Between-hospital Variation Research Articles

Income inequalities in case death of ischaemic heart disease in the Netherlands: a national record-linked study

BackgroundSocioeconomic inequalities in ischaemic heart disease (IHD) mortality have been found in most European countries, but it is unclear to what extent inequalities in survival, as opposed to incidence, contribute...

Hospital-level associations with 30-day patient mortality after cardiac surgery: a tutorial on the application and interpretation of marginal and multilevel logistic regression

BackgroundMarginal and multilevel logistic regression methods can estimate associations between hospital-level factors and patient-level 30-day mortality outcomes after cardiac surgery. However, it is not widely understood how the interpretation of hospital-level effects differs between these methods.MethodsThe Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) registry provided data on 32,354 patients undergoing cardiac surgery in 18 hospitals from 2001 to 2009. The logistic regression methods related 30-day mortality after surgery to hospital characteristics with concurrent adjustment for patient characteristics.ResultsHospital-level mortality rates varied from 1.0% to 4.1% of patients. Ordinary, marginal and multilevel regression methods differed with regard to point estimates and conclusions on statistical significance for hospital-level risk factors; ordinary logistic regression giving inappropriately narrow confidence intervals. The median odds ratio, MOR, from the multilevel model was 1.2 whereas ORs for most patient-level characteristics were of greater magnitude suggesting that unexplained between-hospital variation was not as relevant as patient-level characteristics for understanding mortality rates. For hospital-level characteristics in the multilevel model, 80% interval ORs, IOR-80%, supplemented the usual ORs from the logistic regression. The IOR-80% was (0.8 to 1.8) for academic affiliation and (0.6 to 1.3) for the median annual number of cardiac surgery procedures. The width of these intervals reflected the unexplained variation between hospitals in mortality rates; the inclusion of one in each interval suggested an inability to add meaningfully to explaining variation in mortality rates.ConclusionsMarginal and multilevel models take different approaches to account for correlation between patients within hospitals and they lead to different interpretations for hospital-level odds ratios.

Variations in the Use of an Innovative Technology by Payer

Despite receiving identical reimbursement for treating heart disease patients with bare metal stents (BMS) or drug-eluting coronary stents (DES), cardiologists' use of the new technology (DES) may have varied by patient payer type as DES diffused. Payer-related factors that differ between hospitals and/or differential treatment inside hospitals might explain any overall differences by payer type. To assess the association between payer and DES use and to examine between-hospital and within-hospital variation in DES use over time. We conducted a retrospective analysis of 4.1 million hospitalizations involving DES or BMS from 2003 to 2008 Nationwide Inpatient Sample. We estimated hybrid-fixed effects logit models and calculated the adjusted within-quarter, cross-payer differences in DES use. Coronary stent patients with Medicaid or without insurance were significantly less likely to receive DES than were patients with private insurance throughout the study period. The differences fluctuated over time as the popularity of DES relative to BMS increased and decreased. The within-hospital gaps paralleled the overall differences, and were largest in Q3 2003 (Medicaid: 11.9, uninsured: 10.9% points) and Q4 2008 (Medicaid: 12.8, uninsured: 20.7% points), and smallest in Q4 2004 (Medicaid: 1.4, uninsured: 1.1% points). The between-hospital adjusted differences in DES use by payer were small and rarely significant. We found substantial differences in DES use by payer within hospitals, suggesting physicians selected the new technology for patients in a manner associated with patients' payer type.

Pediatric therapeutic plasma exchange indications and patterns of use in us children's hospitals

