THE RELATIONSHIP BETWEEN DISEASE BURDEN, CARE SETTING, AND LIFE-SUSTAINING TREATMENT CHOICES

Over 70% of the U.S. population is expected to die from advanced chronic illness. We hypothesized that high illness burden, measured by the number and severity of comorbidities and patient age (Charlson Score), leads to low odds for choosing aggressive life-sustaining treatments. Life-sustaining treatment preferences were collected from 593 Physician (or Medical) Orders for Life-Sustaining Treatment forms (MOLST) at three hospitals. Logistic regression models were used to estimate the odds for choosing All Treatment (aggressive) vs. Limitations to Life-sustaining Treatments for patients with Charlson>5 (vs. ≤ 5), controlling for decision-maker (patient/proxy), care setting (with/without a palliative care practice; MD/PA/NP), and patient demographics. Over one-third (36%) chose All Treatment. Mean Charlson Score was 8 (SD=3). Proxy decision-makers signed 43% of the forms. Mean patient age was 71 (SD=15); 49% were male; and 83% were non-Hispanic White. Palliative care clinicians administered 50%, and both non-palliative and palliative MDs administered 52% of the MOLSTs. Patients with Charlson>5 were 70% (OR=0.31; p ≤ 0.001) less likely to choose All Treatment than those with Charlson ≤ 5, when controlling for only patient characteristics (pseudo R 2 =0.148; p ≤ 0.001). Adding care-setting variables showed that Charlson>5 still reduced the odds for choosing All Treatment. However, the participation of palliative care clinicians attenuated the magnitude of the relationship between illness burden and life-sustaining treatment preferences (OR=0.54; p ≤ 0.10) (pseudo R 2 =0.431; p ≤ 0.001). These results support our hypothesis that high illness burden is negatively associated with preferences for All Life-sustaining Treatments, but results also point to the strong influence of palliative care clinicians in these decisions.