Risk-Adjusting Hospital Inpatient Mortality Using Automated Inpatient, Outpatient, and Laboratory Databases

Objectives:To develop a risk-adjustment methodology that maximizes the use of automated physiology and diagnosis data from the time period preceding hospitalization. Design:Retrospective cohort study using split-validation and logistic regression. Setting:Seventeen hospitals in a large integrated health care delivery system. Subjects:Patients (n = 259,699) hospitalized between January 2002 and June 2005. Main Outcome Measures:Inpatient and 30-day mortality. Results:Inpatient mortality was 3.50%; 30-day mortality was 4.06%. We tested logistic regression models in a randomly chosen derivation dataset consisting of 50% of the records and applied their coefficients to the validation dataset. The final model included sex, age, admission type, admission diagnosis, a Laboratory-based Acute Physiology Score (LAPS), and a COmorbidity Point Score (COPS). The LAPS integrates information from 14 laboratory tests obtained in the 24 hours preceding hospitalization into a single continuous variable. Using Diagnostic Cost Groups software, we categorized patients as having up to 40 different comorbidities based on outpatient and inpatient data from the 12 months preceding hospitalization. The COPS integrates information regarding these 41 comorbidities into a single continuous variable. Our best model for inpatient mortality had a c statistic of 0.88 in the validation dataset, whereas the c statistic for 30-day mortality was 0.86; both models had excellent calibration. Physiologic data accounted for a substantial proportion of the model's predictive ability. Conclusion:Efforts to support improvement of hospital outcomes can take advantage of risk-adjustment methods based on automated physiology and diagnosis data that are not confounded by information obtained after hospital admission.

[1]  E. Keeler,et al.  Interpreting Hospital Mortality Data , 1988 .

[2]  T. Shakespeare,et al.  Observational Studies , 2003 .

[3]  A. Zaslavsky,et al.  Profiling Providers on Use of Adjuvant Chemotherapy by Combining Cancer Registry and Medical Record Data , 2006, Medical care.

[4]  J. Selby,et al.  Linking Automated Databases for Research in Managed Care Settings , 1997, Annals of Internal Medicine.

[5]  J. Allison.  Risk adjustment for measuring health care outcomes , 1996 .

[6]  A. Ash,et al.  Refinements to the Diagnostic Cost Group (DCG) model. , 1995, Inquiry : a journal of medical care organization, provision and financing.

[7]  A. Elixhauser,et al.  Modifying DRG-PPS to Include Only Diagnoses Present on Admission: Financial Implications and Challenges , 2007, Medical care.

[8]  T. Osler,et al.  Accuracy of hospital report cards based on administrative data. , 2006, Health services research.

[9]  G. W. Snedecor.  Statistical Methods , 1964 .

[10]  G. Escobar,et al.  SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. , 2001, The Journal of pediatrics.

[11]  X Zhang,et al.  Multivariate Matching and Bias Reduction in the Surgical Outcomes Study , 2001, Medical care.

[12]  D. E. Lawrence,et al.  APACHE—acute physiology and chronic health evaluation: a physiologically based classification system , 1981, Critical care medicine.

[13]  U. Ruttimann,et al.  PRISM III: an updated Pediatric Risk of Mortality score. , 1996, Critical care medicine.

[14]  U. Ruttimann,et al.  Pediatric risk of mortality (PRISM) score. , 1988, Critical care medicine.

[15]  T A Louis,et al.  Random effects models with non-parametric priors. , 1992, Statistics in medicine.

[16]  W. Knaus,et al.  The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. , 1991, Chest.

[17]  F. Harrell,et al.  Regression modelling strategies for improved prognostic prediction. , 1984, Statistics in medicine.

[18]  L I Iezzoni,et al.  Explaining differences in English hospital death rates using routinely collected data , 1999, BMJ.

[19]  Peter Buerhaus,et al.  Nurse-staffing levels and the quality of care in hospitals. , 2002, The New England journal of medicine.

[20]  Duncan Neuhauser,et al.  Risk Adjustment for Measuring Health Care Outcomes , 1995 .

[21]  D Draper,et al.  Interpreting hospital mortality data. The role of clinical risk adjustment. , 1988, JAMA.

[22]  Yuchiao Chang,et al.  Anticoagulation therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical practice? , 2003, JAMA.

[23]  J. Zimmerman,et al.  Acute Physiology and Chronic Health Evaluation (APACHE) IV: Hospital mortality assessment for today’s critically ill patients* , 2006, Critical care medicine.

[24]  L. Iezzoni.  Risk Adjustment for Measuring Healthcare Outcomes , 1994 .

[25]  T. Osler,et al.  Does date stamping ICD-9-CM codes increase the value of clinical information in administrative data? , 2006, Health services research.

[26]  G. Escobar,et al.  Rehospitalisation after birth hospitalisation: patterns among infants of all gestations , 2005, Archives of Disease in Childhood.

[27]  D. Bates,et al.  Using Diagnoses to Describe Populations and Predict Costs , 2000, Health care financing review.

[28]  Jonathan Mant,et al.  Detecting differences in quality of care: the sensitivity of measures of process and outcome in treating acute myocardial infarction , 1995, BMJ.

[29]  L. Forni,et al.  Identifying the sick: can biochemical measurements be used to aid decision making on presentation to the accident and emergency department. , 2005, British journal of anaesthesia.

[30]  L. McMahon,et al.  APACHE-L: A New Severity of Illness Adjuster for Inpatient Medical Care , 1992, Medical care.

[31]  M. Glickman,et al.  Statistical Methods for Profiling Providers of Medical Care: Issues and Applications , 1997 .

[32]  G. Escobar,et al.  Incorporation of physiological trend and interaction effects in neonatal severity of illness scores: an experiment using a variant of the Richardson score , 2007, Intensive Care Medicine.

[33]  Bekele Afessa,et al.  The influence of missing components of the Acute Physiology Score of APACHE III on the measurement of ICU performance , 2005, Intensive Care Medicine.

[34]  J. Birkmeyer,et al.  Surgical mortality as an indicator of hospital quality: the problem with small sample size. , 2004, JAMA.

[35]  L. Iezzoni,et al.  Finally present on admission but needs attention. , 2007, Medical care.

[36]  S Lemeshow,et al.  Factors affecting the performance of the models in the Mortality Probability Model II system and strategies of customization: a simulation study. , 1996, Critical care medicine.

[37]  D. Hoaglin,et al.  Enhancement of claims data to improve risk adjustment of hospital mortality. , 2007, JAMA.

[38]  D. Wagner,et al.  Automated intensive care unit risk adjustment: results from a National Veterans Affairs study. , 2003, Critical care medicine.

[39]  W. Knaus.  The APACHE III Prognostic System , 1992 .

[40]  R. Hayward,et al.  Identifying poor-quality hospitals. Can hospital mortality rates detect quality problems for medical diagnoses? , 1996, Medical care.

[41]  J. A. N. Eedleman,et al.  Nurse-Staffing Levels and the Quality of Care in Hospitals , 2002 .

[42]  D. Cook,et al.  Prediction of risk of death using 30-day outcome: a practical end point for quality auditing in intensive care. , 2004, Chest.

[43]  D. Hosmer,et al.  A review of goodness of fit statistics for use in the development of logistic regression models. , 1982, American journal of epidemiology.