Therapeutic plasma exchange (TPE) has been increasingly used over the past decade as a first-line and lifesaving treatment for various conditions classified by the American society for apheresis (ASFA). To date, the degree to which utilization of TPE in pediatrics mirrors recommendations is unknown. Using inpatient administrative data from 42 children's hospitals we conducted an 8-year retrospective cohort study of children (≤18 years) with an international classification of diseases-9-clinical modification (ICD-9-CM) discharge diagnosis indicating an ASFA Category I or II condition, or a procedure code indicating receipt of TPE during hospitalization. TPE was performed during 4,190 hospitalizations of 3,449 patients, of whom 310 (9.0%) and 77 (2.2%) had a primary discharge diagnosis of an ASFA Category I or II condition, respectively. Rates of TPE use for Category I conditions were highest for children with thrombotic thrombocytopenic purpura (TTP), Goodpasture's syndrome, and myasthenia gravis. TPE use in children's hospitals significantly increased from 2003 to 2010, but TPE was performed during only 13.4 and 9.3% of hospitalizations for ASFA Category I and II conditions, respectively. There was significant between-hospital variation in the use of TPE for Category I conditions as a group and individual Category I conditions including TTP. We found low levels of TPE use across hospitals for key indications, including TTP, a condition for which TPE is considered a first-line and life-saving procedure. The variation identified may contribute to varying clinical outcomes between hospitals, warrants further investigation, and represents an important opportunity to improve quality of care.

Delivery of Physical Therapy in the Acute Care Setting: A Population-Based Study

Population-based studies on physical therapy use in acute care are lacking. The purpose of this study was to examine population-based, hospital discharge data from North Carolina to describe the demographic and diagnostic characteristics of individuals who receive physical therapy and, for common diagnostic subgroups, to identify factors associated with the receipt of and intensity of physical therapy use. This was a cross-sectional, descriptive study. Hospital discharge data for 2006-2007 from the 128 acute care hospitals in the state were examined to identify the most common diagnoses that receive physical therapy and to describe the characteristics of physical therapy users. For 2 of the most common diagnoses, logistic and linear regression analyses were conducted to identify factors associated with the receipt and intensity of physical therapy. Of the more than 2 million people treated in acute care hospitals, 22.5% received physical therapy (mean age=66 years; 58% female). Individuals with osteoarthritis (admitted for joint replacement) and stroke were 2 of the most common patient types to receive physical therapy. Almost all individuals admitted for a joint replacement received physical therapy, with little between-hospital variation. Between-hospital variation in physical therapy use for stroke was greater. Demographic and hospital-related factors were associated with physical therapy use and physical therapy intensity for both diagnoses, after controlling for illness severity and comorbidities. Data from only one state were examined, and the studied variables were limited. The use and intensity of physical therapy for stroke and joint replacement in acute care hospitals in North Carolina vary by clinical and nonclinical factors. Reasons behind the association of hospital characteristics and physical therapy use need further investigation.

Abstract 10: Hospital-Quality Following Percutaneous Coronary Intervention In Massachusetts: Experience From 2003-2008

Background: Little information is available on hospital performance following percutaneous coronary intervention (PCI) using all-comer populations. Using a fully Bayesian approach, Massachusetts has publicly reported all-cause in-hospital mortality (IHM) following PCI procedures performed since 04/01/2003 in all non-federal hospitals. Methods: Trained data managers prospectively collected information using the ACC-NCDR instrument supplemented by Mass-DAC specific elements. Operators and data managers adjudicated selected elements via chart review. Bayesian hierarchical regression models of IHM were estimated annually for two cohorts: (1) cardiogenic shock or ST-elevation myocardial infarction (STEMI); and (2) no shock and no STEMI. Posterior risk-standardized IHM rates were estimated. The odds of death for a patient treated at a hospital 1 SD below average quality relative to treatment at a hospital 1 SD above average quality quantified variation. Results: Presence of shock; emergent or salvage status; and compassionate use were changed after adjudication for more than one-quarter of adjudicated cases. No Shock and No STEMI Cohort: Between 04/03-09/08, the number of annual PCI admissions ranged from 11121-14504; number of hospitals from 14-22; and median post-PCI length of stay (LOS) remained unchanged (2 days). The crude IHM declined from 0.76% (81 of 10689) in 2003 to 0.50% (56 in 11275)in 2007. Model discrimination using 9 risk factors was high (ROC = 0.86). The odds of death at lower versus higher quality hospital ranged between 1.38 and 2.14. Shock or STEMI Cohort: the number of PCI admissions ranged from 2606-2800; number of hospitals from 18-24; and median post-PCI LOS remained unchanged (4 days). IHM declined from 6.86% (135 in 1968) in 2003 to 4.78% (130 in 2721)in 2008. Model discrimination was excellent using 7 covariates (ROC = 0.89). The odds of death at lower versus higher quality hospital ranged from 1.49 to 6.35. Conclusions: Key clinical factors, including shock, status of PCI, and compassionate use require adjudication. A small number of risk factors have excellent discrimination of in-hospital survivors at the patient level. Estimates of between-hospital variation can be used to quantify performance at the hospital level.

