Abstract
Dear Editor:
We appreciate the comments about some of the limitations of our paper on the relationship between treatment intent and end-of-life preferences. In response, we recognize that from a prognosis and treatment point of view, the use of only two categories (curative intent surgery (CIS) or noncurative intent treatment (non-CIT)) may be overly simplistic. However, we chose to dichotomize because we thought people who chose treatment expecting a cure may be different from people who did not expect a cure (in terms of their general treatment preferences expressed in an advance directive). We acknowledge that patients might not fully understand whether treatment is curative or noncurative and, had this been a prospective study, we would have asked about their knowledge of treatment intention specifically.
The authors of the letter also expressed concerns about generalizability, as it wasn't clearly stated whether participants differed from nonparticipants in important ways. Though not reported here, we previously presented data as an oral presentation and a poster (and are working on a manuscript) that describes such differences.1,2 We found that participants and nonparticipants were similar in age, gender, and race, but differed in that nonparticipants lived significantly farther away from the medical center (median distance: 51 versus 25 miles; p < 0.001). Further, the fact that 10% of individuals referred to the study chose to participate speaks to the inherent challenges of doing research on end-of-life issues among a seriously ill population, findings that are similar to other studies.
The authors also expressed an interest in learning more about how participants' end-of-life preferences were measured. Though not included in our manuscript due to space limitations, elsewhere we have described in more detail how MYWK identifies the preferences of patients.3,4 Finally, in response to a concern about the artificial nature of the hypothetical scenarios used in the study, we agree that scenarios are not the same as real life. That said, they do provide an opportunity for people to indicate how they would respond to common situations that result in loss of decisional capacity (such as stroke, trauma, etc.); and these scenarios were developed after much consideration. Further details about the rationale for their use can be found in a previous publication. 5