthcare system remained statistically significantly
as to whether or not the patient had discussed a preference with the physician.
More importantly, race remained an independent predictor of patient-physician
communication, whereas socioeconomic factors, such as level of education
and income, were not. Overall, the results clearly indicated that nonwhites,
as compared to whites, were significantly less likely to have discussed
their preference for life-sustaining care with their physicians.
Interestingly, among nonwhites, “56% of those with a nonwhite provider
had discussed resuscitation compared with 22% of those with a white physician
(P=.03).” However, for white patients, “physician characteristics
were not associated with the discussion of preferences for life sustaining
care.” The authors noted that “it is unlikely that the practices
of a few physician explain [the] findings” since the characteristics
of over 25 physicians were measured and no physician cared for more than
13 patients.
When examining the relationship between patient desire for communication
and race, the authors found that “among those who had not discussed
resuscitation with their physician, 75% of nonwhites wanted to do so,
as did 71% of whites.” The investigators concluded, “patient
desire to discuss life-sustaining care, therefore, does not seem to explain
the observed racial differences in patient-physician communication.”
In summary, “nonwhites appear to be at significant risk for not
discussing their preferences for life-sustaining care, with no clinical
rationale.” Although a possible explanation is that “differences
in communication by race reflect a greater desire for patient-physician
communication among whites,” the results of this study clearly refute
this hypothesis since “the desire to discuss the issue is similar
among whites and nonwhites.” The authors suggest that “the
different patterns of discussion among nonwhite patients in relation to
the race of their provider suggest that the inadequacy in patient-physician
communication may be mediated in part by ‘mismatch’ of patient
and provider race.” Nonetheless, the authors recommend that since
“particular groups of patients appear to be at higher risk for this
failure in communication, interventions to improve communication should
target these populations.”