Impact of Independent Data Adjudication on Hospital-Specific Estimates of Risk-Adjusted Mortality Following Percutaneous Coronary Interventions in Massachusetts

As part of state-mandated public reporting of outcomes after percutaneous coronary interventions (PCIs) in Massachusetts, procedural and clinical data were prospectively collected. Variables associated with higher mortality were audited to ensure accuracy of coding. We examined the impact of adjudication on identifying hospitals with possible deficiencies in the quality of PCI care. From October 2005 to September 2006, 15 721 admissions for PCI occurred in 21 hospitals. Of the 864 high-risk variables from 822 patients audited by committee, 201 were changed, with reassignment to lower acuities in 97 (30%) of the 321 shock cases, 24 (43%) of the 56 salvage cases, and 73 (15%) of the 478 emergent cases. Logistic regression models were used to predict patient-specific in-hospital mortality. Of 241 (1.5%) patients who died after PCI, 30 (12.4%) had a lower predicted mortality with adjudicated than with unadjudicated data. Model accuracy was excellent with either adjudicated or unadjudicated data. Hospital-specific risk-standardized mortality rates were estimated using both adjudicated and unadjudicated data through hierarchical logistic regression. Although adjudication reduced between-hospital variation by one third, risk-standardized mortality rates were similar using unadjudicated and adjudicated data. None of the hospitals were identified as statistical outliers. However, cross-validated posterior-predicted P values calculated with adjudicated data increased the number of borderline hospital outliers compared with unadjudicated data. Independent adjudication of site-reported high-risk features may increase the ability to identify hospitals with higher risk-adjusted mortality after PCI despite having little impact on the accuracy of risk prediction for the entire population.

Studying factors related to pressure ulcers prevention: a marginal scale model for modelling heterogeneity among hospitals

The purpose of this study was to analyse risk factors associated with the presence of pressure ulcer development in patients referred to several Italian hospitals, taking properly into account the within-hospital outcome correlation. We analysed data from a prevalence survey coordinated by the European Pressure Ulcer Advisory Panel on 12,000 hospitalized patients in Italy, collecting information on patient's risk, presence of ulcers and prevention measures. The article describes the bases which generalized estimating equations rely on as well as their statistical properties. The article compares different model specifications in the light of background knowledge of the survey data and model assumptions, and discusses the potential for this modelling approach to apply in similar statistical situations. In accordance with existing literature, factors associated with pressure ulcers in hospitalized patients were identified as Braden scale, age and assistance-connected aspects. Between-hospital variability seemed to be explained by the adopted degree of prevention (use of preventive equipment combined with a repositioning strategy). Modelling the covariance matrix or the scale argument of the correlated binary responses (presence/absence of pressure ulcers) by using moment estimators based on generalized estimating equations prevents optimistic inference and provides an important insight into the role of structural differences among hospitals.

Development and Validation of Hospital “End-of-Life” Treatment Intensity Measures

Health care utilization among decedents is increasingly used as a measure of health care efficiency, but decedent-based measures may be biased estimates of care received by "dying" patients. To develop and validate new measures of hospital "end-of-life" treatment intensity. Retrospective cohort study using Pennsylvania Health Care Cost Containment Council (PHC4) discharge data (April 2001-March 2005) and Centers for Medicare and Medicaid Services (CMS) data (January 1999-December 2003). Patients 65 and older admitted to 174 Pennsylvania acute care hospitals. Hospital-specific standardized ratios of intensive care unit (ICU) and life-sustaining treatment (LST) use among terminal admissions (decedents) and admissions with a high probability of dying, and spending and use of hospitals, ICUs, and physicians among patients in their last 6 months of life. There was marked between-hospital variation in the use of the ICU and LSTs among decedents and admissions with high probability of dying. All hospital decedent and high probability of dying measures were highly correlated (P < 00001). In principal components factor analysis, all 4 of the last-6-months cohort-based measures, the decedent and high-risk admission-based ICU measures, and 8 of the 12 decedent and high probability of dying LST measures loaded onto a single factor, explaining 42% of the variation in the data. Hospitals' end-of-life intensity varies in the use of specific life-sustaining treatments that are somewhat emblematic of aggressive end-of-life care. End-of-life intensity is a relatively stable hospital attribute that is robust to multiple measurement approaches.

