- Email: [email protected]
CHALLENGES IN ESTIMATING THE HOSPITALIZATION RATE FOR HEART FAILURE
NKF 2015 Spring Clinical Meetings Abstracts
205 CHALLENGES IN ESTIMATING THE HOSPITALIZATION RATE FOR HEART FAILURE: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN Chief Medical Officers of 14 US dialysis providers recently proposed a “Volume First” approach addressing fluid intake and volume control (DE Weiner et al, AJKD, 2014). An important metric of performance in this domain is the hospitalization rate for heart failure, as ascertained from Medicare claims. Measuring progress with this metric requires careful consideration of definitions, in light of the difficulty associated with distinguishing between overt heart failure (HF), fluid overload (FO), and pleural effusion (PE) at bedside, as well as coding shifts due to evolution of the Medicare Hospital Inpatient Prospective Payment System. We estimated rates of hospitalization for, HF, FO, and PE in prevalent dialysis patients from 2004 to 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For each year, we identified Medicare beneficiaries undergoing dialysis in a freestanding facility on January 1 and followed them until kidney transplant, death, or December 31. We identified admissions for HF and FO/PE from principal discharge diagnoses on Medicare inpatient facility claims. Between 2004 and 2011, admission rates for HF fell by 23%, from 16.2 to 12.4 admissions per 100 Admissions per 100 PY patient-years (PY). Concurrently, HF FO/PE All 2004 16.2 2.1 18.3 rates for FO/PE increased 157%, 2005 15.9 2.2 18.1 from 2.1 to 5.4 admissions per 100 2006 15.0 2.3 17.3 2007 14.6 3.0 17.6 PY. Rates for the composite of HF 2008 13.4 4.1 17.5 and FO/PE were unchanged, until 2009 13.5 4.6 18.0 2010 13.0 5.3 18.3 falling modestly from 2010 to 2011. 2011 12.4 5.4 17.7 Although the hospitalization rate for HF in dialysis patients has ostensibly fallen, the hospitalization rate for sequelae of excess extracellular volume has risen commensurately. Measuring progress may require a broad definition of heart failure.
206 RECENT DECREASE IN DEATH RATES AMONG DIALYSIS PATIENTS: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN The rate of death in US dialysis patients has historically exceeded the rate of death in the general population. Because the majority of patients are seen by outpatient dialysis providers three times per week, risk reduction likely depends on continuous quality improvement in disease management by dialysis providers. We assessed trends in rates of death among incident and prevalent patients treated in freestanding dialysis facilities between 1996 and 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For incident cohorts per annum, we identified patients with first outpatient dialysis within 3 months of chronic dialysis initiation and in a freestanding facility. For prevalent cohorts per annum, we identified patients undergoing dialysis in a freestanding facility on January 1. In both cohorts, we followed patients until recovery of renal function, kidney transplant, interruption of outpatient dialysis for > 3 months, or death; and maximum follow-up duration for each cohort was 1 year. Mortality rates were estimated without statistical adjustment for differences in case mix. Among incident patients, mortality rates increased by > 10% between 1996 and 2003, from 27.8 to 30.6 deaths per 100 patient-years (PY). From 2003 to 2011, the trend reversed: rates decreased by > 15%, from 30.6 to 26.0 deaths per 100 PY. Among prevalent patients, mortality rates were unchanged between 1996 (deaths per 100 PY, 22.4) and 2003 (22.5). In contrast, mortality rates decreased by > 19% from 2003 to 2011, from 22.5 to 18.3 deaths per 100 PY. By comparison, crude and age-adjusted mortality rates in the US population decreased by only 4.3% and 12.1%, respectively, between 2003 and 2011. Since 2003, the rate of death among dialysis patients in freestanding facilities has decreased markedly and more rapidly than in the general population, but the rate remains high and further progress is needed.
A66
207 RECENT DECREASE IN HOSPITALIZATION RATES AMONG DIALYSIS PATIENTS: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN Hospitalization imposes significant burdens on the quality of life of dialysis patients and on the US Treasury. Admissions of patients are often attributable to cardiovascular (CV) morbidity and infection, the risks of which outpatient dialysis providers may modify. We assessed trends in rates of hospitalization among incident and prevalent patients treated in freestanding dialysis facilities between 1996 and 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For incident cohorts per annum, we identified patients with first outpatient dialysis within 3 months of chronic dialysis initiation and in a freestanding facility. For prevalent cohorts per annum, we identified patients undergoing dialysis in a freestanding facility on January 1. Only patients with Medicare as primary payer were retained. In both cohorts, we followed patients until recovery of renal function, kidney transplant, interruption of outpatient dialysis for > 3 months, or death; and maximum follow-up duration for each cohort was 1 year. Hospitalization rates were estimated without statistical adjustment for differences in case mix. Among incident patients, admission rates increased slightly between 1996 and 2005, from 2.54 to 2.57 admissions per patient-year (PY). From 2005 to 2011, rates fell by > 8%, from 2.57 to 2.36 admissions per PY. Among prevalent patients, admissions rates were unchanged between 1996 (admissions per PY, 2.01) and 2005 (2.02). In contrast, admissions rates decreased by > 8% between 2005 and 2011, from 2.02 to 1.85 admissions per PY, in alignment with an 8.1% decrease among all Medicare beneficiaries. Between 2005 and 2011, admissions due to CV morbidity and infection decreased by 13% and 5%, respectively. Hospitalization rates among dialysis patients in freestanding facilities have decreased since 2005. Differences in rates of decline by cause of admission illustrate opportunities for further progress.
