Coronary artery bypass surgery versus percutaneous interventions for women with multivessel coronary artery disease.

OBJECTIVE: To compare outcomes in women undergoing percutaneous coronary intervention (PCI) and coronary artery bypass graft (CABG) surgery. DESIGN: This retrospective, propensity-score matched cohort study from the New York State cardiac registry (2012-2018) included all women with multivessel coronary artery disease undergoing PCI with everolimus-eluting stents (EES) and CABG surgery. The primary outcome was all-cause mortality. The key secondary outcome was major adverse cardiac events, defined as the composite of all-cause mortality, myocardial infarction, and stroke. RESULTS: PCI with EES was associated with a higher 6-year risk of mortality (25.75% vs 23.57%; adjusted hazard ratio [AHR], 1.29; 95% confidence interval [CI], 1.14-1.45). PCI also was associated with a higher rate of the composite outcome of death, myocardial infarction, and stroke (36.58% vs 32.89%; AHR, 1.28; 95% CI, 1.17-1.41), as well as myocardial infarction (14.94% vs 9.12%; AHR, 1.84; 95% CI, 1.56-2.17), but not stroke (7.07% vs 7.62%; AHR, 0.83; 95% CI, 0.67-1.03). Repeat revascularization rates also were higher for women undergoing PCI (21.53% vs 11.57%; AHR, 1.88; 95% CI, 1.63-2.17). There was no difference in mortality between the 2 interventions when PCI patients received complete revascularization or had noncomplex lesions and for women without diabetes. CONCLUSIONS: For women with multivessel coronary artery disease, CABG surgery is associated with lower 6-year mortality, myocardial infarction, and repeat revascularization rates compared to PCI with EES.

New York Risk Model and Simplified Risk Score for In-Hospital/30-Day Mortality for Percutaneous Coronary Intervention.

Risk models and risk scores derived from those models require periodic updating to account for changes in procedural performance, patient mix, and new risk factors added to existing systems. No risk model or risk score exists for predicting in-hospital/30-day mortality for percutaneous coronary interventions (PCIs) using contemporary data. This study develops an updated risk model and simplified risk score for in-hospital/30-day mortality following PCI. To accomplish this, New York's Percutaneous Coronary Intervention Reporting System was used to develop a logistic regression model and a simplified risk score model for predicting in-hospital/30-day mortality and to validate both models based on New York data from the previous year. A total of 54,770 PCI patients from 2019 were used to develop the models. Twelve different risk factors and 27 risk factor categories were used in the models. Both models displayed excellent discrimination for the development and validation samples (range from 0.894 to 0.896) and acceptable calibration, but the full logistic model had superior calibration, particularly among higher-risk patients. In conclusion, both the PCI risk model and its simplified risk score model provide excellent discrimination and although the full risk model requires the use of a hand-held device for estimating individual patient risk, it provides somewhat better calibration, especially among higher-risk patients.

Ad Hoc Percutaneous Coronary Intervention in Stable Patients With Multivessel or Unprotected Left Main Disease.

BACKGROUND: There is very little information about the use of ad hoc percutaneous coronary intervention (PCI) in stable patients with multivessel (MV) disease or unprotected left main (LM) disease patients for whom a heart team approach is recommended. OBJECTIVE: To identify the extent of ad hoc PCI utilization for patients with multivessel disease or left main disease, and to explore the inter-hospital variation in ad hoc PCI utilization for those patients. METHODS: New York State's cardiac registries were used to examine the use and variation in use of ad hoc PCI for MV/LM disease as a percentage of all MV/LM PCIs and revascularizations (PCIs plus coronary artery bypass graft procedures) during 2018 to 2019 in New York. RESULTS: After exclusions, 6,425 of the 8,196 stable PCI patients with MV/LM disease (78.4%) underwent ad hoc PCI, ranging from 58.7% for patients with unprotected LM disease to 85.4% for patients with 2-vessel proximal left anterior descending (PLAD) disease. Ad hoc PCIs comprised 35.1% of all revascularizations, ranging from 11.5% for patients with unprotected LM disease to 63.9% for patients with 2-vessel PLAD disease. The risk-adjusted utilization of ad hoc PCI as a percentage of all revascularizations varied widely among hospitals (eg, from 15% in the first quartile to 46% in the last quartile for 3-vessel disease). CONCLUSIONS: Ad hoc PCIs occur frequently even among patients with MV/LM disease. This is particularly true among patients with 2-vessel PLAD disease. The frequency of ad hoc PCIs is lower but still high among patients with diabetes and low ejection fraction and higher in hospitals without surgery on-site (SOS). Given the magnitude of hospital- and physician-level variation in the use of ad hoc PCIs for such patients, consideration should be given to a systems approach to achieving heart team consultation and shared decision making that is consistent for SOS and non-SOS hospitals.

