Hospice utilization among Medicare beneficiaries dying from pancreatic cancer

Advance Care Planning (ACP) in Medicare Beneficiaries with Heart Failure

BACKGROUND

Heart failure is a leading cause of death in the USA, contributing to high expenditures near the end of life. Evidence remains lacking on whether billed advance care planning changes patterns of end-of-life healthcare utilization among patients with heart failure. Large-scale claims evaluation assessing billed advance care planning and end-of-life hospitalizations among patients with heart failure can fill evidence gaps to inform health policy and clinical practice.

OBJECTIVE

Assess the association between billed advance care planning delivered and Medicare beneficiaries with heart failure upon the type and quantity of healthcare utilization in the last 30 days of life.

DESIGN

This retrospective cross-sectional cohort study used Medicare fee-for-service claims from 2016 to 2020.

PARTICIPANTS

A total of 48,466 deceased patients diagnosed with heart failure on Medicare.

MAIN MEASURES

Billed advance care planning services between the last 12 months and last 30 days of life will serve as the exposure. The outcomes are end-of-life healthcare utilization and total expenditure in inpatient, outpatient, hospice, skilled nursing facility, and home healthcare services.

KEY RESULTS

In the final cohort of 48,466 patients (median [IQR] age, 83 [76-89] years; 24,838 [51.2%] women; median [IQR] Charlson Comorbidity Index score, 4 [2-5]), 4406 patients had an advance care planning encounter. Total end-of-life expenditure among patients with billed advance care planning encounters was 19% lower (95% CI, 0.77-0.84) compared to patients without. Patients with billed advance care planning encounters had 2.65 times higher odds (95% CI, 2.47-2.83) of end-of-life outpatient utilization with a 33% higher expected total outpatient expenditure (95% CI, 1.24-1.42) compared with patients without a billed advance care planning encounter.

CONCLUSIONS

Billed advance care planning delivery to individuals with heart failure occurs infrequently. Prioritizing billed advance care planning delivery to these individuals may reduce total end-of-life expenditures and end-of-life inpatient expenditures through promoting use of outpatient end-of-life services, including home healthcare and hospice.

Effects of hospice-shared care on terminal cancer patients in Taiwan: A hospital-based observational study

PURPOSE

To assess how hospice-shared care (HSC) affected the likelihood of aggressive medical treatments and the life quality among terminal cancer patients.

METHODS

In the first part, a cohort of 160 late-stage cancer patients who died in non-hospice wards were identified to review their charts in their last 22 days before death. In the second part, a total of 19 late-stage cancer patients with clear consciousness admitted to non-hospice wards were identified to investigate their quality of life for the final 2 weeks before death.

RESULTS

The utilization rate of HSC was 55.6%. Among these, the rate for late referral to HSC (≤7 days before death) was 43.8% and early referral (>3 months before death) was 5.6%. Compared to the non-HSC group, in the last few weeks of life, the HSC group underwent lower incidence of chemotherapy use (10.1% vs. 39.4%, p < .001), signed do-not-resuscitate orders (0% vs. 21.1%, p < .001), emergency room visits (13.5% vs. 40.8%, p < .001), intensive care unit admission or ventilator use (2.2% vs. 11.3%, p = .019), and endotracheal intubation (2.2% vs. 9.9%, p = .038). However, the quality of life did not appear to have obvious differences between the two groups (p > .05).

CONCLUSION

In Taiwan, late HSC referral in terminal cancer patients is common. HSC is associated with a reduced likelihood of aggressive medical utilization. However, the effect of HSC in improving patients' quality of life in the last few weeks needs to be further evaluated.

Exploring the effects of acceptable palliative care models on survival time and healthcare expenditure among patients with cancer: a national longitudinal population-based study

OBJECTIVE

Hospice care ensures better end-of-life quality by relieving terminal symptoms. Prior research has indicated that hospice care could prolong survival and reduce end-of-life medical expenditures among patients with cancer. However, the dearth of studies on the effects of hospice care type and use sequence on survival time and end-of-life medical expenditures substantiates the need for investigation.

DATA SOURCES AND STUDY SETTING

Two million random records were obtained from the National Health Insurance Research Database.

STUDY DESIGN

We estimated the effects of the type and sequence of hospice care use on survival time and medical expenditures among advanced cancer patients. This was a cross-sectional study.

DATA COLLECTION/EXTRACTION METHODS

Patient data were collected from 2 million random records provided by the National Health Insurance Research Database of Taiwan. We included people with cancer and excluded patients under 20 years of age; 2860 patients remained after matching.