208 REGIONAL VARIATION IN MORTALITY RATES OF INCIDENT DIALYSIS PATIENTS: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN Both the quantity and quality of health care resources tend to vary regionally, but relatively little is understood about regional variation in dialysis patient outcomes. We assessed regional variation in mortality rates during the first year of dialysis, as rates are particularly high and thus constitute an important target for quality improvement, with focus on trends in rates between 2004 and 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For incident cohorts per annum, we identified patients with first outpatient dialysis within 3 months of chronic dialysis initiation and in a freestanding facility. We followed patients until recovery of renal function, kidney transplant, interruption of outpatient dialysis for > 3 months, or death; and maximum follow-up duration for each cohort was 1 year. Patients were stratified by US Census Division, on the basis of the state of initial outpatient dialysis facility. In 2011, Division mortality rates differed by 46% from minimum to maximum. The highest mortality rate (deaths per 100 patient-years, 30.9) was observed in the East North Central (IL, IN, OH, MI, WI) states; the lowest rates (21.1 and 22.2) were observed in the Mountain (AZ, CO, ID, MT, NM, NV, UT, WY) and Pacific (AK, CA, HI, OR, WA) states, respectively. Between 2004 and 2011, annual percent changes in mortality rates ranged from +0.4% per year in the East North Central to -3.4% per year in the West South Central (AR, LA, OK, TX) states. Among the East North Central states, mortality rates tended to increase annually in Illinois, Indiana, and Ohio. There is substantial regional variation in mortality rates among incident dialysis patients. Some variation likely represents differences in care before chronic dialysis initiation. However, the secular increases in mortality rates in some areas warrant further attention.
Am J Kidney Dis. 2015;65(4):A1-A93
205 CHALLENGES IN ESTIMATING THE HOSPITALIZATION RATE FOR HEART FAILURE: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN Chief Medical Officers of 14 US dialysis providers recently proposed a “Volume First” approach addressing fluid intake and volume control (DE Weiner et al, AJKD, 2014). An important metric of performance in this domain is the hospitalization rate for heart failure, as ascertained from Medicare claims. Measuring progress with this metric requires careful consideration of definitions, in light of the difficulty associated with distinguishing between overt heart failure (HF), fluid overload (FO), and pleural effusion (PE) at bedside, as well as coding shifts due to evolution of the Medicare Hospital Inpatient Prospective Payment System. We estimated rates of hospitalization for, HF, FO, and PE in prevalent dialysis patients from 2004 to 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For each year, we identified Medicare beneficiaries undergoing dialysis in a freestanding facility on January 1 and followed them until kidney transplant, death, or December 31. We identified admissions for HF and FO/PE from principal discharge diagnoses on Medicare inpatient facility claims. Between 2004 and 2011, admission rates for HF fell by 23%, from 16.2 to 12.4 admissions per 100 Admissions per 100 PY patient-years (PY). Concurrently, HF FO/PE All 2004 16.2 2.1 18.3 rates for FO/PE increased 157%, 2005 15.9 2.2 18.1 from 2.1 to 5.4 admissions per 100 2006 15.0 2.3 17.3 2007 14.6 3.0 17.6 PY. Rates for the composite of HF 2008 13.4 4.1 17.5 and FO/PE were unchanged, until 2009 13.5 4.6 18.0 2010 13.0 5.3 18.3 falling modestly from 2010 to 2011. 2011 12.4 5.4 17.7 Although the hospitalization rate for HF in dialysis patients has ostensibly fallen, the hospitalization rate for sequelae of excess extracellular volume has risen commensurately. Measuring progress may require a broad definition of heart failure.
