Predictors of patient misunderstanding of the goals of care at initiation of chemotherapy.

Abstract

9038 Background: All patients should be apprised of the risks of treatment before starting chemotherapy. According to the ASCO QOPI core measures, providers should also discuss the associated goals of care (GOC) at the time of consent discussion. We sought to determine if attributes of patients and initial patient-MD encounters were associated with subsequent patient-MD concordance regarding GOC. Methods: At a single academic medical center, we surveyed 137 consecutive patients within 30 days of their consent for first chemotherapy to assess their understanding of the GOC (i.e., curative vs. non-curative). The patients had solid tumors and represented all stages of disease. The survey was designed to assess patient understanding of the GOC and to measure several areas that are plausibly relevant to this understanding. We compared this to the GOC documented by their treating oncologists in the institution's chemotherapy order entry system (required). Results: The survey response rate was 91% (125/137). Using the MD GOC as the gold- standard, we found that only 76% (95/125) of patient-MD pairs were concordant regarding the GOC; 16% (20/125) of patients erroneously believed their GOC was curative; and 8% (10/125) erroneously believed their GOC was not curative. In a multivariable logistic regression that controlled for patient demographics and other patient factors, we found that patients greater than 65 years of age had an almost 80% lower odds (OR 0.21, 95% CI: 0.08-0.55) of concordance compared to younger patients; that patients for whom English was not their native language also had an almost 80% lower odds (OR 0.22, 95% CI: 0.05-0.93) of concordance compared to native English speakers; and that patients who received printed materials during the patient-MD consent process had almost three times higher odds (OR 2.97, 95% CI: 1.13-7.79) of concordance compared to those not receiving such materials. Conclusions: Patient misunderstanding of MD's GOC was substantial. Key predictors of discordance included factors potentially amenable to interventions at the time of patient chemotherapy consent including provision of printed material and use of communication aids targeted at elderly and non-native English speaking patients. No significant financial relationships to disclose.