Title

UTILIZATION OF COMMUNITY-BASED PALLIATIVE CARE AND HEALTH OUTCOMES AMONG ELDERLY PANCREATIC CANCER PATIENTS IN THE UNITED STATES: A RETROSPECTIVE STUDY OF SEER-MEDICARE DATA

Date of Award

Fall 12-2018

Degree Name

Doctor of Public Health (DrPH)

Advisor(s)

CECILIA GANDUGLIA CAZABAN, MD, MPH, DRPH

Second Advisor

SRIRAM YENNURAJALINGAM, MD, MS, FAAHPM

Third Advisor

HUI ZHAO, MD, PHD

Abstract

Background: The absence of population-based research on community-based palliative care (CBPC) use and how CBPC influenced overall survival, quality of end-of-life (EOL) care, opioid utilization, and opioid-induced toxicity for elderly pancreatic cancer patients indicated a knowledge gap in understanding their needs of palliative care in communities. This study evaluated patterns and predictors of CBPC utilization, and how CBPC was associated with overall survival, quality of EOL care, opioid utilization and induced toxicity in this population using the SEER-Medicare data. Patients and Methods: 16,106 Medicare beneficiaries diagnosed with pancreatic cancer from 2007 to 2013 were identified and followed up until 2014. Trends in CBPC use and in referral time after pancreatic cancer diagnosis were assessed. The patient-level predictors, geographic-level factors, and health service area characteristics of CBPC use were determined. The influence of CBPC on overall survival over a two-year follow-up and on hospitalizations, emergency department (ED) visits, intensive care unit (ICU) use, hospice referral and stay, chemotherapy use, and life-extending care use were assessed associations between CBPC use and opioids prescribing, and induced toxicity were investigated. Results: 9.5% of study patients used CBPC. CBPC use increased from 8.2% to 9.4% (P = .014) and the median time from diagnosis to referral decreased from 5 to 2.5 months (P < .001) between 2007 and 2013. CBPC use was positively associated with being female, being white, having a diagnose of advanced pancreatic cancer, remaining less comorbid, maintaining cancer-directed treatment, and living in health service areas with a high density of hospital-based palliative care (HBPC) programs. CBPC users had a lower hazards ratio of death than did matched HBPC users (hazards ratio [HR], 0.78; P < .001). They were less likely than matched HBPC users to be admitted to ED (relative risk [RR], 0.64; 95% CI, 0.54 to 0.77), hospital (RR, 0.46; 95% CI, 0.31 to 0.67), and to use life-extending care (odds ratio [OR], 0.63; 95% CI, 0.45 to 0.89). They tended to die at home (OR, 1.43; 95% CI, 1.15 to 1.78) and to have shorter hospital (b = −0.54; 95% CI, −0.91 to −0.18) and ICU stays (b = −0.29; 95% CI, −0.53 to −0.04). Compared to non-palliative care (NPC) users, they had higher likelihood of hospice referral (OR, 1.31; 95% CI, 1.10 to 1.56) and of longer hospice stay (b = 0.59; 95% CI, 0.09 to 1.09). CBPC users used higher (P< .001) adjusted and weighted mean average opioid daily dose within 4 months after starting palliative care but had a lower risk of developing delirium (HR: 0.93, P<.001) than NPC users. Conclusions: CBPC use increased and its referral time from cancer diagnosis decreased over the 6 years among elderly pancreatic cancer patients. Patient-level factors, geographic locations, and regional health care resources predicted CBPC use. CBPC use was associated with improved survival, less intensive EOL care, higher doses of opioid prescribing, and lower incidence of opioid-induced toxicity compared to HBPC use.

This document is currently not available here.

Share

COinS