The Neighborhood Atlas Area Deprivation Index For Measuring Socioeconomic Status: An Overemphasis On Home Value.

The Area Deprivation Index (ADI), popularized by the Neighborhood Atlas, is a multifaceted proxy measure for assessing socioeconomic disadvantage that captures social risk factors that are not available in typical clinical registries and that are related to adverse health outcomes. In applying the ADI to New York State, we found that the downstate regions (New York City and its suburbs) were as deprived as or more deprived than the other regions for thirteen of the seventeen ADI variables (all but the ones measured in dollars), but the Neighborhood Atlas-computed overall ADI deprivation was much less in the downstate areas. Numerous census block groups with high home values (indicating low deprivation) accompanied by high deprivation in the other ADI variables had overall ADI scores as computed by the Neighborhood Atlas in the same or contiguous deciles as the home values. We concluded that Neighborhood Atlas-computed ADI scores for New York block groups are mainly representative of median home value. This can be especially problematic when considering quality assessment, funding, and resource allocation in regions with large variations in cost of living, and it may result in underresourcing for disadvantaged communities with high housing prices. We conclude that the Neighborhood Atlas ADI would be more accurate for comparing block groups if variables were standardized before computing the overall index.

Impact of COVID-19 on percutaneous coronary intervention utilization and mortality in New York.

BACKGROUND: COVID-19 has disrupted the care of all patients, and little is known about its impact on the utilization and short-term mortality of percutaneous coronary intervention (PCI) patients, particularly nonemergency patients. METHODS: New York State's PCI registry was used to study the utilization of PCI and the presence of COVID-19 in four patient subgroups ranging in severity from ST-elevation myocardial infarction (STEMI) to elective patients before (December 01, 2018-February 29, 2020) and during the COVID-19 era (March 01, 2020-May 31, 2021), as well as to examine the impact of different COVID severity levels on the mortality of different types of PCI patients. RESULTS: Decreases in the mean quarterly PCI volume from the prepandemic period to the first quarter of the pandemic ranged from 20% for STEMI patients to 61% for elective patients, with the other two subgroups having decreases in between these values. PCI quarterly volume rebounds from the prepandemic period to the second quarter of 2021 were in excess of 90% for all patient subgroups, and 99.7% for elective patients. Existing COVID-19 was rare among PCI patients, ranging from 1.74% for STEMI patients to 3.66% for elective patients. PCI patients with COVID-19 and acute respiratory distress syndrome (ARDS) who were not intubated, and PCI patients with COVID-19 and ARDS who were either intubated or were not intubated because of Do Not Resuscitate//Do Not Intubate status had higher risk-adjusted mortality ([adjusted ORs = 10.81 [4.39, 26.63] and 24.53 [12.06, 49.88], respectively]) than patients who never had COVID-19. CONCLUSIONS: There were large decreases in the utilization of PCI during COVID-19, with the percentage of decrease being highly sensitive to patient acuity. By the second quarter of 2021, prepandemic volumes were nearly restored for all patient subgroups. Very few PCI patients had current COVID-19 throughout the pandemic period, but the number of PCI patients with a COVID-19 history increased steadily during the pandemic. PCI patients with COVID-19 accompanied by ARDS were at much higher risk of short-term mortality than patients who never had COVID-19. COVID-19 without ARDS and history of COVID-19 were not associated with higher mortality for PCI patients as of the second quarter of 2021.