PRINCIPAL FINDINGS

The results indicated that the average survival time of patients who received inpatient palliative care (1022 days) was significantly shorter than that of patients who did not receive palliative care (P < 0.001), but the health care expenditure during the entire course of cancer therapy was not the lowest. Interestingly, patients who received inpatient palliative care had the lowest health care expenditure at 1 year or month before the end of life (P < 0.001).

CONCLUSION

The type and sequence of palliative care affected the survival time and health care expenditures of cancer patients. Receiving palliative care did not prolong survival but rather reduced health care expenditures. The sequence of receiving palliative care significantly affected health care expenditures.

Characteristics and Procedures Among Adults Discharged to Hospice After Gastrointestinal Tract Surgery in California

IMPORTANCE

Hospice care is associated with improved quality of life and goal-concordant care. Limited data suggest that provision of hospice services after surgery is suboptimal; however, literature in this domain is in its nascency, leaving gaps in our understanding of patients who enroll in hospice after surgery.

OBJECTIVE

To characterize the transition to hospice after gastrointestinal tract surgery and identify areas that warrant further attention and intervention.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective cohort study included patients discharged to hospice after a surgical hospitalization for a digestive disorder in California-licensed hospitals between January 1, 2015, and December 31, 2019. Data were analyzed from August 1 to November 30, 2021.

EXPOSURES

Patient age, race and ethnicity, principal language, payer, and Distressed Community Index (DCI).

MAIN OUTCOMES AND MEASURES

Admission type and most common diagnoses and procedures for surgical hospitalizations that resulted in discharge to hospice, annual hospitalization trend for 3 years preceding hospice enrollment, and most common diagnoses for patients who were readmitted after hospice enrollment were summarized. Age, race and ethnicity, principal language, payer, and DCI were compared between patients who were readmitted after hospice enrollment and those who were not.

RESULTS

Of 2688 patients with surgical hospitalizations resulting in discharge to hospice (mean [SD] age, 73.2 [14.7] years; 1459 women [54.3%]), 2389 (88.9%) had urgent or emergent discharges. The most common diagnoses were cancer (primary and metastatic; 1541 [57.3%]) and bowel obstruction (563 [20.9%]). The most common procedures were bowel resection, fecal diversion, inferior vena cava filter, gastric bypass, and paracentesis. In the 3 years preceding hospice enrollment, this cohort had a mean (SD) of 2.21 (2.77) hospitalizations per patient (1537 of 5953 surgical [25.8%]). Of these, 3594 of 5953 total (60.4%) and 840 of 1537 surgical (54.7%) hospitalizations were within 1 year of hospice enrollment. Three hundred and sixty-eight patients (13.7%) were readmitted after hospice enrollment, with infection being the most common readmission diagnosis. Readmitted patients were more likely to be younger (mean [SD] age, 69.7 [16.4] vs 73.8 [14.3] years; P < .001), to speak a principal language other than English (62 of 368 [16.8%] vs 292 of 2320 [12.6%]; P = .02), to be insured through Medicaid (70 of 368 [19.0%] vs 223 of 2320 [9.6%]; P < .001), and to be from a community with higher DCI (198 of 360 [55.0%] vs 1117 of 2269 [49.2%]; P = .04) and were less likely to be White (195 of 368 [53.0%] vs 1479 of 2320 [63.8%]; P < .001).

CONCLUSIONS AND RELEVANCE

These findings suggest multiple opportunities for advance care planning in this surgical cohort, with a particular focus on emergent care. Further study is needed to understand the reasons for rehospitalization after hospice discharge and identify ways to improve communication and decision-making support for patients who choose to enroll in hospice care. Given the frequent antecedent interactions with the health care system among this population, longitudinal and tailored approaches may be beneficial to promote equitable end-of-life care; however, further research is needed to clarify barriers and understand differing patient needs.

Emergency Department Utilization Following Hepatopancreatic Surgery Among Medicare Beneficiaries

BACKGROUND

Care delivered in hospital-based emergency departments (ED) is a target for cost savings. ED utilization following hepatopancreatic surgery remains poorly defined. We sought to define the rate of ED utilization following liver and pancreatic resection, as well as to identify factors associated with ED visits post-discharge.

METHODS

The Medicare 100% Standard Analytic Files were used to identify Medicare beneficiaries who underwent hepatectomy or pancreatectomy between 2013 and 2017. Claims associated with ED services were identified using the relevant Revenue Center Codes. Patient characteristics and postoperative outcomes associated with ED care within 30 days of discharge were investigated.

