Dataset Information

Hierarchical and joint site-edge methods for medicare hospice service region boundary analysis.

ABSTRACT: Hospice service offers a convenient and ethically preferable health-care option for terminally ill patients. However, this option is unavailable to patients in remote areas not served by any hospice system. In this article, we seek to determine the service areas of two particular cancer hospice systems in northeastern Minnesota based only on death counts abstracted from Medicare billing records. The problem is one of spatial boundary analysis, a field that appears statistically underdeveloped for irregular areal (lattice) data, even though most publicly available human health data are of this type. In this article, we suggest a variety of hierarchical models for areal boundary analysis that hierarchically or jointly parameterize both the areas and the edge segments. This leads to conceptually appealing solutions for our data that remain computationally feasible. While our approaches parallel similar developments in statistical image restoration using Markov random fields, important differences arise due to the irregular nature of our lattices, the sparseness and high variability of our data, the existence of important covariate information, and most importantly, our desire for full posterior inference on the boundary. Our results successfully delineate service areas for our two Minnesota hospice systems that sometimes conflict with the hospices' self-reported service areas. We also obtain boundaries for the spatial residuals from our fits, separating regions that differ for reasons yet unaccounted for by our model.

SUBMITTER: Ma H

PROVIDER: S-EPMC3061258 | biostudies-literature | 2010 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Hierarchical and joint site-edge methods for medicare hospice service region boundary analysis.

Ma Haijun H Carlin Bradley P BP Banerjee Sudipto S

Biometrics 20090723 2

Hospice service offers a convenient and ethically preferable health-care option for terminally ill patients. However, this option is unavailable to patients in remote areas not served by any hospice system. In this article, we seek to determine the service areas of two particular cancer hospice systems in northeastern Minnesota based only on death counts abstracted from Medicare billing records. The problem is one of spatial boundary analysis, a field that appears statistically underdeveloped fo ...[more]

PMID: 19645704

Similar Datasets

Project description:ImportanceOlder adults in Medicare Advantage (MA) enroll in hospice at higher rates than those in traditional Medicare (TM), but it is unclear whether the pathway of care prior to hospice use differs between MA and TM.ObjectiveTo examine the site of care prior to hospice enrollment for MA beneficiaries compared with those in TM.Design, setting, and participantsThis population-based, retrospective cross-sectional study used Medicare claims data for decedents in calendar years 2011, 2013, 2016, and 2018 who enrolled in hospice in the last 90 days of life. Data were analyzed from February 11, 2022, to October 24, 2022.ExposuresEnrollment in MA or TM in the last month of life.Main outcomes and measuresThe main outcome was the site of care prior to hospice enrollment, defined as hospital, nursing home, and home with or without home health, dichotomized as community vs hospital in a logistic regression model. Covariates included decedent demographics, hospice primary diagnosis, and county-level MA penetration. Differences in hospice length of stay between MA beneficiaries and TM beneficiaries were assessed using linear and logistic regression models.ResultsIn this study of 3 164 959 decedents, mean (SD) age was 83.1 (8.6) years, 55.8% were female, and 28.8% were enrolled in MA. Decedents in MA were more likely to enroll in hospice from a community setting than were those in TM, although the gap narrowed over time from an unadjusted 11.1% higher rate of community enrollment in MA vs TM in 2011 (50.1% vs 39.0%) to 8.1% in 2018 (46.4% vs 38.3%). In the primary adjusted analysis over the entire study period, MA enrollment was associated with an 8.09-percentage point (95% CI, 7.96-8.21 percentage points) higher rate of hospice enrollment from the community vs all other sites. This association remained in multiple sensitivity analyses to account for potential differences in the populations enrolled in MA vs TM. The mean overall hospice length of stay was 0.29 days (95% CI, 0.24-0.34 days) longer for MA decedents compared with TM decedents.Conclusions and relevanceCompared with TM beneficiaries, those in MA were more likely to enroll in hospice from community settings vs following inpatient stays. However, hospice length of stay was not substantially different between MA and TM. Further research is needed to understand how MA plans influence hospice use and the direct association with quality of end-of-life care as reported by older adults and their families.

Project description:BackgroundThe use of medications not relieving symptoms or maximizing quality of life should be minimized following hospice enrollment.ObjectiveTo evaluate the frequency of and predictive factors for continuation of medications with limited benefit after hospice admission among those admitted for cancer- and non-cancer-related causes.DesignCohort study using the Surveillance, Epidemiology and End Results-Medicare linked database.PatientsMedicare Part D-enrolled beneficiaries 66 years and older who were admitted to and died under hospice care between January 1, 2008, and December 31, 2013 (N = 70,035).Main measuresPatients were followed from hospice enrollment through death for Part D dispensing of limited benefit medications (LBMs) they had used in the 6 months prior to hospice admission, including anti-hyperlipidemics, anti-hypertensives, oral anti-diabetics, anti-platelets, anti-dementia medications, anti-osteoporotic medications, and proton pump inhibitors. The proportion of patients continuing an LBM after hospice admission was evaluated. Adjusted relative risks (RRs) were estimated for factors associated with LBM continuation.Key resultsOverall, 29.8% and 30.5% of patients admitted to hospice for a cancer- and non-cancer-related cause, respectively, continued at least one LBM after hospice admission. Anti-dementia medications were continued most frequently (29.3%) while anti-osteoporotic medications were continued least often (14.1%). Compared to home hospice, LBM continuation was greater in hospice patients residing in skilled nursing (RR 1.25, 95% CI 1.20-1.29), non-skilled nursing (RR 1.29, 95% CI 1.25-1.32), and assisted living facilities (RR 1.28, 95% CI 1.24-1.32). Patients with hospice stays ≥ 180 days were more likely to continue at least one LBM compared to those with stays of 1 week or less (RR 13.11, 95% CI 12.25-14.02).ConclusionsA substantial proportion of Medicare hospice beneficiaries continued to receive LBMs following hospice enrollment. Providers should evaluate the necessity of continuing non-palliative medications at the end of life through a careful, patient-centric consideration of their potential risks and benefits.

Dataset Information

Hierarchical and joint site-edge methods for medicare hospice service region boundary analysis.

Publications

Hierarchical and joint site-edge methods for medicare hospice service region boundary analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets