Refining case-mix adjustment. The research evidence

External validation of inpatient neonatal mortality prediction models in high-mortality settings

BACKGROUND

Two neonatal mortality prediction models, the Neonatal Essential Treatment Score (NETS) which uses treatments prescribed at admission and the Score for Essential Neonatal Symptoms and Signs (SENSS) which uses basic clinical signs, were derived in high-mortality, low-resource settings to utilise data more likely to be available in these settings. In this study, we evaluate the predictive accuracy of two neonatal prediction models for all-cause in-hospital mortality.

METHODS

We used retrospectively collected routine clinical data recorded by duty clinicians at admission from 16 Kenyan hospitals used to externally validate and update the SENSS and NETS models that were initially developed from the data from the largest Kenyan maternity hospital to predict in-hospital mortality. Model performance was evaluated by assessing discrimination and calibration. Discrimination, the ability of the model to differentiate between those with and without the outcome, was measured using the c-statistic. Calibration, the agreement between predictions from the model and what was observed, was measured using the calibration intercept and slope (with values of 0 and 1 denoting perfect calibration).

RESULTS

At initial external validation, the estimated mortality risks from the original SENSS and NETS models were markedly overestimated with calibration intercepts of - 0.703 (95% CI - 0.738 to - 0.669) and - 1.109 (95% CI - 1.148 to - 1.069) and too extreme with calibration slopes of 0.565 (95% CI 0.552 to 0.577) and 0.466 (95% CI 0.451 to 0.480), respectively. After model updating, the calibration of the model improved. The updated SENSS and NETS models had calibration intercepts of 0.311 (95% CI 0.282 to 0.350) and 0.032 (95% CI - 0.002 to 0.066) and calibration slopes of 1.029 (95% CI 1.006 to 1.051) and 0.799 (95% CI 0.774 to 0.823), respectively, while showing good discrimination with c-statistics of 0.834 (95% CI 0.829 to 0.839) and 0.775 (95% CI 0.768 to 0.782), respectively. The overall calibration performance of the updated SENSS and NETS models was better than any existing neonatal in-hospital mortality prediction models externally validated for settings comparable to Kenya.

CONCLUSION

Few prediction models undergo rigorous external validation. We show how external validation using data from multiple locations enables model updating and improving their performance and potential value. The improved models indicate it is possible to predict in-hospital mortality using either treatments or signs and symptoms derived from routine neonatal data from low-resource hospital settings also making possible their use for case-mix adjustment when contrasting similar hospital settings.

Serum osmolarity does not predict mortality in patients with respiratory failure

We aimed to determine the parameters that affect mortality in pulmonary intensive care units that are faster and inexpensive to determine than existing scoring systems. The relationship between serum osmolarity and prognosis was demonstrated for predialysis patients, in acute pulmonary embolism, heart failure, acute coronary syndrome, myocardial infarction, and acute spontaneous intracerebral hemorrhage in the literature. We hypothesized that serum osmolarity, which is routinely evaluated, may have prognostic significance in patients with respiratory failure.This study comprised 449 patients treated in the Pulmonary Intensive Care Clinic (PICU) of our hospital between January 1, 2020, and December 31, 2020. The modified Charlson Comorbidity Index (mCCI), Acute Physiology and Chronic Health Assessment (APACHE II), Sequential Organ Failure Evaluation Score (SOFA), Nutrition Risk Screening 2002 (NRS-2002), and hospitalization serum osmolarity levels were measured.Of the 449 patients included in the study, 65% (n = 292) were female and the mean age of all patients was 69.86 ± 1.72 years. About 83.1% (n = 373) of the patients included in the study were discharged with good recovery. About 4.9% (n = 22) were transferred to the ward because their intensive care needs were over. About 6.9% (n = 31) were transferred to the tertiary intensive care unit after their status deteriorated. About 5.1% (n = 23) died in the PICU. In the mortality group, APACHE II (P = .005), mCCI (P < .001), NRS-2002 total score (P < .001), and SOFA score (P < .001) were significantly higher. There was no statistically significant difference between the groups in terms of serum osmolarity levels.Although we could not determine serum osmolarity as a practical method to predict patient prognosis in this study, we assume that our results will guide future studies on this subject.

A scoring model to detect abusive medical institutions based on patient classification system: Diagnosis-related group and ambulatory patient group

The detection of medical abuse is essential because medical abuse imposes extra payments on individual insurance fees and increases unnecessary social costs. To reduce the costs due to medical abuse, insurance companies hire medical experts who examine claims, suspected to arise as a result of overtreatment from institutions, and review the suitability of claimed treatments. Owing to the limited number of reviewers and mounting volume of claims, there is need for a comprehensive method to detect medical abuse that uses a scoring model that selects a few institutions to be investigated. Numerous studies for detecting medical abuse have focused on institution-level variables such as the average values of hospitalization period and medical expenses to find the abuse score and selected institutions based on it. However, these studies use simple variables to construct a model that has poor performance with regard to detecting complex abuse billing patterns. Institution-level variables could easily represent the characteristics of institutions but loss of information is inevitable. Hence, it is possible to reduce information loss by using the finest granularity of data with treatment-level variables. In this study, we develop a scoring model by using treatment-level information and it is first of its kind to use a patient classification system (PCS) to improve the detection performance of medical abuse. PCS is a system that classifies patients in terms of clinical significance and consumption of medical resources. Because PCS is based on diagnosis, the patients grouped according to PCS tend to suffer from similar diseases. Claim data segmented by PCS is composed of patients with fewer types of diseases; hence, the data distribution by PCS is more homogeneous than data classified with respect to medical departments. We define an abusive institution as an institution having numerous number of abused treatments and containing their large sum of the abuse amounts, and the main idea of our model is that the abuse score of an institution is approximated as the sum of abuse scores for all treatments claimed from the institution. The proposed method consists of two steps: training a binary classification model to predict the abusiveness of each treatment and yielding an abuse score for each institution by aggregating the predicted abusiveness. The resulting abuse score is used to prioritize institutions to investigate. We tested the performance of our model against the scoring model employed by the insurance review agency in South Korea, making use of the real world claim data submitted to the agency. We compared these models with efficiency which represents the extent to which the model may detect the abused amounts per treatment. Experimental results show that the proposed model has efficiency up to 3.57 times higher than the model employed by the agency. In addition, we put forward an efficient and realistic reviewing process when the proposed scoring model is applied to the existing process. The proposed process has efficiency up to 2.17 times higher than the existing process.

[Not Available]

Zusammenfassung. Diagnose-bezogene Fallgruppen, DRGs, sind eines der seit den 1960er in den USA entwickelten und erstmals 1983 zu Abrechnungszwecken eingeführten Patientenklassifikationssysteme, die den Anspruch erheben, klinisch ähnliche Fälle, die ähnlich teuer sind, zwecks einer erhöhten Vergleichbarkeit zusammenzufassen. Die Hauptziele, welche mit der Einführung von DRGs zu Abrechnungszwecken verfolgt werden, sind einerseits eine Erhöhung der Transparenz der erbrachten Leistungen, sowie andererseits eine Steigerung der Effizienz durch die pauschale Rückvergütung des durchschnittlichen Aufwands der Fälle der selben DRG. In der Schweiz werden seit 2012 sämtliche stationäre, akut-somatische Fälle über DRGs abgerechnet. Obwohl einige Befürchtungen nicht objektiviert werden konnten, haben sich andere bestätigt.

Diagnosis-Related Groups: Past and Future

Introduction of the diagnosis-related group (DRG)-based Medicare Prospective Payment System is one of a series of major innovations that has occurred in the payment and delivery of health care over the past ten years. Changes such as the increased prevalence of health maintenance organizations, preferred provider organizations, third-party utilization review programs, and the peer review organizations for Medicare patients have all altered the way health care is financed and delivered. The DRG-based Medicare Prospective Payment System is the most visible of these changes, given its breadth of application and its radical departure from the previous retrospective reimbursement for hospital care. The Medicare Prospective Payment System has been in effect since October, 1983. As we approach the fifth anniversary of this program, it is a good time to review its history and to make some judgments as to its future.

Predictors of High Profit and High Deficit Outliers under SwissDRG of a Tertiary Care Center

Principles

Case weights of Diagnosis Related Groups (DRGs) are determined by the average cost of cases from a previous billing period. However, a significant amount of cases are largely over- or underfunded. We therefore decided to analyze earning outliers of our hospital as to search for predictors enabling a better grouping under SwissDRG.

Methods

28,893 inpatient cases without additional private insurance discharged from our hospital in 2012 were included in our analysis. Outliers were defined by the interquartile range method. Predictors for deficit and profit outliers were determined with logistic regressions. Predictors were shortlisted with the LASSO regularized logistic regression method and compared to results of Random forest analysis. 10 of these parameters were selected for quantile regression analysis as to quantify their impact on earnings.

Results

Psychiatric diagnosis and admission as an emergency case were significant predictors for higher deficit with negative regression coefficients for all analyzed quantiles (p<0.001). Admission from an external health care provider was a significant predictor for a higher deficit in all but the 90% quantile (p<0.001 for Q10, Q20, Q50, Q80 and p = 0.0017 for Q90). Burns predicted higher earnings for cases which were favorably remunerated (p<0.001 for the 90% quantile). Osteoporosis predicted a higher deficit in the most underfunded cases, but did not predict differences in earnings for balanced or profitable cases (Q10 and Q20: p<0.00, Q50: p = 0.10, Q80: p = 0.88 and Q90: p = 0.52). ICU stay, mechanical and patient clinical complexity level score (PCCL) predicted higher losses at the 10% quantile but also higher profits at the 90% quantile (p<0.001).

Conclusion

We suggest considering psychiatric diagnosis, admission as an emergencay case and admission from an external health care provider as DRG split criteria as they predict large, consistent and significant losses.

Australian diagnosis related groups: Drivers of complexity adjustment

BACKGROUND

In undertaking a major revision to the Australian Refined Diagnosis Related Group (ARDRG) classification, we set out to contrast Australia's approach to using data on additional (not principal) diagnoses with major international approaches in splitting base or Adjacent Diagnosis Related Groups (ADRGs).

METHODS

Comparative policy analysis/narrative review of peer-reviewed and grey literature on international approaches to use of additional (secondary) diagnoses in the development of Australian and international DRG systems.

ANALYSIS

European and US approaches to characterise complexity of inpatient care are well-documented, providing useful points of comparison with Australia's. Australia, with good data sources, has continued to refine its national DRG classification using increasingly sophisticated approaches. Hospital funders in Australia and in other systems are often under pressure from provider groups to expand classifications to reflect clinical complexity. DRG development in most healthcare systems reviewed here reflects four critical factors: these socio-political factors, the quality and depth of the coded data available to characterise the mix of cases in a healthcare system, the size of the underlying population, and the intended scope and use of the classification. Australia's relatively small national population has constrained the size of its DRG classifications, and development has been concentrated on inpatient care in public hospitals.

DISCUSSION AND CONCLUSIONS

Development of casemix classifications in health care is driven by both technical and socio-political factors. Use of additional diagnoses to adjust for patient complexity and cost needs to respond to these in each casemix application.

Clinical emergencies and outcomes in patients admitted to a surgical versus medical service

BACKGROUND

The merit of rapid response systems (RRSs) remains controversial. A tailored approach to specific groups may increase the efficacy of these teams. The purpose of this study was to compare differences in triggers for RRS activation, interventions, and outcomes in patients on medical and surgical services.

METHODS

A retrospective review RRS events was performed. The incidence of out of ICU cardiac arrests and hospital mortality were compared 2 years prior to and following RRS implementation. Call trigger, interventions, and disposition between medical and surgical patients were compared over a 15 month period.

RESULTS

Out of ICU cardiac arrest was significantly more prevalent in the medical group both before and after implementation of RRS. The out of ICU cardiac arrest rate decreased 32% in the surgical group (p=0.05) but hospital mortality did not change. Out of ICU cardiac arrest decreased 40% in the medical group (p<0.001) and hospital mortality decreased 25% (p<0.001) following RRS implementation. There were 1082 RRS activations, 286 surgical and 796 medical. Surgical patients were more likely to have received sedation within 24 h of evaluation (14% vs. 4%, p<0.001). The majority of patients in both cohorts were discharged alive.

CONCLUSION

Implementation of a RRS had greater impact on reduction of out of ICU cardiac arrest and mortality in medical inpatients. Triggers for activation and interventions were similar between groups; however, surgical patients demonstrated substantial risk for decompensation within the first 24 h following operation. More research is needed to evaluate the disproportionate benefit observed between cohorts.

Using risk adjustment approaches in child welfare performance measurement: Applications and insights from health and mental health settings

Federal policymaking in the last decade has dramatically expanded performance measurement within child welfare systems, and states are currently being fiscally penalized for poor performance on defined outcomes. However, in contrast to performance measurement in health settings, current policy holds child welfare systems solely responsible for meeting outcomes, largely without taking into account the effects of factors at the level of the child, and his or her social ecology, that might undermine the performance of child welfare agencies. Appropriate measurement of performance is predicated upon the ability to disentangle individual, as opposed to organizational, determinants of outcomes, which is the goal of risk adjustment methodologies. This review briefly conceptualizes and examines risk adjustment approaches in health and child welfare, suggests approaches to expanding its use to appropriately measure the performance of child welfare agencies, and highlights research gaps that diminish the appropriate use of risk adjustment approaches - and which consequently suggest the need for caution - in policymaking around performance measurement of child welfare agencies.

Potential unintended financial consequences of pay-for-performance on the quality of care for minority patients

OBJECTIVES

The purpose of this study was to determine whether pay-for-performance (PFP) increases existing racial care disparities.

BACKGROUND

Medicare's PFP program provides financial rewards to hospitals whose care performance ranks in the highest quintile relative to peers and reduces funding to hospitals that rank in the lowest quintile. Pay-for-performance is designed to improve care but may disproportionately penalize hospitals caring for large minority populations.

METHODS

Using Medicare data, 3449 US hospitals were ranked by performance on PFP process measures for acute myocardial infarction (AMI), community-acquired pneumonia (CAP), and heart failure (HF). These rankings were compared with the percentage of African American (AA) patients in a center. We determined the eligibility for financial bonus (highest quintile ranking) or penalty (lowest quintile) among centers treating large AA populations (> or = 20%) versus not after adjusting for hospital facility (catheterization, percutaneous coronary intervention, surgery), academic status, number of hospital beds, location, patient volume, and region.

RESULTS

The percentage of AA patients treated by a center was inversely associated with performance for AMI and CAP (P < .01) but not HF (P = .06). Relative to hospitals with < 20% AA, those with > or = 20% AA were less likely eligible for financial bonuses and more likely to face penalties: for AMI, adjusted odds ratio (OR) 0.7 (95% CI 0.5-1.0) and 1.8 (1.4-2.4), respectively; for CAP, OR 0.5 (95% CI 0.3-0.6) and 2.3 (1.8-2.9), respectively; for HF, OR 1.0 (95% CI 0.7-1.2) and 1.2 (0.9-1.5), respectively.

CONCLUSIONS

Hospitals with large minority populations may be at financial risk under PFP. Thus, PFP may worsen existing racial care disparities.

Charlson Index is associated with one-year mortality in emergency department patients with suspected infection

OBJECTIVES

A patient's baseline health status may affect the ability to survive an acute illness. Emergency medicine research requires tools to adjust for confounders such as comorbid illnesses. The Charlson Comorbidity Index has been validated in many settings but not extensively in the emergency department (ED). The purpose of this study was to examine the utility of the Charlson Index as a predictor of one-year mortality in a population of ED patients with suspected infection.

METHODS

The comorbid illness components of the Charlson Index were prospectively abstracted from the medical records of adult (age older than 18 years) ED patients at risk for infection (indicated by the clinical decision to obtain a blood culture) and weighted. Charlson scores were grouped into four previously established indices: 0 points (none), 1-2 points (low), 3-4 points (moderate), and > or =5 points (high). The primary outcome was one-year mortality assessed using the National Death Index and medical records. Cox proportional-hazards ratios were calculated, adjusting for age, gender, and markers of 28-day in-hospital mortality.

RESULTS

Between February 1, 2000, and February 1, 2001, 3,102 unique patients (96% of eligible patients) were enrolled at an urban teaching hospital. Overall one-year mortality was 22% (667/3,102). Mortality rates increased with increasing Charlson scores: none, 7% (95% confidence interval [CI] = 5.4% to 8.5%); low, 22% (95% CI = 19% to 24%); moderate, 31% (95% CI = 27% to 35%); and high, 40% (95% CI = 36% to 44%). Controlling for age, gender, and factors associated with 28-day mortality, and using the "none" group as a reference group, the Charlson Index predicted mortality as follows: low, odds ratio of 2.0; moderate, odds ratio of 2.5; and high, odds ratio of 4.7.

CONCLUSIONS

This study suggests that the Charlson Index predicts one-year mortality among ED patients with suspected infection.

Creating and analyzing a statewide nursing quality measurement database

PURPOSE

To explicate a replicable methodology for designing and analyzing a large ongoing reliable and valid quality database to examine nurse staffing and patient care outcomes in acute care hospitals.

DESIGN

Prospective nurse staffing, process of care, and patient outcomes data based on the American Nurses Association's (ANA) nursing quality indicators collected from a voluntary convenience sample at acute care hospitals in California with rolling-site accrual.

METHODS

The ongoing CalNOC database development and repository project, the largest statewide effort of its kind in the United States (US), currently includes data on hospital nurse staffing, patient days, patient falls, pressure ulcer and restraint prevalence, registered nurse (RN) education, and patients' perceptions of satisfaction with care.

FINDINGS

As of May 2003, the CalNOC database contained staffing data from 842 units in 134 acute care hospitals over 20 quarters from April 1998 to March 2003. The repository also included clinical outcome information on 34,262 reported patient falls, pressure ulcer prevalence data on 41,982 patient observations, and service outcome data on patient satisfaction from 26,461 patients. Participating hospitals receive quarterly reports allowing them to benchmark their own performance against other participating hospitals. CalNOC methods have been adapted and replicated by both the Military Nursing Outcomes Database and VA Nursing Outcomes Database projects, and CalNOC nursing-sensitive measures have been endorsed by the National Quality Forum.

CONCLUSIONS

This working model for collecting reliable and valid data was derived from multiple hospitals across California. The data are the basis for studies to contribute to the development of evidence-based public policy, and for ongoing study of the effects of nurse staffing on clinical and service outcomes.

The use of economic modeling to determine the hospital costs associated with nosocomial infections

Hospital-associated infection is well recognized as a patient safety concern requiring preventive interventions. However, hospitals are closely monitoring expenditures and need accurate estimates of potential cost savings from such prevention programs. We used a retrospective cohort design and economic modeling to determine the excess cost from the hospital perspective for hospital-associated infection in a random sample of adult medical patients. Study patients were classified as being not infected (n=139), having suspected infection (n=8), or having confirmed infection (n=17). Severity of illness and intensive unit care use were both independently associated with increased cost. After controlling for these confounding effects, we found an excess cost of $6767 for suspected infection and $15,275 for confirmed hospital-acquired infection. The economic model explained 56% of the total variability in cost among patients. Hospitals can use these data when evaluating potential cost savings from effective infection-control measures.

Age and laboratory costs for hospitalized medical patients

OBJECTIVE

To examine the hypothesis that older hospitalized patients have higher laboratory costs than younger patients in the same severity-adjusted diagnosis-related group (DRG).

DESIGN

We obtained hospital case mix data sets (1995-1997) from the Massachusetts Division of Health Care Finance and Policy. We selected discharge abstracts from 4 medical DRGs, at 5 large academic hospitals (n = 15,265) and 5 midsized community hospitals (n = 10,540), for analysis. We converted laboratory and blood product charges to direct costs using the department-specific ratio of cost to charges. We adjusted diagnostic groups for severity of comorbid conditions and complications using the refined DRG method.

MAIN OUTCOME MEASURES

Hospital length of stay (LOS), laboratory direct cost (LDC) per hospitalization, LDC per hospital day, and ratio of LDC to total direct cost.

RESULTS

Hospital LOS was longer for older patients in all comparisons. Laboratory direct cost per hospitalization was higher for older patients in some DRGs, but lower in other DRGs. Laboratory direct cost per hospital day was almost always less for older patients than for younger patients, both at academic and community hospitals. Data stratification by gender, admission status, and principal diagnosis yielded substantially the same pattern of cost differences observed within the larger data set.

CONCLUSIONS

Older medical patients have longer hospital stays and generally higher costs. These patients also have a significantly decreased rate of laboratory resource consumption over the course of hospitalization (LDC per hospital day), as well as lower laboratory costs as a proportion of total costs. Age-specific differences in LOS and cost parameters were essentially unchanged after controlling for several potential sources of bias.

Early experience with clinical indicators in surgery

BACKGROUND

In 1997 a set of 53 clinical indicators developed by the Royal Australasian College of Surgeons (RACS) and the Australian Council on Healthcare Standards (ACHS) Care Evaluation Programme (CEP), was introduced into the ACHS Evaluation and Quality Improvement Programme (EQuIP). The clinical indicators covered 20 different conditions or procedures for eight specialty groups and were designed to act as flags to possible problems in surgical care.

METHODS

The development process took several years and included a literature review, field testing, and revision of the indicators prior to approval by the College council. In their first year 155 health-care organizations (HCO) addressed the indicators and this rose to 210 in 1998. Data were received from all states and both public and private facilities.

RESULTS

The collected data for 1997 and 1998 for some of the indicators revealed rates which were comparable with those reported in the international literature. For example, the rates of bile duct injury in laparoscopic cholecystectomy were 0.7 and 0.53%, respectively; the mortality rates for coronary artery graft surgery were 2.5 and 2.1%, respectively; the mortality rates after elective abdominal aortic aneurysm repair were 2.5 and 3.7%, respectively; and the post-tonsillectomy reactionary haemorrhage rates were 0.9 and 1.3%, respectively. Results for some indicators differed appreciably from other reports, flagging the need for further investigation; for example, the negative histology rates for appendectomy in children were 18.6 and 21.2%, respectively, and the rates for completeness of excision of malignant skin tumours were 90.7 and 90%, respectively. The significance of these figures, however, depends upon validation of the data and their reliability and reproducibility. Because reliability can be finally determined only at the hospital level they are of limited value for broader comparison.

CONCLUSION

The process of review established for the indicator set has led to refinement of some indicators through improvement of definitions, and to a considerable reduction in the number of indicators to 29 (covering 18 procedures), for the second version of the indicators (which was introduced for use from January 1999). The clinical indicator programme, as it has with other disciplines, hopefully will provide a stimulus to the modification and improvement of surgical practice. Clinician ownership should enhance the collection of reliable data and hence their usefulness.

Management of gastrointestinal bleeding induced by gastrointestinal endoscopy

Therapeutic gastrointestinal endoscopy has a much greater risk of inducing gastrointestinal hemorrhage than diagnostic endoscopy. For example, colonoscopic polypectomy has a risk of approximately 1.6% of inducing bleeding, compared with a risk of approximately 0.02% for diagnostic colonoscopy. Higher-risk procedures include colonoscopic polypectomy, endoscopic biliary sphincterotomy, endoscopic dilatation, endoscopic variceal therapy, percutaneous endoscopic gastrostomy, and endoscopic sharp foreign body retrieval. The risk of inducing hemorrhage is decreased by meticulous endoscopic technique. Hemorrhage from endoscopy may be immediate or delayed. Immediate hemorrhage should be immediately treated by endoscopic hemostatic therapy, including injection therapy, thermocoagulation, or electrocoagulation. Delayed hemorrhage generally requires repeat endoscopy for diagnosis and for therapy, using the same hemostatic techniques.

Principles of disease management in neonatology

This article emphasizes the emerging facets of disease-management practice that impact directly on establishing a measured care system that can produce the information needed to establish a continuous quality improvement program. The areas discussed are risk assessment, clinical management guidelines and carepaths, and the measurement of system output known as clinical outcomes. The remainder of the article details the aspects of risk assessment, guideline function, and outcome assessment, critical in a disease-managed measured care system.

Variation in hospital use and avoidable patient morbidity

OBJECTIVES

To determine whether geographical areas with relatively low overall hospitalization rates have higher population-based rates of admission of patients with advanced stages of disease.

METHODS

Age- and sex-standardized hospital admission rate were calculated for the residents of the 80 Local Health Units in Lombardia, Italy. Using the Disease Staging classification, advanced stage admissions were identified for six common medical and surgical conditions, which it was presumed would reflect untimely hospital admission. Standardized rates of advanced stage admissions were compared in areas with overall high hospitalization rates (high-use areas).

RESULTS

Hospitalization at advanced stages of disease in the low-use areas were significantly higher for the six conditions combined (55.9 vs 43.0 per 100,000; P = 0.005), and for external hernia, appendicitis and uterine fibroma, but not for bacterial pneumonia, diverticular disease and peptic ulcer. For the six study conditions combined, residents of overall low-use area were 30% more likely to be admitted with advanced stages of disease.

CONCLUSION

Low overall hospitalization rates were found to be associated with greater severity of illness at hospitalization and potentially avoidable morbidity for some conditions. Policies aimed at curbing unnecessary hospital admission should consider preserving access for appropriate treatment.

Accuracy of administrative data for assessing outcomes after knee replacement surgery

OBJECTIVE

To assess the accuracy of information in an administrative database (Canadian Institute for Health Information; CIHI) compared with the hospital record for patients undergoing knee replacement (KR).

METHODS

A stratified random sample of 185 KR recipients from 5 Ontario hospitals were chosen. Their hospital records and corresponding CIHI files were compared to assess percent complete agreement, false negative (FN) and false positive (FP) rates for demographic data, procedures, and diagnoses.

RESULTS

Of 185 records, 175 (95%) were reviewed. Percent complete agreement was greater than 94% for each of patient demographics and procedures (mean FN rates: 0%; mean FP rates: 0-5%). For comorbidities and complications, although mean percent complete agreement was high, and FP rates were low, mean FN rates were 63% for specific comorbid conditions and 70% for organ systems.

CONCLUSIONS

High FN rates have been found in documentation of comorbidities and in-hospital complications for CIHI data compared with the hospital record. Under-coding of comorbidities and in-hospital complications has potential implications for researchers using administrative databases.

