| Background Heart failure(HF)is a leading cause of hospitalization in China,which is experiencing a rapid increase in cardiovascular disease prevalence.Yet,little is known about current practice patterm,quality of care,and treatment outcomes of HF in China.Objectives Our study aimed(1)to characterize the contemporary national landscape of patients hospitalized for HF in China;(2)to evaluate HF care at hospital-and patient-level,and hospital variation in quality of care,and identify factors associated with quality of care;(3)to describe in-patient outcomes and quantify hospital variation in outcomes.Methods In a retrospective analysis of hospital records from 2015,we used two-stage random sampling to create a nationally representative cohort of admissions for HF in China.In the first stage,we used simple random-sampling stratified by economic-geographical regions to generate a list of participating hospitals.In the second stage,we systematically sampled cases admitted for HF from participating hospitals.Data on patient characteristics,management and outcomes were obtained through centralized medical record abstraction;data on hospital characteristics were obtained through hospital survey.We evaluated both process and outcome measures at the patient-and hospital-level.Process measures include adherence to 4 eore performance measures:(1)left ventricular ejection fraction(LVEF)assessment during hospitalization,(2)angiotensin-converting-enzyme inhibitors or angiotensin receptor blockers for eligible patients at discharge,(3)evidence-based P blocker(bisoprolol,carvedilol or metoprolol succinate)for eligible patients at discharge,(4)scheduled appointment at discharge);use of HF medications among patients with contraindications;utilization of diagnostic tests;specifically,we describe frequency of tradition Chinese medicine(TCM)use,and assessed factors influenceing TCM use and the association of TCM use with in-hospital bleeding and mortality using hierarchical logistic regression models.Outcome measures include in-hospital mortality and length of stay(LOS).We fUrther ealculated a composite performance score depending on the number of the performance measures attained divided by the number potentially indicated for each patient at the hospital level(ranging from 0%to 100%).To determine which hospital characteristics were associated with a better performance score,multivariable linear regression modeling with the median composite performance score as the continuous outcome variable was performed.To compare outcomes among hospitals,we calculated risk-standardized in-hospital mortality(RSMR)rate and risk-standardized LOS(RS-LOS).Results A total 10004 patients hospitalized for HF from 189 selected hospitals were included in the study.The median age of the cohort was 73 years(IQR,65-80),and 48.9%were women.More than half(56.2%)of the patients were hospitalized in rural areas(county hospitals).Patients with NYHA class IWTV comprised 71%of the patients.Only 0.2%of patients presented with cardiogenic shock.A high prevalence of comorbidities.Among patients who underwent echocardiography during hospitalization(n=6258,63.6%),the prevalence of LVEF?50%,40-50%and<40%was 60.3%,17.7%and 22.0%,respectively.Adherence to measure(1)was noted in 6258 of 9840(63.6%)eligible admissions,measure(2)in 648 of 1266(51.2%)eligible patients,measure(3)in 262 of 1327(19.7%)eligible patients,and measure(4)in 2325 of 9662(24.1%)eligible admissions.We found misuse of medications when contraindicated like aldosterone antagonist(19.9%).We also identified substantial gaps in underatilization of diagnostic tests such as chest imaging(75.2%)and biomarker testing(56.4%).Across all hospitals,the median rate of adherence to measure(1)was 66.7%(range 0-100%,IQR,45.5%-80.7%),measure(2)was 57.1%(range 0-100%,IQR,36.4%-75.0%),measure(3)was 14.8%(range 0-81.8%,IQR,0-37.5%),and measure(4)was 11.5%(range 0-96.7%,IQR,3.3%-32.8%).The median composite performance score across all hospitals was 40.0%(IQR,26.9%-51.9%),with range from 2.2%to 85.4%.The odds of receiving guideline-recommended care varied,on average,by two-to five-fold between hospitals for the core measures.Having an independent cardiology department was associated with an increase o16.5%(Cl 9.9%-23.1%;p<0.001),on average,in the performance score.TCM was used in 74.8%of patients hospitalized for HF(83.8%administered intravenously).The most commonly used agent was Salvia miltiorrhiza(51.2%).Patients with coronary artery disease(OR:1.7;95%Cl:1.5-2.0)or stroke(OR:1.3;95%Cl:1.2-1.5)were more likely to receive TCM;there was no correlation with evidence-based therapy use.Nearly all hospitals(99.4%)used TCM,with average odds of a patient receiving TCM at one random hospital versus another varied 3.3-fold(95%Cl:2.8-3.8).In-patient bleeding(OR:1.4,95%Cl:1.03-1.9)and mortality(OR:1.4,95%Cl:1.04-1.8)were higher with Salvia miltiorrhiza,though not with other TCMs.The combined rate of in-hospital mortality and treatment withdrawal in our study was 3.5%.The median RSMR was 3.3%(IQR:3.0%-3.9%,range 2.5%-6.9%).Average odds of a patient died at one random hospital versus another varied 1.7-fold(95%Cl:1.4-2.0).The median LOS was 9 days(IQR 7-13 days).RS-LOS ranged from 6.7 to 22.4 days across all hospitals.Conclusions Patients admitted with acute HF in China have distinctive epidemiology and receive substandard care.Given the overall high use of TCM,along with the signal for potential harm with the most commonly used TCM,there is an urgent need for the study of the safety and efficacy of TCM.The reasons for relatively lower rates of inpatient mortality despite longer LOS are tnclear but could likely be due to hospitalization of patients who could otherwise be treated in the outpatient setting.Quality of care and outcomes for HF varied substantially by hospital.These findings provide opportunities for streamlining efficiencies while improving quality and minimize heterogenity of inpatient HF care in China. |
PDF Full Text Request