Continuous assessment and improvement in quality of care. A model from the Department of Veterans Affairs Cardiac Surgery

Remote monitoring following adult cardiac surgery: A paradigm shift?

Background

The Perfect Care (PC) initiative engages, educates, and enrolls adult cardiac surgery patients into a transformational program that includes an app for appointment scheduling, tracking biometric data and patient-reported outcomes, audiovisual visits, and messaging, paired with a digital health kit (consisting of a fitness tracker, scale, and sphygmomanometer). PC aims to reduce postoperative length of stay (LOS) as well as 30-day readmission and mortality.

Methods

This was a retrospective review of patients who underwent coronary artery bypass (CAB), valve, or combined CAB and valve procedures at either of the 2 participating hospitals between April 2018 and March 2022. Patients who participated in the PC quality improvement initiative were compared to propensity-matched controls (1:1 matching). The evaluation focused on postoperative LOS and a novel composite measure comprising 30-day readmission and mortality.

Results

Remote monitoring (PC) was associated with a shorter postoperative LOS, lower combined rate of 30-day readmission and mortality, and less variation compared to matched non-PC controls.

Conclusions

Integrated improvements in postoperative remote monitoring of adult cardiac surgery patients may reduce time in the hospital and post-acute care facilities. Future prioritized efforts include the development of additional, personalized biometric monitoring devices, use of biometric data to augment risk assessment, and investigation of the value of remote monitoring on various patient risk profiles to address potential disparities in care.

Outcomes of intermediate-risk patients treated with transcatheter and surgical aortic valve replacement in the Veterans Affairs Healthcare System: A single center 20-year experience

BACKGROUND

Transfemoral transcatheter aortic valve replacement (TAVR) was superior to surgical aortic valve replacement (SAVR) in the placement of aortic transcatheter valves (PARTNER) 2A trial (P2). The generalizability of the trial results to the broader population of patients with intermediate surgical risk remains unknown.

OBJECTIVE

To compare the outcomes of SAVR and TAVR among patients with intermediate surgical risk treated in the VA Healthcare System.

METHODS

We retrospectively analyzed the clinical characteristics and outcomes on all SAVR (1987-2014) and TAVR procedures (2015-2017) performed at the Minneapolis VA Healthcare System. Patients were divided into three groups based on their estimated 30-day mortality risk. The primary outcome was a composite of death or stroke at 30-days.

RESULTS

A total of 1,049 patients underwent SAVR with (n = 468, 45%) or without CABG (n = 581, 55%) and 110 underwent TAVR during the study period. Intermediate-risk patients represented 29.4% and 40% of patients undergoing SAVR and TAVR, respectively. The predicted 30-day mortality risk of intermediate-risk patients was 5.5% for the SAVR group and 5.2% for the TAVR group (P = 0.54). The observed combined rate of stroke or death at 30-days for intermediate-risk patients treated with SAVR and TAVR was 11% and 2.2%, respectively (P = 0.05). The results for SAVR and TAVR at the VA were comparable to the P2 trial and STS database (all P = NS). The results did not change when the analysis was restricted to a more contemporary (2005-2014) surgical cohort or isolated SAVR. The number needed to treat to prevent one death/stroke with TAVR was 10.

CONCLUSIONS

Adoption of TAVR as the preferred treatment modality in intermediate-risk patients may result in significant improvements in morbidity and mortality.

Quality Management and Key Performance Indicators in Oncologic Esophageal Surgery

Ranking systems and comparisons of quality and performance indicators will be of increasing relevance for complex "high-risk" procedures such as esophageal cancer surgery. The identification of evidence-based standards relevant for key performance indicators in esophageal surgery is essential for establishing monitoring systems and furthermore a requirement to enhance treatment quality. In the course of this review, we analyze the key performance indicators case volume, radicality of resection, and postoperative morbidity and mortality, leading to continuous quality improvement. Ranking systems established on this basis will gain increased relevance in highly complex procedures within the national and international comparison and furthermore improve the treatment of patients with esophageal carcinoma.