The Association of Socioeconomic Factors With Percutaneous Coronary Intervention Outcomes.

BACKGROUND: Numerous studies have identified the association of socioeconomic factors with outcomes of cardiac surgical procedures. Most have focused on easily measured demographic factors or on socioeconomic characteristics of patients' 5-digit zip codes. The impact of socioeconomic information that is derived from smaller geographic regions has rarely been studied. METHODS: The association of the Area Deprivation Index (ADI) with short-term mortality and readmissions was tested for patients undergoing percutaneous coronary intervention (PCI) in New York while adjusting for numerous patient risk factors, including race, ethnicity, and payer. Changes in hospitals' risk-adjusted outcomes and outlier status with the addition of socioeconomic factors were examined. RESULTS: After adjustment, patients in the 2 most deprived ADI quintiles were more likely to experience in-hospital and 30-day mortality after PCI (adjusted odds ratios [95% confidence intervals] 1.39 [1.18-1.65] and 1.24 [1.03-1.49], respectively), than patients in the first quintile (least deprived). Also, patients in the second and fifth ADI quintiles had higher 30-day readmissions rates than patients in the first quintile (1.12 [1.01-1.25] and 1.17 [1.04-1.32], respectively). Medicare patients had higher mortality and readmission rates, Hispanics had lower mortality, and Medicaid patients had higher readmission rates. CONCLUSIONS: Patients with the most deprived ADIs are more likely to experience short-term mortality and readmissions after PCI. Ethnicity and payer are significantly associated with adverse outcomes even after adjusting for ADI. This information should be considered when identifying patients who are at the highest risk for adverse events after PCI and when risk-adjusting hospital outcomes and assessing quality of care.

The Association of Socioeconomic Factors With Outcomes for Coronary Artery Bypass Surgery.

BACKGROUND: Numerous studies have identified the associations of socioeconomic factors with outcomes of cardiac procedures. The majority have focused on easily measured factors like sex, race, and insurance status, or on socioeconomic characteristics of patients' 5-digit zip codes. The impact of more granular census-derived socioeconomic information on outcomes has rarely been studied. METHODS: The independent impact of the Area Deprivation Index (ADI) on short-term mortality and readmissions was tested on patients undergoing isolated coronary artery bypass grafting (CABG) surgery in New York by using it in logistic regression models in conjunction with patient risk factors and typical disparities measures (race, ethnicity, payer). Changes in hospitals' risk-adjusted outcomes and outlier status with the addition of socioeconomic measures were also tested. RESULTS: After adjusting for numerous patient characteristics, patients in the fourth and fifth highest ADI quintiles (most deprived) were more likely to experience in-hospital/30-day mortality after CABG surgery (adjusted odds ratio [AOR] 1.54, 95% confidence interval [CI] 1.08, 2.20; and AOR 1.50, 95% CI 1.02, 2.21), respectively. ADI was not associated with readmissions, but African Americans (AOR 1.49, 95% CI 1.18, 1.87), Hispanics (AOR 1.33, 95% CI 1.06, 1.65) and Medicaid patients (AOR 1.34, 95% CI 1.09, 1.64) were more likely to be readmitted. CONCLUSIONS: Patients with high ADIs are more likely to experience short-term mortality after CABG surgery. African Americans, Hispanics, and Medicaid patients are more likely to experience 30-day readmissions. This information should be taken into account when monitoring patients to reduce adverse events following surgery, and more studies related to ADI are needed to fully understand its implications.

Hybrid coronary revascularization vs. percutaneous coronary interventions for multivessel coronary artery disease.