RESULTS

Among 37,707 patients who underwent hepatopancreatic surgery, 10,323 (27.4%) had at least one ED visit within 30 days of discharge. Patients presenting to the ED were more likely to be male (OR 1.13, 95%CI 1.07-1.18). Patients undergoing a pancreatectomy (OR 1.39, 95%CI 1.32-1.47), as well as patients who had a perioperative complication (OR 1.16, 95%CI 1.10-1.23) and patients not discharged home (OR 1.41, 95%CI 1.33-1.49), were more likely to require ED care. In contrast, patients undergoing resection for cancer or surgery for an elective basis were less likely to present to the ED postoperatively (OR 0.92, 95%CI 0.87-0.97 and OR 0.22, 95%CI 0.20-0.23, respectively). Patients often had multiple ED visits within 30 days of discharge as 37.2% of patients presented to the ED with at least 2 visits. Visits were also most common in the immediate postoperative period, with 30.9% of ED visits taking place in the first 2 days from discharge. Among patients requiring postoperative ED care, 53.9% were readmitted within 30 days.

CONCLUSION

More than 1 in 4 patients undergoing hepatopancreatic surgery presented to the ED within 30 days of discharge, with most patients returning to the ED within the first week of discharge. A subset of patients had multiple ED visits. Future efforts should target patients most likely to be high ED utilizers to avoid the need for early post-discharge ED use.

End-of-Life Hospice Use and Medicare Expenditures Among Patients Dying of Hepatocellular Carcinoma

BACKGROUND

The increasing incidence of hepatocellular carcinoma (HCC) coupled with rising health care costs contributes to high end-of-life expenditures. The current study aimed to characterize health care expenditures and hospice use among patients with HCC using a large, national database.

METHODS

The Surveillance, Epidemiology, and End Results (SEER)-Medicare-linked database was used to identify patients with HCC. Logistic regression was used to identify factors associated with overall hospice use and end-of-life expenditures among individuals who died of HCC.

RESULTS

Among 14,369 Medicare beneficiaries with HCC, 8069 (63.7 %) used hospice. Racial/ethnic minority patients were less likely to use hospice services during the last year of life than white patients (no hospice: n = 2034 [44.3 %] vs. hospice: n = 2513 [31.1 %]). Social vulnerability also had an impact on the likelihood of patients using hospice services; in particular, the probability of hospice use among patients declined as social vulnerability increased (P < 0.05). Hospice use was associated with an approximate $10,000 decrease in inpatient expenditures (hospice: US$7900 [IQR, US$0-26,600] vs. no hospice: US$18,000 [IQR $400-49,100]; P < 0.001) and $1300 decrease in outpatient expenditures (hospice: US$900 [IQR, US$0-4500] vs. non-hospice: US$2200 [IQR, US$200-7900; P < 0.001) compared with individuals who did not use hospice.

CONCLUSIONS

Minority patients and individuals residing in high-vulnerability areas were less likely to use hospice. Patients who used hospice at the end of life had a reduction in inpatient and outpatient Medicare claims. Patients with HCC in need of hospice services should be ensured timely referral regardless of race/ethnicity or social vulnerability.

Race/Ethnicity and County-Level Social Vulnerability Impact Hospice Utilization Among Patients Undergoing Cancer Surgery

BACKGROUND

Integration of palliative care services into the surgical treatment plan is important for holistic patient care. We sought to examine the association between patient race/ethnicity and county-level vulnerability relative to patterns of hospice utilization.

PATIENTS AND METHODS

Medicare Standard Analytic Files were used to identify patients undergoing lung, esophageal, pancreatic, colon, or rectal cancer surgery between 2013 and 2017. Data were merged with the Centers for Disease Control and Prevention's social vulnerability index (SVI). Logistic regression was utilized to identify factors associated with overall hospice utilization among deceased individuals.

RESULTS

A total of 54,256 Medicare beneficiaries underwent lung (n = 16,645, 30.7%), esophageal (n = 1427, 2.6%), pancreatic (n = 6183, 11.4%), colon (n = 26,827, 49.4%), or rectal (n = 3174, 5.9%) cancer resection. Median patient age was 76 years (IQR 71-82 years), and 28,887 patients (53.2%) were male; the majority of individuals were White (91.1%, n = 49,443), while a smaller subset was Black or Latino (racial/ethnic minority: n = 4813, 8.9%). Overall, 35,416 (65.3%) patients utilized hospice services prior to death. Median SVI was 52.8 [interquartile range (IQR) 30.3-71.2]. White patients were more likely to utilize hospice care compared with minority patients (OR 1.24, 95% CI 1.17-1.31, p < 0.001). Unlike White patients, there was reduced odds of hospice utilization (OR 0.97, 95% CI 0.96-0.99) and early hospice initiation (OR 0.94, 95% CI 0.91-0.97) as SVI increased among minority patients.

CONCLUSIONS

Patients residing in counties with high social vulnerability were less likely to be enrolled in hospice care at the time of death, as well as be less likely to initiate hospice care early. The effects of increasing social vulnerability on hospice utilization were more profound among minority patients.