Secondary diagnoses as predictive factors for survival or mortality in Medicare patients with acute pneumonia

We wished to determine if a claims-based method for severity adjustment would predict mortality or survival in pneumonia based on age, gender, and secondary diagnoses. We used a discriminant analysis model of severity of illness developed from Medicare Part A claims data. Our data base was taken from a hospitalized population age 65 years or older coded as DRG 89 (pneumonia with complications/comorbidities). There were 35,677 cases with a mortality = 11.2% in the derivation cohort from 1989 to 1990, and 19,915 cases with a mortality = 9.8% in the validation cohort from 1991. In the derivation cohort, 98% of patients predicted to live, lived, whereas 18% of patients predicted to die, died. Of the three variables, secondary diagnoses had greatest explanatory power. Receiver operating characteristic curves showed that the model performed best at 40% survival. Results were confirmed with the 1991 validation cohort. The model could be applied to hospitals with as few as 172 discharges. This simple, claims-based method can predict survival in pneumonia. It may be useful in selecting medical records for intensified review of medical quality.

Does severity explain differences in hospital length of stay for pneumonia patients?

OBJECTIVES

In the USA, the role of patient severity in determining hospital resource use has been questioned since Medicare adopted prospective hospital payment based on diagnosis-related groups (DRGs). Exactly how to measure severity, however, remains unclear. We examined whether assessments of severity-adjusted hospital lengths of stay (LOS) varied when different measures were used for severity adjustment.

METHODS

The complete study sample included 18,016 patients receiving medical treatment for pneumonia at 105 acute care hospitals. We studied 11 severity measures, nine based on patient demographic and diagnosis and procedure code information and two derived from clinical findings from the medical record. For each severity measure, LOS was regressed on patient age, sex, DRG, and severity score. Analyses were performed on trimmed and untrimmed data. Trimming eliminated cases with LOS more than three standard deviations from the mean on a log scale.

RESULTS

The trimmed data set contained 17,976 admissions with a mean (S.D.) LOS of 8.9 (6.1) days. Average LOS ranged from 5.0-11.8 days among the 105 hospitals. Using trimmed data, the 11 severity measures produced R-squared values ranging from 0.098-0.169 for explaining LOS for individual patients. Across all severity measures, predicted average hospital LOS varied much less than the observed LOS, with predicted mean hospital LOS ranging from about 8.4-9.8 days.

DISCUSSION

No severity measure explained the two-fold differences among hospitals in average LOS. Other patient characteristics, practice patterns, or institutional factors may cause the wide differences across hospitals in LOS.

Searching for an improved clinical comorbidity index for use with ICD-9-CM administrative data

We studied approaches to comorbidity risk adjustment by comparing two ICD-9-CM adaptations (Deyo, Dartmouth-Manitoba) of the Charlson comorbidity index applied to Massachusetts coronary artery bypass surgery data. We also developed a new comorbidity index by assigning study-specific weights to the original Charlson comorbidity variables. The 2 ICD-9-CM coding adaptations assigned identical Charlson comorbidity scores to 90% of cases, and specific comorbidities were largely found in the same cases (kappa values of 0.72-1.0 for 15 of 16 comorbidities). Meanwhile, the study-specific comorbidity index identified a 10% subset of patients with 15% mortality, whereas the 5% highest-risk patients according to the Charlson index had only 8% mortality (p = 0.01). A model using the new index to predict mortality had better validated performance than a model based on the original Charlson index (c = 0.74 vs. 0.70). Thus, in our population, the ICD-9-CM adaptation used to create the Charlson score mattered little, but using study-specific weights with the Charlson variables substantially improved the power of these data to predict mortality.

Quality assurance and medical outcomes in the era of cost containment

Market forces are driving health care organizations to "prove" quality while diminishing costs. Payers for health care, led by large employers and insurance companies, are demanding clinical, financial, and satisfaction outcomes from providers. To meet the challenge, traditional quality assurance based on inspection and rooting out "bad apples" is rapidly being replaced by the industrial engineering principles of continuous quality improvement. A philosophical shift is occurring from a focus on episodes of care delivered by physicians to the delivery of processes of care by teams of health care personnel. We are seeing a shift in emphasis from a fascination with intensive care delivered to sick patients to cost-effective preventive services delivered to populations of well patients. The locus of care delivery is moving from inpatient hospitals to ambulatory clinics and home care. The need for this information is leading to innovation in computer systems and health care organizations. New partnerships are emerging between physicians, nurses, and hospitals. Traditional oversight bodies including the JCAHO and the HCFA-sponsored PROs are restructuring to meet these new demands. New organizations such as the National Committee on Quality Assurance and state governmental agencies are being established to fill the perceived void. Individual surgeons have begun to receive performance data on their individual and group practices. Professional societies have collaborated in the development of clinical guidelines and outcomes data bases. This massive reorganization will take several more years to play out. With careful development it has the potential to dramatically improve patient care through the efficient application of new scientific knowledge and the sustained flow of information back to physicians and patients.

[Severity of illness explains the inadequacy between diagnosis-related groups and intensive care patients. Groupe GHM]

OBJECTIVE

To assess the relationship between diagnosis related groups (DRG) and severity of illness in intensive care unit (ICU) patients in semf1tical and economical terms.

STUDY DESIGN

Prospective, multicentric study including 13 medical and surgical ICUs for adults.

MATERIAL

Discharge reports of 3,215 ICU admissions including age, gender, diagnosis, organ system failures, length of stay (LOS) and severity of illness evaluated with severity scores (SS): simplified acute physiological score (SAPS). Apache II, Glasgow score and physiological score (PS).

METHODS

Semantical homogeneity was evaluated from the percentage of well-classified patients established from the comparison between the official computerized method and a non-computerized method applied by three clinical experts. Economical homogeneity was evaluated from the relationship between SS and LOS.

RESULTS

In total, 88% (CI: 87.7-88.2) of ICU stays were classified in eight main categories of diagnosis (MCD). According to the MCD, the percentage of well-classified patients varied from 28% (CI: 27.6-28.3) to 97% (CI: 96.8-97.1), decreasing with the association of several diagnoses and organ system failures. There was a large variability in the LOS of DRG and a significant correlation between LOS and SS was found in only 8/16 DRG.

CONCLUSION

The fact that the severity of illness is not taken into account in the elaboration of DRGs explains the inadequacy of the DRG system in intensive care.

The Australian Council on Healthcare Standards care Evaluation Program

With the assistance of the medical colleges, the Australian Council on Healthcare Standards (ACHS), through its Care Evaluation Program, has established clinical performance measures which will assist both internal and external review of care and enable hospitals to compare their quality of patient care with that of other hospitals.

Effect of patient factors on hospital costs for major bowel surgery: implications for managed health care

BACKGROUND

We examined the effects of patient factors on hospital resource consumption for patients who had undergone major bowel operation (diagnosis-related groups [DRGs] 148 and 149) at an urban, university hospital.

METHODS

We performed cross-sectional analysis of computerized hospital discharge abstracts and charts of 491 consecutive discharges in these DRGs. Total hospital charges and length of stay were dependent variables. Independent variables included admission status, admission service, previous admissions, payer type, service type, diagnosis, reoperation, and death.

RESULTS

Patient factors accounted for significant variability in resource consumption. By univariate analysis all of the above variables significantly affected total charges, and all but service type significantly affected length of stay. By multivariate analysis DRGs 148/149 alone explained 4.2% of the variance, whereas all the variables together increased R2 to 52.1%. Logistic regression of reoperation and of death as dependent variables suggested that patient factors also accounted for significant variance in these outcomes.

CONCLUSIONS

Because patient factors may not be directly controllable by hospitals or physicians, differences among hospitals in costs and in "quality" may relate more to differences in patient mix than to efficiency. DRGs alone are not a sufficient management tool, and additional measures are needed to adequately measure both efficiency and quality.

Explaining cost variations in DRGs 'Acute Myocardial Infarction' by severity of illness

The empirical relationship is analyzed between the severity of illness and costs of medical care for 464 patients classified into DRGs 121-123, Acute Myocardial Infarction (AMI), in the University Hospital, Maastricht. Severity of cardiac and cardiovascular disorders characteristic of acute myocardial infarction is defined and operationalized in a sense that closely resembles the clinical practice of cardiologists. The effect of the severity of illness on DRG cost variations is studied separately for the costs of acute care (such as thrombolytic therapy, cardiac catheterization and percutaneous transluminal coronary angioplasty (PTCA)), length of hospital stay, costs of intensive nursing care at the coronary care unit (CCU) and the costs of ECGs, laboratory tests, echocardiography, exercise tests and drugs. For AMI patients, severity of illness measured by specific clinical criteria is found to give better predictions (higher R2) for costs of medical care than the DRG classification.

Measuring hospital workload in general medicine

The study described here used routine data sources to consider three aspects of hospital workload-volume and case-mix of inpatients; volume of outpatients; and volume of procedures.

A simple severity of disease index for systemic lupus erythematosus

Although indices of activity for systemic lupus erythematosus have been developed and validated, a disease staging system requires the measurement of severity as well. We have constructed such a scale, the Lupus Severity of Disease Index (Lupus SDI). Accepted clinical, pathologic and physiologic classification schemes were employed to validate this index at two separate research institutions. The Lupus SDI allows homogenization of patient populations for the purposes of research and, possibly, for case mix adjustment.

Comparison of postoperative mortality in VA and private hospitals

OBJECTIVE

This study compared unselected VA (Department of Veterans Affairs) and private multi-hospital postoperative mortality rates. In the absence of national standards for postoperative mortality rates and in view of the unique volume and range of surgical procedures studied, the second objective is to help establish national standards through the dissemination of these postoperative mortality norms.

SUMMARY BACKGROUND DATA

Public Law 99-166, Section 204, enacted by Congress December 3, 1985, required that the VA compare postoperative mortality and morbidity rates for each type of surgical procedure it performs with the prevailing national standard and analyze any deviation between such rates in terms of patient characteristics.

METHODS

The authors compared postoperative mortality in the VA to that in private hospitals, adjusting for the patient characteristics of age, diagnosis, comorbidity, or severity of illness. We used a total of 830,000 patients discharge records (323,000 VA and 507,000 private patients) from 1984 through 1986 among 309 individual surgical procedures within 113 comparison surgical procedures or procedure groups.

RESULTS

The authors found no significant differences in postoperative mortality rates between the VA and private hospital systems for 105 of the 113 surgical procedures or procedure groups. VA postoperative mortality rates that were higher than those in private hospitals were found for suture of ulcer, revision of gastric anastomosis, small-to-small intestinal anastomosis, appendectomy, and reclosure of postoperative disruption of abdominal wall (p = 0.05). Vascular bypass surgery, portal systemic venous shunt, and esophageal surgery showed a significantly lower postoperative mortality in the VA as compared with that in private hospitals (p = 0.05).

CONCLUSIONS

VA postoperative mortality in 113 surgical procedures or procedure groups is comparable to that in private hospitals.

Measuring case mix and severity of illness in Canada: case mix groups versus refined diagnosis related groups

This study compares the effectiveness of Case Mix Groups (CMG*) groups and Refined Diagnosis Related Groups (RDRG) in reducing Canadian length of stay (LOS) variability. The effectiveness of the two case mix grouping methodologies was assessed with a common data base, 282,459 abstracts with ICD-9 CM diagnosis codes reported to the Hospital Medical Records Institute (HMRI) from January to March 1989. Death, signouts, transfers to or from acute care institutions and cases with an outlier LOS ("atypical" cases) were excluded from the analysis. HMRI utilization management reports to acute care hospitals use a data base defined in this way. On the basis of the variance reduction statistic (R2) from ordinary least squares regression analysis, CMG groups were found to be slightly more effective than RDRGs in reducing LOS variability. R2 statistics were 45.7 and 43.8 for CMG groups and RDRGs, respectively. Within subgroups of cases, CMG groups were found to be markedly more effective with the newborn/neonate group and to a lesser extent with non-surgical cases. The severity of illness categories within RDRGs did not, over all "typical" cases in the data base, yield more homogeneous groups of cases than CMG groups, which have half the number of categories. The value of tailoring severity measurement to Canadian medical practice and Canadian diagnosis coding is highlighted.

Comparing in-patient classification systems: a problem of non-nested regression models

Since 1983, hospitals in the United States have been receiving prospective payment for their in-hospital patient admissions covered under Medicare. Under such schemes each patient is placed in a group by a classification system, known as the Diagnosis Related Groups (DRG), and the hospital is reimbursed by the Health Care Financing Administration according to some predetermined group average, adjusted for hospital level characteristics, such as size, location and teaching activity. Recent interest has focused on refining the DRG system or considering totally different systems of classification. Studies designed to compare the ability of different systems to account for between-patient variability in resource consumption in the same dataset lead to the problem of model selection between large non-nested regressions, where resource consumption, measured by length of hospital stay or costs, is regressed on dummy-indicator variables representing different patient groups. We use a simple measure of fit to develop a symmetric test of the null hypothesis that the two systems account equally well for variability in resource consumption. With this method, unlike methods such as Akaike's AIC criterion, we can quantify the probability of a false positive, and thereby limit the probability of choosing one system over another when it is no better at accounting for variability in resource consumption.