OBJECTIVE: Hybrid coronary revascularization (HCR) combines a minimally invasive surgical approach to the left anterior descending (LAD) artery with percutaneous coronary intervention (PCI) for non-LAD diseased coronary arteries. It is associated with shorter hospital lengths of stay and recovery times than conventional coronary artery bypass surgery, but there is little information comparing it to isolated PCI for multivessel disease. Our objective is to compare long-term outcomes of HCR and PCI for patients with multivessel disease. METHODS: This cohort study used data from New York's cardiac surgery and PCI registries in 2010-2016 to examine mortality and repeat revascularization rates for patients with multivessel coronary artery disease who underwent HCR and PCI. Cox proportional hazards methods were used to reduce selection bias. Patients were followed for a median of four years. RESULTS: There was a total of 335 HCR patients (1.2%) and 25,557 PCI patients (98.8%) after exclusions. There was no difference in 6-year risk adjusted survival between HCR and PCI patients (83.17% vs. 81.65%, adjusted hazard ratio (aHR) = 0.90 (95% CI: 0.67-1.20). However, HCR patients were more likely to be free from repeat revascularization in the LAD artery (91.13% vs. 83.59%, aHR = 0.51 (95% CI: 0.34-0.77)). CONCLUSIONS: For patients with multi-vessel coronary artery disease, HCR is rarely performed. There are no differences in mortality rates after four years, but HCR is associated with lower repeat revascularization rates in the LAD artery, presumably due to better longevity in left arterial mammary grafts.

Percutaneous Coronary Intervention for ST-Elevation Myocardial Infarction Before and During COVID in New York.

Little is known about regional differences in volume, treatment, and outcomes of STEMI patients undergoing PCI during the pandemic. The objectives of this study were to compare COVID-19 pandemic and prepandemic periods with respect to regional volumes, outcomes, and treatment of patients undergoing percutaneous coronary intervention (PCI) for ST-elevation myocardial infarction (STEMI) between January 1, 2019 and March 14, 2020 (pre-COVID period) and between March 15, 2020 and April 4, 2020 (COVID period) in 51 New York State hospitals certified to perform PCI. The hospitals were classified as being in either high-density or low-density COVID-19 counties on the basis of deaths/10,000 population. There was a decrease of 43% in procedures/week in high-density COVID-19 counties (p <0.0001) and only 4% in low-density counties (p = 0.64). There was no difference in the change in risk-adjusted in-hospital mortality rates in either type of county, but STEMI PCI patients in high-density counties had longer times from symptom onset to hospital arrival and lower cardiac arrest rates in the pandemic period. In conclusion, the decrease in STEMI PCIs during the pandemic was mainly limited to counties with a high density of COVID-19 deaths. The decrease appears to be primarily related to patients not presenting to hospitals in high-density COVID regions, rather than PCI being avoided in STEMI patients or a reduction in the incidence of STEMI. Also, high-density COVID-19 counties experienced delayed admissions and less severely ill STEMI PCI patients during the pandemic. This information can serve to focus efforts on convincing STEMI patients to seek life-saving hospital care during the pandemic.

Revascularization for Isolated Proximal Left Anterior Descending Artery Disease.

BACKGROUND: Most studies of patients with isolated proximal left anterior descending (PLAD) coronary artery disease do not include all 3 procedural options: percutaneous coronary intervention (PCI), conventional coronary artery bypass graft (CABG) surgery, or minimally invasive CABG. METHODS: New York's cardiac registries were used to identify patients who underwent revascularization for isolated PLAD disease between January 1, 2010, and November 30, 2016, in New York State. After exclusions, 14,327 patients, of whom 13,115 received PCI, 1001 of whom underwent CABG surgery, and 211 of whom underwent minimally invasive CABG were monitored through the end of 2017 to compare outcomes. Registry data were matched to vital statistics data to obtain deaths occurring after discharge and matched to claims data to obtain subsequent admissions for myocardial infarction and stroke. RESULTS: There were no significant differences in mortality or in mortality/myocardial infarction/stroke after 7 years (with median follow-up times in excess of 4 years) among the 3 procedures after adjusting for differences in patient risk factors. However, conventional CABG surgery was associated with a lower subsequent revascularization rate than PCI (adjusted hazard ratio, 0.45; 95% confidence interval, 0.35-0.58) and minimally invasive CABG surgery (adjusted hazard ratio, 0.46; 95% confidence interval, 0.32-0.66). CONCLUSIONS: Among patients with isolated PLAD disease undergoing any of 3 revascularization options (PCI, conventional CABG surgery, or minimally invasive CABG surgery), conventional CABG surgery was associated with lower subsequent revascularization rates, but there were no differences in mortality or mortality/myocardial infarction/stroke rates.