The validity of the VA surgical risk tool in predicting postoperative mortality among octogenarians

BACKGROUND

To examine the validity of Veterans Affair-VA risk assessment tool in predicting the perioperative and overall mortality among octogenarians.

METHODS

This is a single-institution retrospective observational study, in which the clinical information of 1,618 octogenarians were extracted from the VA Surgical Quality Improvement Program database. VA risk assessment tool and ASA classification were used to predict the probability of postoperative mortality and morbidity. Multiple risk groups were compared for mortality using multiple logistic regressions.

RESULTS

There were 570 survivors and 1,048 nonsurvivors. VA risk tool strongly predicted perioperative 30-day mortality in receiver operator characteristic curve analysis (area under the curve: .82 ± .02). The power of this tool, while acceptable, was less in predicting overall mortality (area under the curve: .68 ± .01). Age, dialysis, a history of congestive heart failure, functional status, transfusion, and weight loss were also associated with increased rate of death within 30 days.

CONCLUSIONS

VA risk tool predicted both perioperative and overall mortality. Relatively strong power of this tool in predicting overall mortality may be unique to this age group because of their advanced age.

The MELD system and liver transplant waiting-list mortality in developing countries: lessons learned from São Paulo, Brazil

OBJECTIVE

The MELD system has not yet been tested as an allocation tool for liver transplantation in the developing countries. In 2006, MELD (Model for End-stage Liver Disease) was launched as a new liver allocation system in São Paulo, Brazil. This study was designed to assess the results of the new allocation policy on waiting list mortality.

METHODS

The State of São Paulo liver transplant database was retrospectively reviewed from July 2003 through July 2009. Patients were divided into those who were transplanted before (Pre-MELD Group) and those who were transplanted after (post-MELD Group) the implementation of the MELD system. Only adult liver transplant candidates were included. Waiting list mortality was the primary endpoint.

RESULTS

The unadjusted death rate in waiting list decreased significantly after the implementation of the MELD system (from 91.2 to 33.5/1,000 patients per year; p<0.0001). Multivariate analysis showed a significant drop in risk of waiting list death for post-MELD patients (HR 0.34; p<0.0001). Currently, 48% of patients are transplanted within 1-year of listing (versus 23% in the pre-MELD era; p<0.0001). Patient and graft survival did not change with MELD implementation.

CONCLUSION

There was a reduction in waiting time and list mortality after implementation of the MELD system in São Paulo. Patients listed in the post-MELD era had a significant reduction in risk for the waiting list mortality. There were no changes in post-transplant outcomes. MELD can be successfully utilized for liver transplant allocation in developing countries.

Risk-adjusted comparison of blood pressure and low-density lipoprotein (LDL) noncontrol in primary care offices

OBJECTIVES

Population-level control of modifiable cardiovascular disease (CVD) risk factors is suboptimal. The objectives of this study were (1) to demonstrate the use of electronically downloaded electronic health record (EHR) data to assess guideline concordance in a large cohort of primary care patients, (2) to provide a contemporary assessment of blood pressure (BP) and low-density lipoprotein (LDL) noncontrol in primary care, and (3) to demonstrate the effect of risk adjustment of rates of noncontrol of BP and LDL for differences in patient mix on these clinic-level performance measures.

METHODS

This was an observational comparative effectiveness study that included 232,172 adult patients ≥18 years old with ≥1 visit within 2 years in 33 primary care clinics with EHRs. The main measures were rates of BP and LDL noncontrol based on current guidelines and were calculated from electronically downloaded EHR data. Rates of noncontrol were risk-adjusted using multivariable models of patient-level variables.

RESULTS

Overall, 16.0% of the 227,122 patients with known BP and 14.9% of the 136,771 patients with known LDL were uncontrolled. Clinic-level, risk-adjusted BP noncontrol ranged from 7.7% to 26.5%, whereas that for LDL ranged from 5.8% to 23.6%. Rates of noncontrol exceeded an achievable benchmark for 85% (n = 28) and 79% (n = 26) of the 33 clinics for BP and LDL, respectively. Risk adjustment significantly influences clinic rank order for rate of noncontrol.

