Hospital Variation in Carotid Stenting Outcomes

Stability of hospital quality indicators over time: A multi-year observational study of German hospital data

BACKGROUND

Retrospective hospital quality indicators can only be useful if they are trustworthy signals of current or future quality. Despite extensive longitudinal quality indicator data and many hospital quality public reporting initiatives, research on quality indicator stability over time is scarce and skepticism about their usefulness widespread.

OBJECTIVE

Based on aggregated, widely available hospital-level quality indicators, this paper sought to determine whether quality indicators are stable over time. Implications for health policy were drawn and the limited methodological foundation for stability assessments of hospital-level quality indicators enhanced.

METHODS

Two longitudinal datasets (self-reported and routine data), including all hospitals in Germany and covering the period from 2004 to 2017, were analysed. A logistic regression using Generalized Estimating Equations, a time-dependent, graphic quintile representation of risk-adjusted rates and Spearman's rank correlation coefficient were used.

RESULTS

For a total of eight German quality indicators significant stability over time was demonstrated. The probability of remaining in the best quality cluster in the future across all hospitals reached from 46.9% (CI: 42.4-51.6%) for hip replacement reoperations to 80.4% (CI: 76.4-83.8%) for decubitus. Furthermore, graphical descriptive analysis showed that the difference in adverse event rates for the 20% top performing compared to the 20% worst performing hospitals in the two following years is on average between 30% for stroke and AMI and 79% for decubitus. Stability over time has been shown to vary strongly between indicators and treatment areas.

CONCLUSION

Quality indicators were found to have sufficient stability over time for public reporting. Potentially, increasing case volumes per hospital, centralisation of medical services and minimum-quantity regulations may lead to more stable and reliable quality of care indicators. Finally, more robust policy interventions such as outcome-based payment, should only be applied to outcome indicators with a higher level of stability over time. This should be subject to future research.

Editor's Choice - Distribution of Care and Hospital Incidence of Carotid Endarterectomy and Carotid Artery Stenting: A Secondary Analysis of German Hospital Episode Data

OBJECTIVE

This is a description of the German healthcare landscape regarding carotid artery disease, assessment of hospital incidence time courses for carotid endarterectomy (CEA) and carotid artery stenting (CAS), and simulation of potential effects of minimum hospital caseload requirements for CEA and CAS.

METHODS

The study is a secondary data analysis of diagnosis related group statistics data (2005-2016), provided by the German Federal Statistical Office. Cases encoded by German operation procedure codes for CEA or CAS and by International Classification of Diseases (ICD-10) codes for carotid artery disease were included. Hospitals were categorised into quartiles according to annual caseloads. Linear distances to the closest hospital fulfilling hypothetical caseload requirements were calculated.

RESULTS

A total of 132 411 and 33 709 patients treated with CEA and CAS from 2012 to 2016 were included. CEA patients had lower rates of myocardial infarction (1.4% vs. 1.8%) and death (1.2% vs. 4.0%), and CAS patients were more often treated after emergency admission (38.1% vs. 27.1%). Age standardised annual hospital incidences were 67.2 per 100 000 inhabitants for CEA and 16.3 per 100 000 inhabitants for CAS. The incidence for CEA declined from 2005 to 2016, with CAS rising again until 2016 after having declined from 2010 to 2013. Regarding distance from home to hospital, centres offering CEA are distributed more homogeneously across Germany, compared with those performing CAS. Hypothetical introduction of minimum annual caseloads (> 20 for CEA; > 10 for CAS) imply that 75% of the population would reach their hospital after travelling 45 km for CEA and 70 km for CAS.

CONCLUSION

Differences in spatial distribution mean that statutory minimum annual caseloads would have a greater impact on CAS accessibility than CEA in Germany. Presumably because of a decline in carotid artery disease and a transition towards individualised therapy for asymptomatic patients, hospital incidence for CEA has been declining.

Hospital variation of 30-day readmission rate following transcatheter aortic valve implantation

OBJECTIVES

Thirty-day readmission rate is one of the hospital quality metrics. Outcomes of transcatheter aortic valve implantation (TAVI) have improved significantly, but it remains unclear whether hospital-level variance in 30-day readmission rate exists in the contemporary TAVI era.

METHODS

From the 2017 US Nationwide Readmission Database, endovascular TAVI were identified. The unadjusted 30-day readmission rate and 30-day risk-standardised readmission rate (RSRR) were calculated and we then conducted model testing to determine the relative contribution of hospital characteristics, patient-level covariates and economic status to the variation in readmission rates observed between the hospitals.

RESULTS

A total of 44 899 TAVI from 338 hospitals were identified. The range of unadjusted 30-day readmission rate and 30-day RSRR was 2.0%-33.3% and 9.4%-15.3%, respectively. Median 30-day RSRR was 11.8% and there was a significant hospital-level variation (median OR 1.22, 95% CI 1.16 to 1.32, p<0.01) and this was similar in both readmissions caused due to major cardiac and non-cardiac conditions. Patient, hospital and economic factors accounted for 9.6%, 1.9% and 3.8% of the variability in hospital readmission rate, respectively.