Hybrid Coronary Revascularization Versus Conventional Coronary Artery Bypass Surgery: Utilization and Comparative Outcomes.

BACKGROUND: Hybrid coronary revascularization (HCR) treats multivessel coronary artery disease by combining a minimally invasive surgical approach to the left anterior descending artery with percutaneous coronary intervention for non-left anterior descending diseased coronary arteries. The objective of this study is to compare HCR and conventional coronary artery bypass graft (CABG) surgery medium-term outcomes. METHODS: Data from multivessel disease patients in New York's cardiac surgery and percutaneous coronary intervention registries in 2010 to 2016 were used to compare mortality and repeat revascularization rates for HCR and conventional CABG after using propensity matching to reduce selection bias. RESULTS: There was a total of 303 HCR (0.80%) patients and 37 556 conventional CABG patients after exclusions. After propensity matching, the respective median follow-up times were 3.72 years and 3.76 years. There was no difference between HCR and conventional CABG in survival at 6 years (80.9% versus 85.8%%, adjusted hazard ratio, 1.44 [0.90-2.31]), but HCR had higher mortality excluding deaths during the first year (adjusted hazard ratio, 1.88 [1.10-3.23]). Conventional CABG patients were more likely to be free from repeat revascularization at 6 years than HCR patients (88.2% versus 76.6%; hazard ratio, 2.22 [1.44-3.42]). CONCLUSIONS: HCR is rarely performed for patients with multivessel coronary artery disease. HCR and conventional CABG had no different 6-year mortality rates, but HCR had higher mortality after 1 year and higher rates of subsequent revascularization that were caused by both the need for repeat revascularization in the left anterior descending artery where minimally invasive CABG was performed, and in the coronary arteries where percutaneous coronary intervention was performed. Graphic Abstract: A graphic abstract is available for this article.

Out-of-Hospital 30-day Deaths After Cardiac Surgery Are Often Underreported.

BACKGROUND: Operative mortality (in-hospital during the index admission or within 30 days of the procedure after discharge) is commonly used as a quality of care measure for public reporting of cardiac surgery outcomes, but the ability to capture out-of-hospital deaths accurately remains undetermined. The objective of the study was to estimate the impact of incomplete reporting of out-of-hospital deaths on hospital risk-adjusted mortality and outlier status. METHODS: New York State's 2014 to 2016 cardiac registry data were used to compare the capture of 30-day postprocedure deaths after discharge with and without the use of national and state-level vital statistics data for all 54,442 patients undergoing isolated coronary artery bypass graft, cardiac valve surgery, or both. Hospital risk-adjusted operative mortality rates and mortality outliers were compared based on statistical models that were developed with and without the use of vital statistics data. RESULTS: Thirty-day deaths postprocedure after discharge ranged from 10% to 39% of all operative deaths among cardiac surgical procedures. More than 30% of these deaths were missing without vital statistics confirmation for 7 of the 10 cardiac procedures examined, and more than 40% were missing for 5 of the procedures examined. When vital statistics data were used to confirm 30-day postprocedure deaths after discharge, an additional high outlier for valve surgery was identified. CONCLUSIONS: Operative mortality after cardiac surgery is often underreported owing to a considerable percentage of out-of-hospital cardiac surgery deaths that are missed by reporting centers. This can adversely affect the assessment of hospital risk-adjusted mortality in public reports.