206 RECENT DECREASE IN DEATH RATES AMONG DIALYSIS PATIENTS: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN The rate of death in US dialysis patients has historically exceeded the rate of death in the general population. Because the majority of patients are seen by outpatient dialysis providers three times per week, risk reduction likely depends on continuous quality improvement in disease management by dialysis providers. We assessed trends in rates of death among incident and prevalent patients treated in freestanding dialysis facilities between 1996 and 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For incident cohorts per annum, we identified patients with first outpatient dialysis within 3 months of chronic dialysis initiation and in a freestanding facility. For prevalent cohorts per annum, we identified patients undergoing dialysis in a freestanding facility on January 1. In both cohorts, we followed patients until recovery of renal function, kidney transplant, interruption of outpatient dialysis for > 3 months, or death; and maximum follow-up duration for each cohort was 1 year. Mortality rates were estimated without statistical adjustment for differences in case mix. Among incident patients, mortality rates increased by > 10% between 1996 and 2003, from 27.8 to 30.6 deaths per 100 patient-years (PY). From 2003 to 2011, the trend reversed: rates decreased by > 15%, from 30.6 to 26.0 deaths per 100 PY. Among prevalent patients, mortality rates were unchanged between 1996 (deaths per 100 PY, 22.4) and 2003 (22.5). In contrast, mortality rates decreased by > 19% from 2003 to 2011, from 22.5 to 18.3 deaths per 100 PY. By comparison, crude and age-adjusted mortality rates in the US population decreased by only 4.3% and 12.1%, respectively, between 2003 and 2011. Since 2003, the rate of death among dialysis patients in freestanding facilities has decreased markedly and more rapidly than in the general population, but the rate remains high and further progress is needed.
A66
207 RECENT DECREASE IN HOSPITALIZATION RATES AMONG DIALYSIS PATIENTS: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN Hospitalization imposes significant burdens on the quality of life of dialysis patients and on the US Treasury. Admissions of patients are often attributable to cardiovascular (CV) morbidity and infection, the risks of which outpatient dialysis providers may modify. We assessed trends in rates of hospitalization among incident and prevalent patients treated in freestanding dialysis facilities between 1996 and 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For incident cohorts per annum, we identified patients with first outpatient dialysis within 3 months of chronic dialysis initiation and in a freestanding facility. For prevalent cohorts per annum, we identified patients undergoing dialysis in a freestanding facility on January 1. Only patients with Medicare as primary payer were retained. In both cohorts, we followed patients until recovery of renal function, kidney transplant, interruption of outpatient dialysis for > 3 months, or death; and maximum follow-up duration for each cohort was 1 year. Hospitalization rates were estimated without statistical adjustment for differences in case mix. Among incident patients, admission rates increased slightly between 1996 and 2005, from 2.54 to 2.57 admissions per patient-year (PY). From 2005 to 2011, rates fell by > 8%, from 2.57 to 2.36 admissions per PY. Among prevalent patients, admissions rates were unchanged between 1996 (admissions per PY, 2.01) and 2005 (2.02). In contrast, admissions rates decreased by > 8% between 2005 and 2011, from 2.02 to 1.85 admissions per PY, in alignment with an 8.1% decrease among all Medicare beneficiaries. Between 2005 and 2011, admissions due to CV morbidity and infection decreased by 13% and 5%, respectively. Hospitalization rates among dialysis patients in freestanding facilities have decreased since 2005. Differences in rates of decline by cause of admission illustrate opportunities for further progress.
208 REGIONAL VARIATION IN MORTALITY RATES OF INCIDENT DIALYSIS PATIENTS: Peer Kidney Care Initiative (CDRG and the Chief Medical Officers of 13 US Dialysis Providers), Minneapolis, MN Both the quantity and quality of health care resources tend to vary regionally, but relatively little is understood about regional variation in dialysis patient outcomes. We assessed regional variation in mortality rates during the first year of dialysis, as rates are particularly high and thus constitute an important target for quality improvement, with focus on trends in rates between 2004 and 2011. Data were ascertained from the Centers for Medicare & Medicaid Services End Stage Renal Disease database. For incident cohorts per annum, we identified patients with first outpatient dialysis within 3 months of chronic dialysis initiation and in a freestanding facility. We followed patients until recovery of renal function, kidney transplant, interruption of outpatient dialysis for > 3 months, or death; and maximum follow-up duration for each cohort was 1 year. Patients were stratified by US Census Division, on the basis of the state of initial outpatient dialysis facility. In 2011, Division mortality rates differed by 46% from minimum to maximum. The highest mortality rate (deaths per 100 patient-years, 30.9) was observed in the East North Central (IL, IN, OH, MI, WI) states; the lowest rates (21.1 and 22.2) were observed in the Mountain (AZ, CO, ID, MT, NM, NV, UT, WY) and Pacific (AK, CA, HI, OR, WA) states, respectively. Between 2004 and 2011, annual percent changes in mortality rates ranged from +0.4% per year in the East North Central to -3.4% per year in the West South Central (AR, LA, OK, TX) states. Among the East North Central states, mortality rates tended to increase annually in Illinois, Indiana, and Ohio. There is substantial regional variation in mortality rates among incident dialysis patients. Some variation likely represents differences in care before chronic dialysis initiation. However, the secular increases in mortality rates in some areas warrant further attention.
Am J Kidney Dis. 2015;65(4):A1-A93