Melissa D. Aldridge

Methods for Constructing and Assessing Propensity Scores

OBJECTIVES To model the steps involved in preparing for and carrying out propensity score analyses by providing step-by-step guidance and Stata code applied to an empirical dataset. STUDY DESIGN Guidance, Stata code, and empirical examples are given to illustrate (1) the process of choosing variables to include in the propensity score; (2) balance of propensity score across treatment and comparison groups; (3) balance of covariates across treatment and comparison groups within blocks of the propensity score; (4) choice of matching and weighting strategies; (5) balance of covariates after matching or weighting the sample; and (6) interpretation of treatment effect estimates. EMPIRICAL APPLICATION We use data from the Palliative Care for Cancer Patients (PC4C) study, a multisite observational study of the effect of inpatient palliative care on patient health outcomes and health services use, to illustrate the development and use of a propensity score. CONCLUSIONS Propensity scores are one useful tool for accounting for observed differences between treated and comparison groups. Careful testing of propensity scores is required before using them to estimate treatment effects.

Education, implementation, and policy barriers to greater integration of palliative care: A literature review

Background: Early integration of palliative care into the management of patients with serious disease has the potential to both improve quality of life of patients and families and reduce healthcare costs. Despite these benefits, significant barriers exist in the United States to the early integration of palliative care in the disease trajectory of individuals with serious illness. Aim: To provide an overview of the barriers to more widespread palliative care integration in the United States. Design and data sources: A literature review using PubMed from 2005 to March 2015 augmented by primary data collected from 405 hospitals included in the Center to Advance Palliative Care’s National Palliative Care Registry for years 2012 and 2013. We use the World Health Organization’s Public Health Strategy for Palliative Care as a framework for analyzing barriers to palliative care integration. Results: We identified key barriers to palliative care integration across three World Health Organization domains: (1) education domain: lack of adequate education/training and perception of palliative care as end-of-life care; (2) implementation domain: inadequate size of palliative medicine–trained workforce, challenge of identifying patients appropriate for palliative care referral, and need for culture change across settings; (3) policy domain: fragmented healthcare system, need for greater funding for research, lack of adequate reimbursement for palliative care, and regulatory barriers. Conclusion: We describe the key policy and educational opportunities in the United States to address and potentially overcome the barriers to greater integration of palliative care into the healthcare of Americans with serious illness.

Has Hospice Use Changed? 2000-2010 Utilization Patterns.

Background:Hospice use has increased substantially during the past decade by an increasingly diverse patient population; however, little is known about patterns of hospice use and how these patterns have changed during the past decade. Objective:To characterize Medicare hospice users in 2000 and 2010 and estimate the prevalence of (1) very short (⩽1 wk) hospice enrollment; (2) very long (>6 mo) hospice enrollment; and (3) hospice disenrollment and how these utilization patterns have varied over time and by patient and hospice characteristics. Research Design:Cross-sectional analysis of Medicare hospice claims data from 2000 and 2010. Subjects:All US Medicare Hospice Benefit enrollees in 2000 (N=529,573) and 2010 (N=1,150,194). Results:As of 2010, more than half (53.4%) of all Medicare decedents who used hospice had either very short (⩽1 wk, 32.4%) or very long (>6 mo, 13.9%) hospice enrollment or disenrolled from hospice before death (10.6%). This represents an increase of 4.9 percentage points from 2000. In multivariable analysis, patients with noncancer diagnoses, the fastest growing group of hospice users, were approximately twice as likely as those with cancer to have very short or long enrollment periods and to disenroll from hospice. Conclusion:The substantial proportion of hospice users with very short or long enrollment, or enrollments that end before death, underscores the potential for interventions to improve the timing and appropriateness of hospice referral so that the full benefits of hospice are received by patients and families.

Transitions Between Healthcare Settings of Hospice Enrollees at the End of Life

To characterize the number and types of care transitions in the last 6 months of life of individuals who used hospice and to examine factors associated with having multiple transitions in care.