CONCLUSIONS

There was significant hospital-level variation in 30-day RSRR following TAVI. Further measures are required to mitigate this variance in the readmission rate.

The Relationship Between Carotid Revascularization Procedural Volume and Perioperative Outcomes in Australia and New Zealand

Carotid endarterectomy (CEA) and carotid artery stenting (CAS) prevent stroke in selected patients. However, each intervention carries a risk of perioperative complications including stroke or death (S/D). We aimed to determine the relationship between operator volume, hospital volume, and the perioperative risk of S/D in carotid revascularization in Australia and New Zealand. Retrospective analysis was performed on prospectively collected data extracted from the Australasian Vascular Audit between 2010 and 2017. Annual caseload volume was analyzed in quintiles (Q) using multivariate regression to assess its impact on perioperative S/D. Carotid endarterectomy procedures (n = 16 765) demonstrated higher S/D rates for lower-volume operators (2.21% for Q1-Q3 [1-17 annual cases] vs 1.76% for Q4-Q5 [18-61 annual cases]; odds ratio [OR]: 1.28; 95% CI: 1.001-1.64; P = .049). Carotid artery stenting procedures (n = 1350) also demonstrated higher S/D rates for lower-volume operators (2.63% for Q1-Q3 [1-11 annual cases] vs 0.37% for Q4-Q5 [12-31 annual cases]; OR: 6.11; 95% CI: 1.27-29.33; P = .024). No significant hospital volume-outcome effect was observed for either procedure. An inverse relationship was demonstrated between operator volume and perioperative S/D rates following CEA and CAS. Consideration of minimum operator thresholds, restructuring of services and networked referral pathways of care in Australia and New Zealand, would likely result in improved patient outcomes.

The Potential Impact of "Take the Volume Pledge" on Outcomes After Carotid Artery Stenting

BACKGROUND

The "Volume Pledge" aims to centralize carotid artery stenting (CAS) to hospitals and surgeons performing ≥10 and ≥5 procedures annually, respectively.

OBJECTIVE

To compare outcomes after CAS between hospitals and surgeons meeting or not meeting the Volume Pledge thresholds.

METHODS

We queried the Nationwide Inpatient Sample for CAS admissions. Hospitals and surgeons were categorized as low volume and high volume (HV) based on the Volume Pledge. Multivariable hierarchical regression models were used to examine the impact of hospital volume (2005-2011) and surgeon volume (2005-2009) on perioperative outcomes.

RESULTS

Between 2005 and 2011, 22 215 patients were identified. Most patients underwent CAS by HV hospitals (86.4%). No differences in poor outcome (composite endpoint of in-hospital mortality, postoperative neurological or cardiac complications) were observed by hospital volume but HV hospitals did decrease the likelihood of other complications, nonroutine discharge, and prolonged hospitalization. From 2005 to 2009, 9454 CAS admissions were associated with physician identifiers. Most patients received CAS by HV surgeons (79.2%). On multivariable analysis, hospital volume was not associated with improved outcomes but HV surgeons decreased odds of poor outcome (odds ratio [OR] 0.76, 95% confidence interval [CI] 0.59-0.97; P = .028), complications (OR 0.56, 95% CI 0.46-0.71, P < .001), nonroutine discharge (OR 0.70, 95% CI 0.57-0.87; P = .001), and prolonged hospitalization (OR 0.52, 95% 0.44-0.61, P < .001).

CONCLUSION

Most patients receive CAS by hospitals and providers meeting the Volume Pledge threshold for CAS. Surgeons but not hospitals who met the policy's volume standards were associated with superior outcomes across all measured outcomes.

Expanding opportunities to understand quality and outcomes of peripheral vascular interventions: The ACC NCDR PVI Registry

Lower extremity peripheral artery disease (PAD) and cerebrovascular disease (CeVD) are prevalent conditions in the United States, and both are associated with significant morbidity (eg, stroke, myocardial infarction, and limb loss) and increased mortality. With a growth in invasive procedures for PAD and CeVD, this demands a more clear responsibility and introduces an opportunity to study how patients are treated and evaluate associated outcomes. The American College of Cardiology (ACC) National Cardiovascular Data Registry (NCDR) Peripheral Vascular Intervention (PVI) Registry is a prospective, independent collection of data elements from individual patients at participating centers, and it is a natural extension of the already robust NCDR infrastructure. As of September 20, 2018, data have been collected on 45,316 lower extremity PVIs, 12,417 carotid artery stenting procedures, and 11,027 carotid endarterectomy procedures at 208 centers in the United States. The purpose of the present report is to describe the patient and procedural characteristics of the overall cohort and the methods used to design and implement the registry. In collecting these data, ACC and ACC PVI Registry have the opportunity to play a pivotal role in scientific evidence generation, medical device surveillance, and creation of best practices for PVI and carotid artery revascularization.