A description and clinical assessment of the Computerized Severity Index

Recent initiatives expanding health care data networks have increasingly emphasized severity of illness information, both to improve fairness of hospital payment and to assist in widespread assessment of hospital and physician quality. The Computerized Severity Index (CSI), one of the newest severity tools to generate interest, is disease specific and produces scores from 1 to 4 at both the disease and overall patient levels. Severity is defined as "the treatment difficulty presented to physicians due to the extent and interactions of patient's diseases." The clinical logic of the severity rating system is readily available through the "severity matrices" associated with over 820 disease groups. Questions exist about the CSI's dependence on diagnostic coding and the qualitative nature of some of the clinical criteria. More study is required to assess the utility of the CSI for various health policy purposes.

Utilization management: a literature review for Canadian health care administrators

Utilization management (UM), the attempt to measure, understand and reduce inappropriate hospital use, has been in development for over 20 years. It is an outgrowth of two related phenomena: (1) the increasing responsibility of large institutional third party payers for health care costs and the increasing demand of those payers for accountability; and (2) in Canada, particularly, the debate surrounding the adequacy of hospital funding and the perceived inadequacy of cost control using global budgeting. Given the interest in UM, hospital administrators, provincial and federal associations representing hospitals, hospital employees and physicians would find a review of UM programs useful in terms of what is known about their effectiveness, and the specific initiatives in Canada. The authors underscore the critical need for formal evaluation of UM programs; to date there has been little systematic research into issues related to its implementation and impact. This issue is particularly pertinent because UM programs have not been widely implemented in Canada.

Striving for benchmark infection rates: progress in control for patient mix

The outcome of clinical events has become the major focus for quality assurance programs in the United States. Assessment of outcome depends on the availability of accurate benchmark rates appropriate to the clinical situation. Although mortality, length of stay, cost, and other resource utilization measures are important outcome events for review, nosocomial infection is one of the best-studied outcome events for which benchmark rates and distributions of rates are available. To monitor nosocomial infection acquisition, the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) has proposed indicators for wound infections after clean and clean-contaminated surgical procedures, for pneumonia in intensive care units (ICUs), and for intravascular device infections in ICUs. From the National Nosocomial Infections Surveillance (NNIS) system, we know that the mean rate of surgical wound infections (SWIs) for clean-contaminated wounds is 2.7%. However, the range is too broad and depends on the types of surgical procedures and the patient risk factors in each category. As an example, the SWI rate for appendectomies is 2.2% while for colon surgery it is 7.1%. In addition, the patients should be stratified by underlying risk factors. For example, the NNIS risk index is composed of wound class, duration of surgery, and American Society of Anesthesiology score. Other commonly used severity of illness scores could be used, such as a comorbidity score or APACHE II. Other proposed JCAHO indicators are reviewed. Controlling for case mix and for severity of illness will be necessary for the development of benchmark infection rate distributions if clinical indicators are to be used as reliable quality assurance tools in the 1990s.

Patient intensity for nursing index: the measurement model

One part of the psychometric evaluation of the Patient Intensity for Nursing Index (PINI), a new measure of nursing intensity, is reported. The PINI has four interrelated components: (a) severity of illness, (b) dependency, (c) complexity of care, and (d) time. Taken together, the 10 items that make up these four components comprise the multidimensional construct of nursing intensity. Using the factor analytic approach and a model specification search, the measurement model for the Patient Intensity for Nursing Index was identified. The structure consists of Dependency, Severity, and Complexity with time (hours of nursing care) loading on each factor. When results were cross-validated using data from four other hospitals, support emerged for a consistent pattern of factor loadings. The loadings themselves, however, are only partially invariant. Further examination of the relationship between severity and time is suggested and areas for future research are identified.

Clinical outcomes management and disease staging

Clinical outcomes management includes multiple approaches for evaluating and improving the quality and cost effectiveness of medical care. The usefulness of outcomes assessments depends, in part, on how well the clinical issues have been specified and whether the analyses are sensitive to the diverse clinical characteristics of the patients receiving the medical care in question. Measures of severity of illness and, in particular, Disease Staging, have an important role in outcomes assessment by classifying diseases along dimensions that have prognostic significance. This article reviews current applications of Disease Staging for outcomes assessment and management.

International comparison of long-term care: the need for resident-level classification

Differences between long-term care facilities in Stockholm (1134 residents) and New York (95,000 residents statewide) were examined. The comparison employed a resident classification system, Resource Utilization Groups (RUG-II), which links individuals' characteristics to resource use. Distributions of Activity of Daily Living functionality and RUG-II categories demonstrated significant differences between these two populations, with the Stockholm facilities more akin to the heavier care skilled nursing facilities in New York. These differences may indicate different uses of long-term care beds in the United States and Sweden and demonstrate the need for resident-level classification systems in cross-national studies.

Direct costs of emergency medical care: a diagnosis-based case-mix classification system

STUDY OBJECTIVE

To develop a diagnosis-based case mix classification system for emergency department patient visits based on direct costs of care designed for an outpatient setting.

DESIGN

Prospective provider time study with collection of financial data from each hospital's accounts receivable system and medical information, including discharge diagnosis, from hospital medical records.

SETTING

Three community hospital EDs in Los Angeles County during selected times in 1984.

MEASUREMENTS AND MAIN RESULTS

Only direct costs of care were included: health care provider time, ED management and clerical personnel excluding registration, nonlabor ED expense including supplies, and ancillary hospital services. Indirect costs for hospitals and physicians, including depreciation and amortization, debt service, utilities, malpractice insurance, administration, billing, registration, and medical records were not included. Costs were derived by valuing provider time based on a formula using annual income or salary and fringe benefits, productivity and direct care factors, and using hospital direct cost to charge ratios. Physician costs were based on a national study of emergency physician income and excluded practice costs. Patients were classified into one of 216 emergency department groups (EDGs) on the basis of the discharge diagnosis, patient disposition, age, and the presence of a limited number of physician procedures. Total mean direct costs ranged from $23 for follow-up visit to $936 for trauma, admitted, with critical care procedure. The mean total direct costs for the 16,771 nonadmitted patients was $69. Of this, 34% was for ED costs, 45% was for ancillary service costs, and 21% was for physician costs. The mean total direct costs for the 1,955 admitted patients was $259. Of this, 23% was for ED costs, 63% was for ancillary service costs, and 14% was for physician costs. Laboratory and radiographic services accounted for approximately 85% of all ancillary service costs and 38% of total direct costs for nonadmitted patients versus 80% of ancillary service costs and 51% of total direct costs for admitted patients.

