Open AccessClinical and Socioeconomic Predictors of Palliative Care Utilization
Author(s) -
Rohan Tummala,
Andrew de Jesus,
Natasha Tillett,
Jeffrey Nelson,
Christine Lamey
Publication year - 2021
Language(s) - English
Resource type - Reports
DOI - 10.21007/com.lsp.2020.0006
Subject(s) - palliative care , medicine , socioeconomic status , poisson regression , family medicine , curative care , health care , ambulatory care , population , nursing , environmental health , economics , economic growth
INTRODUCTION: Palliative care continues to gain recognition among primary care providers, as patients suffering from chronic conditions may benefit from use of this growing service. OBJECTIVES: This single-institution quality improvement study investigates the clinical characteristics and socioeconomic status (SES) of palliative care patients and identifies predictors of palliative care utilization. METHODS: Retrospective chart review was used to compare clinical and SES parameters for three groups of patients: (1) palliative care patients who attended at least one visit since the inception of the University Clinical Health Palliative Care Clinic in Memphis, TN in October 2018 (n = 61), (2) palliative care patients who did not attend any appointments (n = 19), and (3) a randomized group of age-matched primary care patients seen by one provider from May 2018 to May 2019 (n = 36). A Poisson regression model with backward conditional variable selection was used to determine predictors of palliative care utilization. RESULTS: Patients across the three care groups did not differ in demographic parameters. Compared to palliative care-referred non-users and primary care patients, palliative care patients tended to have lower health risk (p < 0.001). Palliative care patients did not differ from primary care patients in socioeconomic status but did differ in comorbidity distribution, having a higher prevalence of cancer (2 = 14.648, df = 7, p = 0.041). Chance of 10-year survival did not differ across risk categories for palliative care patients but was significantly lower for very high-risk compared to moderate-risk primary care patients (30% vs. 78%, p = 0.019). Significant predictors of palliative care use and their corresponding incidence rate ratios (IRR) were hospital referral (IRR = 1.471; p = 0.039), higher number of prescribed medications (IRR = 1.045; p = 0.003), lower Charlson Comorbidity Index (IRR = 0.907; p = 0.003), and lower systolic blood pressure (IRR = 0.989; p = 0.004). CONCLUSIONS: Patients who are expected to benefit from and of being high utilizers of palliative care may experience greater clinical benefit from earlier referral to this service.