Reduction in Acute Myocardial Infarction Mortality in the United States

During the last 2 decades, health care professional, consumer, and payer organizations have sought to improve outcomes for patients hospitalized with acute myocardial infarction (AMI). However, little has been reported about improvements in hospital short-term mortality rates or reductions in between-hospital variation in short-term mortality rates. To estimate hospital-level 30-day risk-standardized mortality rates (RSMRs) for patients discharged with AMI. Observational study using administrative data and a validated risk model to evaluate 3,195,672 discharges in 2,755,370 patients discharged from nonfederal acute care hospitals in the United States between January 1, 1995, and December 31, 2006. Patients were 65 years or older (mean, 78 years) and had at least a 12-month history of fee-for-service enrollment prior to the index hospitalization. Patients discharged alive within 1 day of an admission not against medical advice were excluded, because it is unlikely that these patients had sustained an AMI. Hospital-specific 30-day all-cause RSMR. At the patient level, the odds of dying within 30 days of admission if treated at a hospital 1 SD above the national average relative to that if treated at a hospital 1 SD below the national average were 1.63 (95% CI, 1.60-1.65) in 1995 and 1.56 (95% CI, 1.53-1.60) in 2006. In terms of hospital-specific RSMRs, a decrease from 18.8% in 1995 to 15.8% in 2006 was observed (odds ratio, 0.76; 95% CI, 0.75-0.77). A reduction in between-hospital heterogeneity in the RSMRs was also observed: the coefficient of variation decreased from 11.2% in 1995 to 10.8%, the interquartile range from 2.8% to 2.1%, and the between-hospital variance from 4.4% to 2.9%. Between 1995 and 2006, the risk-standardized hospital mortality rate for Medicare patients discharged with AMI showed a significant decrease, as did between-hospital variation.

Calidad de las variables administrativas del Conjunto Mínimo Básico de Datos de Osakidetza-Servicio Vasco de Salud

IntroductionThe Minimum Basic Data Set (MBDS) is the most used database in the National Health System. The objective of this study is to evaluate the quality, in terms of coding errors of the administrative variables of the MBDS in the Basque Country Health Service. Material and methodsA retrospective, descriptive study by means of reviewing 768 clinical histories from 11 hospitals. Study period: the year 2005. The data obtained in the clinical histories were considered comparison standard and the level of discrepancy with the MBDS was established. The administrative variable measured were: sex, dates of birth, admission, discharge and of the first intervention, admission type and circumstances at discharge, department and physician responsible at discharge. The χ2 was used to measure the between-hospital variability of the percentage discrepancy between the clinical histories and the MBDS records. ResultsThere were 213,091 discharge reports analysed. The percentage errors in the MBDS varied between 0.7% in the sex variable and 24.9% in the physician responsible variable. The variability between hospitals was statistically significant (p=0.001). Good quality was observed in the records of the six variables: sex, date of birth, admission date, discharge date, admission type and medical department responsible. The records were lacking in the date of first intervention and the physician responsible variables.

Measuring and explaining mortality in Dutch hospitals; The Hospital Standardized Mortality Rate between 2003 and 2005

BackgroundIndicators of hospital quality, such as hospital standardized mortality ratios (HSMR), have been used increasingly to assess and improve hospital quality. Our aim has been to describe and explain variation in new HSMRs for the Netherlands.MethodsHSMRs were estimated using data from the complete population of discharged patients during 2003 to 2005. We used binary logistic regression to indirectly standardize for differences in case-mix. Out of a total of 101 hospitals 89 hospitals remained in our explanatory analysis. In this analysis we explored the association between HSMRs and determinants that can and cannot be influenced by hospitals. For this analysis we used a two-level hierarchical linear regression model to explain variation in yearly HSMRs.ResultsThe average HSMR decreased yearly with more than eight percent. The highest HSMR was about twice as high as the lowest HSMR in all years. More than 2/3 of the variation stemmed from between-hospital variation. Year (-), local number of general practitioners (-) and hospital type were significantly associated with the HSMR in all tested models.ConclusionHSMR scores vary substantially between hospitals, while rankings appear stable over time. We find no evidence that the HSMR cannot be used as an indicator to monitor and compare hospital quality. Because the standardization method is indirect, the comparisons are most relevant from a societal perspective but less so from an individual perspective. We find evidence of comparatively higher HSMRs in academic hospitals. This may result from (good quality) high-risk procedures, low quality of care or inadequate case-mix correction.