2017 Versus 2012 Appropriate Use Criteria for Percutaneous Coronary Interventions: Impact on Appropriateness Ratings.

OBJECTIVES: The purpose of this study is to revisit cases rated as "inappropriate" in the 2012 appropriate use criteria (AUC) using the 2017 AUC. BACKGROUND: AUC for coronary revascularization in patients with stable ischemic heart disease (SIHD) were released in January 2017. Earlier 2012 AUC identified a relatively high percentage of New York State patients for whom percutaneous coronary intervention (PCI) was rated as "inappropriate" versus optimal medical therapy alone. METHODS: New York State's PCI registry was used to rate inappropriateness of patients undergoing PCI in 2014 using the 2012 and 2017 AUC, and to examine patient characteristics for patients rated differently. RESULTS: A total of 911 of 9,261 (9.8%) patients who underwent PCI in New York State in 2014 with SIHD without prior coronary artery bypass grafting were rated as "inappropriate" using the 2012 AUC, but only 171 (1.8%) patients were rated as "rarely appropriate" ("inappropriate" in 2012 AUC terminology) using the 2017 AUC. A total of 26% of all 8,407 patients undergoing PCI in New York State with 1- to 2-vessel SIHD were without high-risk findings on noninvasive testing and were either asymptomatic or without antianginal therapy. No current or past randomized controlled trials have focused on these patients. CONCLUSIONS: The percentage of 2014 New York State PCI patients with SIHD who are rated "rarely appropriate" has decreased substantially using 2017 AUC in comparison with the older 2012 AUC. However, for many low-risk patients undergoing the procedure, the relative benefits of optimal medical therapy with and without PCI are unknown. Randomized controlled trials are needed to study these groups.

Changes in Percutaneous Coronary Interventions Deemed "Inappropriate" by Appropriate Use Criteria.

BACKGROUND: Recent studies have demonstrated relatively high rates of percutaneous coronary interventions (PCIs) classified as "inappropriate." The New York State Department of Health shared rates with hospitals and announced the intention of withholding reimbursement pending demonstration of clinical rationale for Medicaid patients with inappropriate PCIs. OBJECTIVES: The objective was to examine changes over time in the number and rate of inappropriate PCIs. METHODS: Appropriate use criteria were applied to PCIs performed in New York in patients without acute coronary syndromes or previous coronary artery bypass graft surgery in periods before (2010 through 2011) and after (2012 through 2014) efforts were made to decrease inappropriateness rates. Changes in the number of appropriate PCIs were also assessed. RESULTS: The percentage of inappropriate PCIs for all patients dropped from 18.2% in 2010 to 10.6% in 2014 (from 15.3% to 6.8% for Medicaid patients, and from 18.6% to 11.2% for other patients). The total number of PCIs in patients with no acute coronary syndrome/no prior coronary artery bypass graft surgery that were rated as inappropriate decreased from 2,956 patients in 2010 to 911 patients in 2014, a reduction of 69%. For Medicaid patients, the decrease was from 340 patients to 84 patients, a decrease of 75%. For a select set of higher-risk scenarios, there were higher numbers of appropriate PCIs per year in the period from 2012 to 2014. CONCLUSIONS: The inappropriateness rate for PCIs and the use of PCI for elective procedures in New York has decreased substantially between 2010 and 2014. This decrease has occurred for a large proportion of PCI hospitals.

The Impact of Excluding Shock Patients on Hospital and Physician Risk-Adjusted Mortality Rates for Percutaneous Coronary Interventions: The Implications for Public Reporting.