National hospice survey results: for-profit status, community engagement, and service.

IMPORTANCE The impact of the substantial growth in for-profit hospices in the United States on quality and hospice access has been intensely debated, yet little is known about how for-profit and nonprofit hospices differ in activities beyond service delivery. OBJECTIVE To determine the association between hospice ownership and (1) provision of community benefits, (2) setting and timing of the hospice population served, and (3) community outreach. DESIGN, SETTING, AND PARTICIPANTS Cross-sectional survey (the National Hospice Survey), conducted from September 2008 through November 2009, of a national random sample of 591 Medicare-certified hospices operating throughout the United States. EXPOSURES For-profit or nonprofit hospice ownership. MAIN OUTCOMES AND MEASURES Provision of community benefits; setting and timing of the hospice population served; and community outreach. RESULTS A total of 591 hospices completed our survey (84% response rate). For-profit hospices were less likely than nonprofit hospices to provide community benefits including serving as training sites (55% vs 82%; adjusted relative risk [ARR], 0.67 [95% CI, 0.59-0.76]), conducting research (18% vs 23%; ARR, 0.67 [95% CI, 0.46-0.99]), and providing charity care (80% vs 82%; ARR, 0.88 [95% CI, 0.80-0.96]). For-profit compared with nonprofit hospices cared for a larger proportion of patients with longer expected hospice stays including those in nursing homes (30% vs 25%; P = .009). For-profit hospices were more likely to exceed Medicares aggregate annual cap (22% vs 4%; ARR, 3.66 [95% CI, 2.02-6.63]) and had a higher patient disenrollment rate (10% vs 6%; P < .001). For-profit were more likely than nonprofit hospices to engage in outreach to low-income communities (61% vs 46%; ARR, 1.23 [95% CI, 1.05-1.44]) and minority communities (59% vs 48%; ARR, 1.18 [95% CI, 1.02-1.38]) and less likely to partner with oncology centers (25% vs 33%; ARR, 0.59 [95% CI, 0.44-0.80]). CONCLUSIONS AND RELEVANCE Ownership-related differences are apparent among hospices in community benefits, population served, and community outreach. Although Medicares aggregate annual cap may curb the incentive to focus on long-stay hospice patients, additional regulatory measures such as public reporting of hospice disenrollment rates should be considered as the share of for-profit hospices in the United States continues to increase.

Association Between Hospice Use and Depressive Symptoms in Surviving Spouses

IMPORTANCE Family caregivers of individuals with serious illness are at risk for depressive symptoms and depression. Hospice includes the provision of support services for family caregivers, yet evidence is limited regarding the effect of hospice use on depressive symptoms among surviving caregivers. OBJECTIVE To determine the association between hospice use and depressive symptoms in surviving spouses. DESIGN, SETTING, AND PARTICIPANTS We linked data from the Health and Retirement Study, a nationally representative longitudinal survey of community-dwelling US adults 50 years or older, to Medicare claims. Participants included a propensity score-matched sample of 1016 Health and Retirement Study decedents with at least 1 serious illness and their surviving spouses interviewed between August 2002 and May 2011. We compared the spouses of individuals enrolled in hospice with the spouses of individuals who did not use hospice, performing our analysis between January 30, 2014, and January 16, 2015. EXPOSURES Hospice enrollment for at least 3 days in the year before death. MAIN OUTCOMES AND MEASURES Spousal depressive symptom scores measured 0 to 2 years after death with the Center for Epidemiologic Studies Depression Scale, which is scored from 0 (no symptoms) to 8 (severe symptoms). RESULTS Of the 1016 decedents in the matched sample, 305 patients (30.0%) used hospice services for 3 or more days in the year before death. Of the 1016 spouses, 51.9% had more depressive symptoms over time (mean [SD] change, 2.56 [1.65]), with no significant difference related to hospice use. A minority (28.2%) of spouses of hospice users had improved Center for Epidemiologic Studies Depression Scale scores compared with 21.7% of spouses of decedents who did not use hospice, although the difference was not statistically significant (P = .06). Among the 662 spouses who were the primary caregivers, 27.3% of spouses of hospice users had improved Center for Epidemiologic Studies Depression Scale scores compared with 20.7% of spouses of decedents who did not use hospice; the difference was not statistically significant (P = .10). In multivariate analysis, the odds ratio for the association of hospice enrollment with improved depressive symptoms after the spouses death was 1.63 (95% CI, 1.00-2.65). CONCLUSIONS AND RELEVANCE After bereavement, depression symptoms increased overall for surviving spouses regardless of hospice use. A modest reduction in depressive symptoms was more likely among spouses of hospice users than among spouses of nonhospice users.

