BACKGROUND: Previous studies have documented that hospice enrollment by terminally ill cancer patients varies substantially by patient characteristics and across broad geographic regions, but little is known about how local practice patterns and individual physicians contribute to these variations. We examined hospice use within a regional integrated health care delivery system that provides consistent insurance coverage and hospice availability for its members to evaluate the relative importance of patient characteristics, physician characteristics, individual physicians, and local health centers in explaining variations in hospice enrollment. METHODS: We examined data for 3805 Kaiser Permanente of Northern California health plan enrollees who were diagnosed with and died of lung, colorectal, breast, or prostate cancer from January 1, 1996, through June 30, 2001. We used a random-effects linear probability hierarchical model to estimate adjusted hospice enrollment rates and identify factors associated with hospice enrollment. All P values are two-sided. RESULTS: Overall, 65.4% of patients enrolled in hospice care before death. In adjusted analyses, hospice enrollment did not vary by patients' race/ethnicity or marital status (all P>.2) but varied substantially among the 11 health centers where patients were treated (standard deviation [SD] of random effect = 10.0 percentage points, corresponding to an estimated adjusted hospice enrollment rate for two-thirds of centers (2 SDs) ranging from 55% to 75%, P = .02). Hospice enrollment varied less among the 675 individual physicians (SD = 4.6 percentage points; P = .09). CONCLUSIONS: Health care within a large integrated delivery system has the potential to eliminate racial and ethnic disparities in hospice use, but substantial variation in hospice use persists among local health centers. Focused efforts to understand how patients, physicians, and hospices interact at the local level are important to ensure equal access to hospice care for all terminally ill cancer patients.
BACKGROUND: Previous studies have documented that hospice enrollment by terminally ill cancerpatients varies substantially by patient characteristics and across broad geographic regions, but little is known about how local practice patterns and individual physicians contribute to these variations. We examined hospice use within a regional integrated health care delivery system that provides consistent insurance coverage and hospice availability for its members to evaluate the relative importance of patient characteristics, physician characteristics, individual physicians, and local health centers in explaining variations in hospice enrollment. METHODS: We examined data for 3805 Kaiser Permanente of Northern California health plan enrollees who were diagnosed with and died of lung, colorectal, breast, or prostate cancer from January 1, 1996, through June 30, 2001. We used a random-effects linear probability hierarchical model to estimate adjusted hospice enrollment rates and identify factors associated with hospice enrollment. All P values are two-sided. RESULTS: Overall, 65.4% of patients enrolled in hospice care before death. In adjusted analyses, hospice enrollment did not vary by patients' race/ethnicity or marital status (all P>.2) but varied substantially among the 11 health centers where patients were treated (standard deviation [SD] of random effect = 10.0 percentage points, corresponding to an estimated adjusted hospice enrollment rate for two-thirds of centers (2 SDs) ranging from 55% to 75%, P = .02). Hospice enrollment varied less among the 675 individual physicians (SD = 4.6 percentage points; P = .09). CONCLUSIONS: Health care within a large integrated delivery system has the potential to eliminate racial and ethnic disparities in hospice use, but substantial variation in hospice use persists among local health centers. Focused efforts to understand how patients, physicians, and hospices interact at the local level are important to ensure equal access to hospice care for all terminally ill cancerpatients.
Authors: Claire Selsky; Barbara Kreling; Gheorghe Luta; Solomon B Makgoeng; Jessika Gomez-Duarte; Andrea Gabriela A Barbo; Jeanne S Mandelblatt Journal: J Palliat Med Date: 2012-06-25 Impact factor: 2.947
Authors: Kristin M Sheffield; Casey A Boyd; Jamie Benarroch-Gampel; Yong-Fang Kuo; Catherine D Cooksley; Taylor S Riall Journal: Cancer Date: 2011-04-14 Impact factor: 6.860
Authors: Elie K Mehanna; Paul J Catalano; Daniel N Cagney; Daphne A Haas-Kogan; Brian M Alexander; James A Tulsky; Ayal A Aizer Journal: J Natl Cancer Inst Date: 2020-12-14 Impact factor: 13.506
Authors: Paul K Maciejewski; Andrea C Phelps; Elizabeth L Kacel; Tracy A Balboni; Michael Balboni; Alexi A Wright; William Pirl; Holly G Prigerson Journal: Psychooncology Date: 2011-03-29 Impact factor: 3.894