Hospital and operator variations in drug-eluting stent use: a multi-level analysis of 5967 consecutive patients in Scotland

To determine whether drug-eluting stent (DES) use varies among Scottish hospitals, and the extent to which any variations are explained by differences between operators, patients and lesions. Multi-level analysis of consecutive patients treated with percutaneous coronary intervention (PCI) between April 2005 and March 2006 in Scotland, using the Scottish Coronary Revascularization Registry. A total of 38 operators performed 5967 PCI procedures on 8489 lesions. Crude level of DES use was 47.6%, and the results varied among hospitals (range 30.6-61.8%, chi(2) = 341.6, P < 0.0001). There was significant between-operator variation in the null model. This was attenuated by the addition of hospital as a fixed effect. Nonetheless, the final model demonstrated significant between-operator variability [sigma(2) = 0.486 (0.249-0.971)] and between-hospital variation, after case-mix adjustment. Within Scotland, marked variation existed among hospitals in the use of DES. Operator was the most important factor at patient level, and hospital of treatment, rather than case-mix, was the most important modifier of between-operator variation. Patient selection for DES is complex and may contribute to much of the variations demonstrated. Consensus criteria would provide more detail than is included in current guidance, may aid decision-making for individual patients, reduce opportunity costs and ensure equity of access.

Factors associated with the use of evidence-based therapies after discharge among elderly patients with myocardial infarction

In an accompanying article, we report moderate between-hospital variation in the postdischarge use of beta-blockers, angiotensin-modifying drugs and statins by elderly patients who had been admitted to hospital with acute myocardial infarction. Our objective was to identify the characteristics of patients, physicians, hospitals and communities associated with differences in the use of these medications after discharge. For this retrospective, population-based cohort study, we used linked administrative databases. We examined data for all patients aged 65 years or older who were discharged from hospital in 2005/06 with a diagnosis of myocardial infarction. We determined the effect of patient, physician, hospital and community characteristics on the rate of postdischarge medication use. Increasing patient age was associated with lower postdischarge use of medications. The odds ratios (ORs) for a 1-year increase in age were 0.98 (95% confidence interval [CI] 0.97-0.99) for beta-blockers, 0.97 (95% CI 0.97-0.98) for angiotensin-converting-enzyme inhibitors and angiotensin-receptor blockers and 0.94 (95% CI 0.93-0.95) for statins. Having a general or family practitioner, a general internist or a physician of another specialty as the attending physician, relative to having a cardiologist, was associated with lower postdischarge use of beta-blockers, angiotensin-modifying agents and statins (ORs ranging from 0.46 to 0.82). Having an attending physician with 29 or more years experience, relative to having a physician who had graduated within the past 15 years, was associated with lower use of beta-blockers (OR 0.71, 95% CI 0.60-0.84) and statins (OR 0.81, 95% CI 0.67-0.97). Patients who received care from noncardiologists and physicians with at least 29 years of experience had substantially lower use of evidence-based drug therapies after discharge. Dissemination strategies should be devised to improve the prescribing of evidence-based medications by these physicians.