OBJECTIVES: The authors examined the impact of including shock patients in public reporting of percutaneous coronary intervention (PCI) risk-adjusted mortality. BACKGROUND: There is concern that an unintended consequence of statewide public reporting of medical outcomes is the avoidance of appropriate interventions for high-risk patients. METHODS: New York State's PCI registry was used to compare hospital and physician risk-adjusted mortality rates and outliers from New York's public report models with rates and outliers based on statistical models that include refractory shock patients and exclude both refractory shock and other shock patients. RESULTS: Correlations between the public report model and each of the other 2 models were above 0.92 for hospital risk-adjusted rates and were 0.99 for all physician risk-adjusted rates (p < 0.0001). There were 11 physicians with lower than expected mortality rates (low outliers) and 41 physicians with higher than expected mortality rates (high outliers) across the 3 time periods in the public report, compared with 10 low outliers and 40 high outliers if all shock patients had been excluded. There was considerable overlap among outliers identified by the 3 models. Findings were similar for hospital outliers. CONCLUSIONS: Risk-adjusted hospital and physician mortality rates are highly correlated regardless of whether shock patients are included in public reporting. The numbers of outliers are similar, and outlier changes are minimal, although 10% to 15% of cardiologists who were outliers in either exclusion rule were not outliers in the other one. This information can form a basis for subsequent discussions regarding the exclusion of high-risk patients from public reporting.

Appending Limited Clinical Data to an Administrative Database for Acute Myocardial Infarction Patients: The Impact on the Assessment of Hospital Quality.

BACKGROUND: Hospitals' risk-standardized mortality rates and outlier status (significantly higher/lower rates) are reported by the Centers for Medicare and Medicaid Services (CMS) for acute myocardial infarction (AMI) patients using Medicare claims data. New York now has AMI claims data with blood pressure and heart rate added. OBJECTIVE: The objective of this study was to see whether the appended database yields different hospital assessments than standard claims data. METHODS: New York State clinically appended claims data for AMI were used to create 2 different risk models based on CMS methods: 1 with and 1 without the added clinical data. Model discrimination was compared, and differences between the models in hospital outlier status and tertile status were examined. RESULTS: Mean arterial pressure and heart rate were both significant predictors of mortality in the clinically appended model. The C statistic for the model with the clinical variables added was significantly higher (0.803 vs. 0.773, P<0.001). The model without clinical variables identified 10 low outliers and all of them were percutaneous coronary intervention hospitals. When clinical variables were included in the model, only 6 of those 10 hospitals were low outliers, but there were 2 new low outliers. The model without clinical variables had only 3 high outliers, and the model with clinical variables included identified 2 new high outliers. CONCLUSION: Appending even a small number of clinical data elements to administrative data resulted in a difference in the assessment of hospital mortality outliers for AMI. The strategy of adding limited but important clinical data elements to administrative datasets should be considered when evaluating hospital quality for procedures and other medical conditions.

Assessing hospital performance for acute myocardial infarction: how should emergency department transfers be attributed.

BACKGROUND: The Centers for Medicare and Medicaid Services publicly reports risk-standardized mortality rates (RSMRs) to assess quality of care for hospitals that treat acute myocardial infarction patients, and the outcomes for inpatient transfers are attributed to transferring hospitals. However, emergency department (ED) transfers are currently ignored and therefore attributed to receiving hospitals. METHODS: New York State administrative data were used to develop a statistical model similar to the one used by Centers for Medicare and Medicaid Services to risk-adjust hospital 30-day mortality rates. RSMRs were calculated and outliers were identified when ED transfers were attributed to: (1) the transferring hospital and (2) the receiving hospital. Differences in hospital outlier status and RSMR tertile between the 2 attribution methods were noted for hospitals performing and not performing percutaneous coronary interventions (PCIs). RESULTS: Although both methods of attribution identified 3 high outlier non-PCI hospitals, only 2 of those hospitals were identified by both methods, and each method identified a different hospital as a third outlier. Also, when transfers were attributed to the referring hospital, 1 non-PCI hospital was identified as a low outlier, and no non-PCI hospitals were identified as a low outlier with the other attribution method. About one sixth of all hospitals changed their tertile status. Most PCI hospitals (89%) that changed status moved to a higher (worse RSMR) tertile, whereas the majority of non-PCI hospitals (68%) that changed status were moved to a lower (better) RSMR tertile when ED transfers were attributed to the referring hospital. CONCLUSIONS: Hospital quality assessments for acute myocardial infarction are affected by whether ED transfers are assigned to the transferring or receiving hospital. The pros and cons of this choice should be considered.