CONCLUSION

We have developed a diagnosis-based case mix classification system for ED patient visits based on direct costs of care designed for an outpatient setting which, unlike diagnosis-related groups, includes the measurement of time-based cost for physician and nonphysician services. This classification system helps to define direct costs of hospital and physician emergency services by type of patient.

The increased needs of patients in nursing homes and patients receiving home health care

To evaluate the effects of Medicare's prospective payment system and Medicaid's preadmission regulations on long-term care, we constructed clinical profiles in 1982 and 1986 of about 500 randomly selected patients from each of three types of facilities: nursing homes with relatively high proportions of Medicare patients (high-Medicare nursing homes; n = 23), traditional nursing homes (n = 19), and home health agencies (n = 18). Data were obtained directly from the care givers on the medical problems, problems requiring skilled nursing, and functional problems of these representative patients from 12 states. For Medicare patients in high-Medicare nursing homes, the prevalence of medical problems and problems requiring skilled nursing increased substantially, whereas the prevalence of functional problems remained relatively unchanged. For example, from 1982 to 1986 there was a marked increase in the frequency of tube feedings (21 to 29 percent), oxygen use (6 to 14 percent), urinary tract infection (7 to 13 percent), and diastolic hypertension (1 to 10 percent), but not difficulty in eating (48 to 51 percent) or speaking (28 to 29 percent). In contrast, in traditional nursing homes there was an increase in the prevalence of functional disability, but virtually no change in that of problems requiring medical and skilled nursing care. In home health care the functional care needs of Medicare patients increased significantly, and there was a slight increase in the prevalence of problems requiring medical and skilled nursing care. We conclude that from 1982 to 1986 the needs of patients in long-term care increased substantially. This trend appears to result from Medicare's prospective payment system, which encourages earlier hospital discharge to long-term care settings, and from Medicaid's policy of de-institutionalization. Meeting this greater need for care will be costly. We require a better system of reimbursing for long-term care and ensuring its quality.

Regulation, retrenchment--the DRG experience: problems from changing reimbursement practice

A study of 227,771 discharge abstracts from one U.S. state's short-term, acute care hospitals compares changes in the inpatient market available to the oldest old Medicare patients (85 and older) with those less than 70 and those 70-84 between 1981, the last year when all hospitals were under cost-based reimbursement, and 1984, the first year in which all hospitals were under a prospective payment system based on diagnosis related groups (DRGs). All three populations experienced retrenchment in services as hospitals pursued practice changes to enhance revenue potential. An older, sicker client was admitted as hospitals implemented changes in admission patterns to avoid denial of reimbursement for an admission deemed inappropriate by the Peer Review Organization (PRO). Evidence demonstrates compression in service markets and retrenchment in services for less profitable DRGs and/or cohorts. Inpatient services were reduced the most for the oldest old population although this cohort was the sickest. Changes were observed in utilization of special care units, such as in coronary and intensive care units. Large increases in readmissions in all three cohorts suggests that DRG incentives to reduce length of hospital stay may have promoted premature discharge. Or, perhaps these readmissions resulted from 'unbundling', a practice of splitting patient problems into multiple admissions, as hospitals sought ways to enhance revenue instead of practicing cost-containment. Policy, perceived to be economically stringent, can affect hospital practice and produce undesired results with long-reaching untoward effects on certain segments of the population.

Using severity information for quality assessment: a review of three cases by five severity measures

Five severity measurement systems--APACHE II, Computerized Severity Index, Disease Staging (both Q-Scale and clinical criteria version), MedisGroups, and Patient Management Categories--were applied to three clinical cases. Severity results and recommendations for quality review were then compared and contrasted. Systems that define severity based on resource need produce different impressions than those that define severity in more clinical terms. In-depth quality review is generally suggested when severity scores later in the hospital stay are higher than at admission. Recommendations for review may be automatic or tied to other factors. Some systems use normative data, based on observed severity patterns and patient outcomes, to guide decisions about quality reviews.

Hospital costs and resource characteristics for cardiothoracic surgical hospital deaths

No major changes in the federal Medicare diagnostic-related group (DRG) prospective hospital payment system have been implemented by the United States Congress. We analyzed hospital resource consumption for 1,567 cardiothoracic surgical patients by outcome (ie, survivors versus nonsurvivors). The 76 patients who died had a much greater intensity of hospital resource utilization and represented a substantial financial risk under DRG pricing schemes compared with the 1,491 survivors. Only patients who died within 1 week of admission to the hospital generated a financial surplus under DRGs. A long hospital stay for nonsurvivors produced a substantial deficit (patients with a stay greater than 60 days generated a $154,433 loss per patient). The cardiothoracic patients admitted on an emergency basis who died tended to have a shorter length of stay and represented a lower financial risk under DRGs compared with patients admitted on a nonemergency basis who died. Among nonsurvivors, patients referred for cardiothoracic surgical procedures from other clinical services had lower resource utilization and financial risk under DRGs compared with nonreferrals. These data suggest significant inequities in the current DRG prospective payment system vis-à-vis cardiothoracic surgical patients who die. Variables predictive of greater hospital resource utilization by outcome included a longer hospital stay, nonemergency admission, and admission directly to the cardiothoracic surgical service. Methods to improve the equity of DRG payment vis-à-vis cardiothoracic surgical nonsurvivors should be implemented in the future.

Financial risk and hospital cost in stratified, peripheral vascular surgical DRGs without complications and comorbidities

The purpose of this study was to analyze hospital resource consumption in any of the nine non-complicating conditions covered under stratified, peripheral vascular surgical Diagnosis-Related Groups (DRGs) using the new DRG prospective All Payer System in effect in New York. There is significant health policy debate at the level of the United States Congress regarding the improvement of the equity of DRG hospital payment. Although there have been a number of recommendations by federal advisory bodies to further stratify DRGs by complications and comorbidities, no major revision of DRGs along these lines has been implemented. The all payer system uses the DRG format to pay for all patients. We analyzed 788 peripheral vascular surgical patients by payer (Medicare, Medicaid, Blue Cross and commercial insurance) in these stratified, peripheral vascular surgical DRGs with no complications or comorbidities for a three year period. Our study demonstrated that patients within each DRG with more complications and comorbidities generated higher total hospital costs, a longer hospital length of stay, a greater percentage of procedures per patient, financial risk under DRG payment, more outliers, and a higher mortality, compared to patients in these same DRGs with fewer complications and comorbidities. These findings suggest that new, prospective DRG all payer systems may be inequitable to certain groups of patients or types of hospitals in these stratified peripheral vascular surgical DRGs with no complication or comorbidities. This analysis suggests that these peripheral vascular surgical DRGs should be stratified by the numbers and types of complications and comorbidities to more equitably reimburse hospitals under DRGs.(ABSTRACT TRUNCATED AT 250 WORDS)