Assessing the Accuracy of Hospital Clinical Performance Measures

To control costs and improve quality, payers are designing new hospital reimbursement policies that link payment to quality. The authors determine the extent to which quality measures discriminate hospitals into tiers in 2 geographic areas. Administrative and medical record data for patients discharged with acute myocardial infarction (AMI) in 368 California and 81 Massachusetts hospitals. Through simulation, the minimum numbers of patients per hospital needed to identify high-quality hospitals with sensitivity ranging from 75% to 95% under a variety of clinical scenarios are determined. Massachusetts hospitals had twice the number of eligible patients per hospital than California hospitals. Regardless of state, few hospitals had sufficient sample size needed to achieve >85% sensitivity for high-variation quality measures. Reliability of quality-based reimbursement systems relies on the distribution of the hospital sample sizes within geographic areas and the size of practice differences. Selection of conformance thresholds and sensitivity levels depends on the user of the information. To assess the usefulness of performance measures to tier hospitals, information regarding between-hospital variation in quality for specific clinical measures needs to be collected and reported.

Indirect vs. Direct Hospital Quality Indicators for Very-Low-Birth-Weight Infants

An increasing number of insurers today rely on evidence-based selective referral strategies to be sure that care is delivered by high-quality providers. Lacking direct-quality measures based on patient outcomes, current standards for many conditions are based on indirect quality measures such as patient volume. This retrospective study was designed to show whether volume is a useful quality indicator for the care of very-low-birth-weight (VLBW) infants. The study population consisted of 94,110 VLBW infants weighing 501 to 1500 g who were born at 332 U.S. hospitals having neonatal intensive-care units in the years 1995-2000. The average hospital admitted approximately 80 VLBW infants each year, but 25% admitted fewer than 40 per year and 10% admitted fewer than 25. The hospitals, in general, provided a high level of care; most had either a level B or a level C neonatal intensive-care unit. At hospitals admitting fewer than 50 VLBW infants, considered low-volume hospitals, an additional 10 admissions correlated with an 11% reduction in mortality (95% confidence interval [CI], 5-16%; P <0.001). The annual number of admissions explained only 9% of the variation in mortality rates at the participating hospitals. Other easily available hospital features explained another 7% of variation. Because historical volume did not relate significantly to mortality rates in 1999-2000, case volume did not prospectively identify high-quality providers. In contrast, hospitals in the lowest mortality quintile in 1995-1998 had significantly lower mortality rates in 1999-2000 (odds ratio [OR], 0.64; 95% CI, 0.55-0.76). In addition, hospitals in the highest mortality quintile had significantly higher mortality in 1999-2000 (OR, 1.37; 95% CI, 1.16-1.64). Whereas the highest and lowest mortality quintiles explained 34% of between-hospital variation in mortality in 1999-2000, the corresponding figure for the highest and lowest quintiles of patient volume was only 1%. These findings bring into question the referral of VLBW infants based on indirect quality indicators such as patient volume. Direct measures based on outcomes are more useful in guiding selective referral.

Using hierarchical modeling to measure ICU quality

To determine whether hierarchical modeling agrees with conventional logistic regression modeling on the identity of ICU quality outliers within a large multi-institutional database. Retrospective database analysis. Subset of the Project IMPACT database consisting of 40435 adult patients admitted to surgical, medical, and mixed surgical-medical ICUs ( n=55) between 1997 and 1999 who met inclusion criteria for SAPS II. The SAPS II score was customized to this database using conventional logistic regression and using a hierarchical (random coefficients) model. Both models exhibited excellent discrimination ( Cstatistic) and calibration (Hosmer-Lemeshow statistic). The hierarchical and nonhierarchical models had C statistics of.870 and.865, and HL statistics of 3.71 ( p>.88, df=8) and 8.94 ( p>.35, df=8), respectively. Since the random effects component of the hierarchical model accounts for between-hospital variability, only the fixed-effects coefficients were used to calculate the expected mortality rate based on the hierarchical model. The ratio and 95% confidence intervals of the observed to expected mortality rate were calculated using both models for each ICU. ICUs whose observed/expected ratio was either less than 1 or greater than 1, and whose 95% confidence interval did not include 1 were labeled as either high-performance or low-performance outliers, respectively. Analysis using kappa statistic revealed almost perfect agreement between the two models (nonhierarchical vs. hierarchical) on the identity of ICU quality outliers. Models obtained by customizing SAPS II using a nonhierarchical and a hierarchical approach exhibit excellent agreement on the identity of ICU quality outliers.