Hospital quality variation matters - A time-trend and cross-section analysis of outcomes in German hospitals from 2006 to 2014

Awareness of care variation and associated differences in outcome quality is important for patients to recognize and leverage the benefits of hospital choice and for policy makers, providers, and suppliers to adapt initiatives to improve hospital quality of care. We examine panel data on outcome quality in German hospitals between 2006 and 2014 for cholecystectomy, pacemaker implantation, hip replacement, percutaneous coronary intervention (PCI), stroke, and acute myocardial infarction (AMI). We use risk-adjusted and unadjusted outcomes based on 16 indicators. Median outcome and outcome variation trends are examined via box plots, simple linear regressions and quintile differences. Outcome trends differ across treatment areas and indicators. We found positive quality trends for hip replacement surgery, stroke and AMI 30-day mortality, and negative quality trends for 90-day stroke and AMI readmissions and PCI inpatient mortality. Variation of risk-adjusted outcomes ranges by a factor of 3-12 between the 2nd and 5th quintile of hospitals, both at the national and regional level. Our results show that simply measuring and reporting hospital outcomes without clear incentives or regulation - "carrots and sticks" - to improve performance and to centralize care in high performing hospitals has not led to broad quality improvements. More substantial efforts must be undertaken to narrow the outcome spread between high- and low-quality hospitals.

Impact of hospital volume on long-term neurological outcome in patients undergoing carotid artery stenting

BACKGROUND

The impact of hospital volume on long-term outcome after carotid artery stenting (CAS) remains unknown.

OBJECTIVES

We designed a nationwide cohort study to elucidate the impact of hospital volume on the incidence of stroke after CAS.

METHODS

The Taiwan National Health Insurance Research database was used to identify all patients admitted for CAS from 2008 to 2012. We defined high-volume hospitals as those performing more than 20 CAS per year. The primary outcome was new ischemic stroke after discharging from the index CAS. Propensity score-matching was performed to create two matched groups for comparison.

RESULTS

A total of 3,248 patients underwent 3,576 CAS procedures were enrolled. There were 56 hospitals performing CAS during the study period. Among these 3,248 patients, 2,226 (68.5%) were performed in high-volume hospitals. A propensity score-matching created two groups with 1,000 patients in each group. During a median of 2.06 years follow-up, 35 (3.5%) and 52 (5.2%) patients in high-volume hospitals and low-volume hospitals developed new ischemic stroke 30 days after discharging from the index CAS, respectively (for low-volume hospitals, HR 1.50, 95%CI 1.06-2.12, P = 0.023). The use of embolic protection device did not result in different periprocedural or postdischarge strokes. The periprocedural (within 30 days after CAS) ischemic stroke or all-cause mortality rates during follow-up period were similar between two groups.

CONCLUSIONS

CAS performed in high-volume hospitals was associated with less new ischemic stroke after discharging from the index CAS, compared to those in low-volume hospitals. © 2017 Wiley Periodicals, Inc.

Significant Association of Annual Hospital Volume With the Risk of Inhospital Stroke or Death Following Carotid Endarterectomy but Likely Not After Carotid Stenting

Background—

Associations between hospital volume and the risk of stroke or death following carotid endarterectomy (CEA) and carotid artery stenting (CAS) on a national level in Germany were analyzed.

Methods and Results—

Secondary data analysis using microdata from the nationwide statutory German quality assurance database on all surgical or endovascular carotid interventions on the extracranial carotid artery between 2009 and 2014. Hospitals were categorized into empirically determined quintiles according to the annual case volume. The resulting volume thresholds were 10, 25, 46, and 79 for CEA and 2, 6, 12, and 26 for CAS procedures. The primary outcome was any stroke or death before hospital discharge. For risk-adjusted analyses, a multilevel regression model was applied. The analysis included 161 448 CEA and 17 575 CAS procedures. In CEA patients, the crude risk of stroke or death decreased monotonically from 4.2% (95% confidence interval, 3.6%–4.9%) in low-volume hospitals (first quintile 1–10 CEA per year) to 2.1% (2.0%–2.2%) in hospitals providing ≥80 CEA per year (fifth quintile; P <0.001 for trend). The overall risk of any stroke or death in CAS patients was 3.7% (3.5%–4.0%), but no trend on annual volume was seen ( P =0.304). Risk-adjusted analyses confirmed a significant inverse relationship between hospital volume (categorized or continuous) and the risk of stroke or death after CEA but not CAS procedures.

Conclusions—

An inverse volume–outcome relationship in CEA-treated patients was demonstrated. No significant association between hospital volume and the risk of stroke or death was found for CAS.