Predictors of aggressive end-of-life care among Hispanic and white advanced cancer patients

9538

Background: Some ethnic/racial minority cancer patients (e.g. African Americans) have been shown to receive more life-prolonging care at the end-of-life (EOL) than white patients. Nevertheless, few studies have investigated whether Hispanic cancer patients receive more aggressive care than white patients and limited information exists on the predictors of aggressive care among Hispanic patients. The present study examined rates and predictors of aggressive EOL care among Hispanic and white cancer patients. Methods: Subjects participated in a US multi-site, prospective study between September 2002 - August 2008. Data were derived from interviews, conducted in English or Spanish, of 292 self-reported Hispanic (N=58) or white (N=234) stage IV cancer patients and their caregivers. Patients were followed until death, a median of 118.5 days later. Caregiver post-mortem interviews documented patient care in the last week of life. “Aggressive EOL care” was operationalized as cardiopulmonary resuscitation and/or ventilation and death in an intensive care unit. Aggressive EOL care was regressed on the following baseline, patient-reported, independent variables: a preference for life-prolonging care, EOL treatment discussion, do-not-resuscitate (DNR) order, terminal illness acknowledgement, and religious coping. Logistic regression models were stratified by ethnic status and controlled for significant confounds (e.g. socioeconomic status). Results: Hispanic and white advanced cancer patients were equally likely to receive aggressive EOL care (5.2% and 3.4%, respectively; p=0.878). Although religious coping and treatment preferences predicted aggressive EOL care for white patients (adjusted odds ratio 7.76 [p=0.025] and 13.20 [p=0.008]), they were not predictive among Hispanic patients. Hispanic patients were less likely than white patients to have DNR orders (22.4% vs 50.4%; p=0.007). For both white and Hispanic cancer patients, no patient who reported an EOL discussion or DNR order at baseline received aggressive EOL care. Conclusions: Given that EOL discussions and DNR orders may prevent aggressive EOL care among Hispanic cancer patients, further efforts to engage Hispanic patients and their caregivers in these activities may reduce aggressive EOL care.

No significant financial relationships to disclose.

Personalized end-of-life care: Associations between patient preferences and treatment intensity near death

6516

Background: Medical treatment at the end-of-life (EOL) should be responsive to patient values, but research shows it may instead reflect the regional supply of health services and/or physician practice patterns. We sought to determine whether patients' preferences are associated with EOL medical care and place of death. Methods: Coping with Cancer is an NCI/NIMH-funded, multi-site prospective, longitudinal cohort study of patients with advanced cancer, conducted from September 2002-February 2008. Analyses were based upon 317 deceased patients interviewed at baseline and followed until death, a median of 4.4 months later. EOL care was obtained from chart review, and regressed on patients' baseline preferences for life-extending therapy. Analyses were adjusted for significant confounds; i.e., socio-demographics, cancer type, patient denial, EOL discussion, do-not resuscitate (DNR) order, and treatment center. A sub-analysis examined EOL care within three different medical centers to determine whether patients' preferences and treatment intensity varied by region/practice type. Results: 90 of 317 patients (28.4%) reported a desire to receive life-extending therapy at baseline. Patients' preferences did not differ by proximity to death or performance status, but patients who preferred life-extending therapy had higher rates of denial (75.0% vs. 55.6%, p = 0.002) and were less likely to report having an EOL conversation (24.4% vs. 48.9%, p = 0.0002) or a DNR order (24.4% vs. 48.9%, p < 0.0001). These patients were more likely to undergo ICU admission (AOR 4.6, 95% CI 1.9–11.1) and ventilation (AOR 3.0, 95% CI 1.2–7.3), and die in an ICU (AOR 6.7, 95% CI 2.4–18.7) instead of at home (AOR 0.6, 95% CI 0.3–0.9). A sub-analysis of patients treated at Yale, Parkland, and New Hampshire Oncology-Hematology confirmed that patients' preference for life-extending treatment was associated with higher rates of ICU admission in all three centers. Conclusions: Advanced cancer patients' treatment preferences may play a more important role in determining the intensity of medical care at the EOL than previously recognized. Future research should determine whether these preferences are informed and thus reflect patients' true values for EOL care.

No significant financial relationships to disclose.

Spiritual support of patients with advanced illness and associations with end-of-life care and quality of death

9514

Background: Little is known about whether spiritual support is associated with advanced cancer patients' medical care and quality of death (QoD) at the end of life (EoL). Methods: Coping with Cancer is an NCI/NIMH-funded, prospective, longitudinal, multi-institutional study of advanced, incurable cancer patients conducted from September 2002-August 2008. Analyses were based on 343 deceased patients who were interviewed at baseline and then followed until death a median of 117 days later. At baseline, spiritual support was assessed by (1) patient-rated support of spiritual needs by the medical system (eg, physicians, nurses, chaplains) scored from 0 (not at all) to 5 (completely supported) and (2) patient-reported receipt of hospital/clinic pastoral care services. Outcomes measured included medical care received in the last week of life (hospice; receipt of aggressive EoL care defined as ICU admission, resuscitation, ventilation or chemotherapy in the last week of life; and death in an acute care facility) and QoD. QoD assessments (possible 0–30, with increasing scores reflecting better QoD) were obtained by post-mortem interviews of a caregiver present in the patient's last week of life. Multivariable analyses examined (1) associations between spiritual support variables and EoL care outcomes, controlling for baseline confounds (eg, race, religiousness, patient EoL care preferences) and (2) associations between spiritual support variables and QoD with adjustment for confounds (eg, baseline quality of life, aggressive EoL care). Results: In adjusted analyses, greater medical system spiritual support was associated with increased receipt of hospice care [OR = 2.97 (1.24–7.11), p = .01], but not with receipt of aggressive EoL care or death in an acute care facility. Receipt of pastoral care services was not associated with any EoL care outcome. In adjusted analyses, spiritual support from the medical system and receipt of pastoral care services were significantly associated with better patient QoD (standardized β = 0.16, p = .009 and β = 0.20, p = .0005, respectively). Conclusions: Support of advanced cancer patients' spiritual needs by the medical care team is associated with increased receipt of hospice care and improved patient QoD.

No significant financial relationships to disclose.