The impact of reported hospice preferred practices on hospital utilization at the end of life

Background:The Affordable Care Act requires hospices to report quality measures across a range of processes and practices. Yet uncertainties exist regarding the impact of hospice preferred practices on patient outcomes. Objective:Assess the impact of 6 hospice preferred practices and hospice organizational characteristics on hospital utilization and death using the first national data on hospice preferred practices. Design:Longitudinal cohort study (2008–2011) of Medicare beneficiaries (N=149,814) newly enrolled in a national random sample of hospices (N=577) from the National Hospice Survey (84% response rate) and followed until death. Outcome Measures:The proportion of patients at each hospice admitted to the hospital, emergency department (ED), and intensive care unit (ICU), and who died in the hospital after hospice enrollment. Results:Hospices that reported assessing patient preferences for site of death at admission had lower odds of being in the highest quartile for hospital death (AOR=0.36; 95% CI, 0.14–0.93) and ED visits (AOR=0.27; 95% CI, 0.10–0.76). Hospices that reported more frequently monitoring symptoms had lower odds of being in the highest quartile for ICU stays (AOR=0.48; 95% CI, 0.24–0.94). In adjusted analyses, a higher proportion of patients at for-profit compared with nonprofit hospices experienced a hospital admission (15.3% vs. 10.9%, P<0.001), ED visit (21.8% vs. 15.6%, P<0.001), and ICU stay (5.1% vs. 3.0%, P<0.001). Conclusions:Hospitalization of patients following hospice enrollment varies substantially across hospices. Two of the 6 preferred practices examined were associated with hospitalization rates and for-profit hospices had persistently high hospitalization rates regardless of preferred practice implementation.

Hospice Services for Complicated Grief and Depression: Results from a National Survey

To describe the prevalence of screening for complicated grief (CG) and depression in hospice and access to bereavement therapy and to examine whether screening and access to therapy varied according to hospice organizational characteristics or staff training and involvement.

End-of-Life Care Transition Patterns of Medicare Beneficiaries

To characterize the patterns of transitions in care and factors associated with multiple transitions in the last 6 months of life of U.S. decedents (N = 660,132).

Prospective cohort study of hospitalized adults with advanced cancer: Associations between complications, comorbidity, and utilization

BACKGROUND: Inpatient hospital stays account for more than a third of direct medical cancer care costs. Evidence on factors driving these costs can inform planning of services, as well as consideration of equity in access. OBJECTIVE: To measure the association between hospital costs, and demographic, clinical, and system factors, for a cohort of adults with advanced cancer. DESIGN: Prospective multisite cohort study. SETTING: Four medical and cancer centers. PATIENTS: Adults with advanced cancer admitted to a participating hospital between 2007 and 2011, excluding those with dementia. Final analytic sample included 1020 patients. METHODS: With receipt of palliative care controlled for, the associations between hospital cost and patient factors were estimated. Factors covered the domains of demographics (age, sex, race), socioeconomics and systems (education, insurance, living will, proxy), clinical care (diagnoses, complications deemed to pose a threat to life or bodily functions, comorbidities, symptom burden, activities of daily living), and prior healthcare utilization (home help, analgesic prescribing). OUTCOME MEASURE: Direct hospital costs. RESULTS: A major (markedly abnormal) complication (+