Risk factors for postoperative death following elective surgical repair of abdominal aortic aneurysm

Abstract Background In regional and national studies death within 30 days of elective open surgical repair of an abdominal aortic aneurysm has been reported to be 7–8 per cent or even higher. Identification of preoperative factors associated with this mortality risk is important for informing surgical policy and may suggest suitable preoperative interventions. Methods During the course of the UK Small Aneurysm trial, 820 patients aged 60–80 years underwent elective open surgical repair of an abdominal aortic aneurysm. The relationship between 30-day mortality rate and 13 prespecified potential prognostic factors (including age, sex, aneurysm diameter, electrocardiographic evidence of myocardial ischaemia, lung function, creatinine level, hospital centre and tube or bifurcated graft) was studied. Results The postoperative mortality rate was 5·6 per cent overall (46 deaths in 820 patients). In univariate analyses postoperative mortality risk was associated significantly with higher serum creatinine (P = 0·002), poor lung function with low forced expiratory volume in 1 s (FEV1) (P = 0·003) and older age (P = 0·03). Evidence of the relationship between age and postoperative death was weaker (P = 0·08) after adjustment for creatinine and FEV1. The predicted postoperative mortality risk ranged from 2·7 per cent in younger patients with a low creatinine level and high FEV1 to 7·8 per cent in older patients with above average creatinine levels and below average FEV1. There was no evidence of between-hospital or between-surgeon variability in postoperative risk. Conclusion Poor preoperative lung and renal function were associated strongly with postoperative death. Age was less important once these two prognostic indicators were taken into account. The potential for preoperative improvement in lung and renal function to reduce the postoperative mortality rate should be addressed in future studies.

Effect of clinical factors on length of stay after coronary artery bypass surgery: Results of the Cooperative Cardiovascular Project

Background Rising health care costs have prompted careful review of comparative hospital resource use. Length of stay after bypass surgery has received particular attention. However, many providers assert that these variations are caused by differences in the clinical mix of patients treated. Our goals were to identify the major clinical predictors of postoperative length of stay (PLOS) after coronary artery bypass graft surgery (CABG), document variations in PLOS among 28 hospitals, and assess the degree to which patient characteristics account for hospital variations in PLOS. Methods Detailed clinical data on 3605 Medicare patients undergoing CABG in 28 Alabama and Iowa hospitals were analyzed by stepwise linear regression to identify significant clinical predictors of PLOS. Analysis of variance was used to compare hospitals’ PLOS while controlling for significant patient risk factors. Results The mean age was 72.1 years, 34.7% were female, and the inhospital mortality rate was 5.6%. The median and mean PLOS were 8 and 11.1 days, respectively. Significant predictors of longer PLOS included increasing age, female sex, history of chronic obstructive pulmonary disease, cerebrovascular disease, or mitral valve disease, elevated admission blood urea nitrogen, and preoperative placement of an intraaortic balloon pump. Hospitals varied significantly ( P = .0001) in their unadjusted PLOS. These hospital-level variations persisted despite adjustment for both preoperative patient characteristics ( P = .0001) and postoperative complications and death ( P = .0001). Conclusions This study found significant between-hospital variations in PLOS that were not explained by patient factors. This finding suggests the potential for increased efficiency in the care of patients undergoing CABG at many institutions. Further research is needed to determine the practice patterns contributing to variations in length of stay after bypass surgery. (Am Heart J 1999;138:69-77.)

Cardiologists' use of clinical information for management decisions for patients with unstable angina: a policy analysis.

Previous studies of management of unstable angina have revealed substantial differences in management between different hospitals, especially with respect to the use of coronary angiography. Physicians in a hospital with angiography facilities were more inclined to perform angiography than were physicians in hospitals without these facilities, even when differences in patient populations were taken into account. The authors compared the management strategies of 18 cardiologists, working in hospitals with and without angiography facilities, using a series of paper-case summaries, in order to assess the contribution of individual variability between physicians to practice differences. Physicians who worked in a hospital with in-house angiography facilities were more inclined to request angiography in similar case summaries, but the inter-individual variation exceeded the between-hospital variation. The variation in individual policies with respect to the decision to initiate coronary angiography could be associated with differences in weighting clinical information. These results confirm that practice variations may have many causes: variability in patients' characteristics, variations in how physicians react to these, differences in the availability of services, and variability in thresholds for action.