Risk-adjusting Hospital Mortality Using a Comprehensive Electronic Record in an Integrated Health Care Delivery System
ABSTRACT : Using a comprehensive inpatient electronic medical record, we sought to develop a risk-adjustment methodology applicable to all hospitalized patients. Further, we assessed the impact of specific data elements on model discrimination, explanatory power, calibration, integrated discrimination improvement, net reclassification improvement, performance across different hospital units, and hospital rankings.
: Retrospective cohort study using logistic regression with split validation.
: A total of 248,383 patients who experienced 391,584 hospitalizations between January 1, 2008 and August 31, 2011.
: Twenty-one hospitals in an integrated health care delivery system in Northern California.
: Inpatient and 30-day mortality rates were 3.02% and 5.09%, respectively. In the validation dataset, the greatest improvement in discrimination (increase in c statistic) occurred with the introduction of laboratory data; however, subsequent addition of vital signs and end-of-life care directive data had significant effects on integrated discrimination improvement, net reclassification improvement, and hospital rankings. Use of longitudinally captured comorbidities did not improve model performance when compared with present-on-admission coding. Our final model for inpatient mortality, which included laboratory test results, vital signs, and care directives, had a c statistic of 0.883 and a pseudo-R of 0.295. Results for inpatient and 30-day mortality were virtually identical.
: Risk-adjustment of hospital mortality using comprehensive electronic medical records is feasible and permits one to develop statistical models that better reflect actual clinician experience. In addition, such models can be used to assess hospital performance across specific subpopulations, including patients admitted to intensive care.
- SourceAvailable from: David W Bates[Show abstract] [Hide abstract]
ABSTRACT: The US health care system is rapidly adopting electronic health records, which will dramatically increase the quantity of clinical data that are available electronically. Simultaneously, rapid progress has been made in clinical analytics-techniques for analyzing large quantities of data and gleaning new insights from that analysis-which is part of what is known as big data. As a result, there are unprecedented opportunities to use big data to reduce the costs of health care in the United States. We present six use cases-that is, key examples-where some of the clearest opportunities exist to reduce costs through the use of big data: high-cost patients, readmissions, triage, decompensation (when a patient's condition worsens), adverse events, and treatment optimization for diseases affecting multiple organ systems. We discuss the types of insights that are likely to emerge from clinical analytics, the types of data needed to obtain such insights, and the infrastructure-analytics, algorithms, registries, assessment scores, monitoring devices, and so forth-that organizations will need to perform the necessary analyses and to implement changes that will improve care while reducing costs. Our findings have policy implications for regulatory oversight, ways to address privacy concerns, and the support of research on analytics.Health Affairs 07/2014; 33(7):1123-31. DOI:10.1377/hlthaff.2014.0041 · 4.32 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: BACKGROUND Sepsis, the most expensive cause of hospitalization in the United States, is associated with high morbidity and mortality. However, healthcare utilization patterns following sepsis are poorly understood.OBJECTIVE To identify patient-level factors that contribute to postsepsis mortality and healthcare utilization.DESIGN, SETTING, PATIENTSA retrospective study of sepsis patients drawn from 21 community-based hospitals in Kaiser Permanente Northern California in 2010.MEASUREMENTSWe determined 1-year survival and use of outpatient and facility-based healthcare before and after sepsis and used logistic regression to identify the factors that contributed to early readmission (within 30 days) and high utilization (≥15% of living days spent in facility-based care).RESULTSAmong 6344 sepsis patients, 5479 (86.4%) survived to hospital discharge. Mean age was 72 years with 28.9% of patients aged <65 years. Postsepsis survival was strongly modified by age; 1-year survival was 94.1% for <45 year olds and 54.4% for ≥85 year olds. A total of 978 (17.9%) patients were readmitted within 30 days; only a minority of all rehospitalizations were for infection. After sepsis, adjusted healthcare utilization increased nearly 3-fold compared with presepsis levels and was strongly modified by age. Patient factors including acute severity of illness, hospital length of stay, and the need for intensive care were associated with early readmission and high healthcare utilization; however, the dominant factors explaining variability—comorbid disease burden and high presepsis utilization—were present prior to sepsis admission.CONCLUSION Postsepsis survival and healthcare utilization were most strongly influenced by patient factors already present prior to sepsis hospitalization. Journal of Hospital Medicine 2014;. © 2014 Society of Hospital MedicineJournal of Hospital Medicine 08/2014; 9(8). DOI:10.1002/jhm.2197 · 2.08 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Adherence to evidence-based recommendations for acute myocardial infarction (AMI) remains unsatisfactory. Quantifying association between using an electronic AMI order set (AMI-OS) and hospital processes and outcomes. Retrospective cohort study. Twenty-one community hospitals. A total of 5879 AMI patients were hospitalized between September 28, 2008 and December 31, 2010. We ascertained whether patients were treated using the AMI-OS or individual orders (a la carte). Dependent process variables were use of evidence-based care; outcome variables were mortality and rehospitalization. Use of individual and combined therapies improved outcomes (eg, 50% lower odds of 30-day mortality for patients with ≥3 therapies). The 3531 patients treated using the AMI-OS were more likely to receive evidence-based therapies (eg, 50% received 5 different therapies vs 36% a la carte). These patients had lower 30-day mortality (5.7% vs 8.5%) than the 2348 treated using a la carte orders. Although AMI-OS patients' predicted mortality risk was lower (3.2%) than that of a la carte patients (4.8%), the association of improved processes and outcomes with the use of the AMI-OS persisted after risk adjustment. For example, after inverse probability weighting, the relative risk for inpatient mortality in the AMI-OS group was 0.67 (95% confidence interval: 0.52-0.86). Inclusion of use of recommended therapies in risk adjustment eliminated the benefit of the AMI-OS, highlighting its mediating effect on adherence to evidence-based treatment. Use of an electronic order set is associated with increased adherence to evidence-based care and better AMI outcomes. Journal of Hospital Medicine 2014. © 2014 Society of Hospital Medicine.Journal of Hospital Medicine 03/2014; 9(3). DOI:10.1002/jhm.2149 · 2.08 Impact Factor