CONCLUSIONS

We demonstrated that the use of electronic collection of data from a large cohort of patients from fee-for-service primary care clinics is feasible for the audit of and feedback on BP and LDL noncontrol. Rates of noncontrol for most clinics are substantially higher than those achievable. Risk adjustment of noncontrol rates results in a rank-order of clinics very different from that achieved with nonadjusted data.

Impact of functional status on survival after coronary artery bypass grafting in a veteran population

BACKGROUND

Although functional impairment has been shown to be an adverse outcome of frailty, little is known of its effect on patients after cardiac operations. We aimed to assess the effect of limited functional status on long-term survival after coronary artery bypass grafting (CABG).

METHODS

We reviewed prospectively gathered data from 1,503 consecutive patients who underwent isolated CABG between 1997 and 2009. We compared the outcomes of 318 patients with limited functional status and 1,185 patients without any functional impairment. The mean follow-up period was 65 months (range, 1 to 157 months). We assessed the relationship between functional status impairment and long-term survival by Cox regression analysis adjusted for confounding factors.

RESULTS

Functionally impaired patients were slightly older (63±9 vs 62±8 years, p=0.05) and had more risk factors for adverse outcomes than patients who were functionally unimpaired. After adjustment for potential confounding variables by multivariate logistic regression analysis, preoperative limited functional status was not an independent predictor (odds ratio [95% confidence interval]) of 30-day mortality (1.4 [0.3 to 5.8], p=0.67) or major adverse cardiac events (1.3 [0.5 to 3.3], p=0.71), nor was it predictive of reduced long-term survival (10-year hazard ratio 1.0 [0.7 to 1.4], p=0.85).

CONCLUSIONS

Limited functional status was not an independent risk factor for early postoperative complications or death. Long-term survival in patients whose functional status was impaired before they underwent CABG was similar to that of patients who were functionally independent.

Measuring sustainability within the Veterans Administration Mental Health System Redesign initiative

OBJECTIVE

To examine how attributes affecting sustainability differ across Veterans Health Administration organizational components and by staff characteristics.

SUBJECTS

Surveys of 870 change team members and 50 staff interviews within the Veterans Affairs' Mental Health System Redesign initiative.

METHODS

A 1-way ANOVA with a Tukey post hoc test examined differences in sustainability by Veteran Integrated Service Networks, job classification, and tenure from staff survey data of the Sustainability Index. Qualitative interviews used an iterative process to identify "a priori" and "in vivo" themes. A simple stepwise linear regression explored predictors of sustainability.

RESULTS

Sustainability differed across Veteran Integrated Service Networks and staff tenure. Job classification differences existed for the following: (1) benefits and credibility of the change and (2) staff involvement and attitudes toward change. Sustainability barriers were staff and institutional resistance and nonsupportive leadership. Facilitators were commitment to veterans, strong leadership, and use of quality improvement tools. Sustainability predictors were outcomes tracking, regular reporting, and use of Plan, Do, Study, Adjust cycles.

CONCLUSIONS

Creating homogeneous implementation and sustainability processes across a national health system is difficult. Despite the Veterans Affairs' best evidence-based implementation efforts, there was significant variance. Locally tailored interventions might better support sustainability than "one-size-fits-all" approaches. Further research is needed to understand how participation in a quality improvement collaborative affects sustainability.

Evolution of quality and surgical risk assessment in the USA

As health-care reforms progress, quality and risk assessment in the health-care system of the USA surface as critical issues. This review considers past, present and possible future changes in quality assessment along with formal programs of complication reduction and pay for performance (PFP) as related to surgery and vascular interventions. Strategies for quality improvement include aggregate and risk-adjusted outcome measurement, process compliance with the Surgical Complication Improvement Program, oversight and PFP, now policies of the Centers for Medicare and Medicaid Services (CMS). Advantages, disadvantages and unintended consequences of these policies are discussed. While ongoing system changes will influence vascular surgical practice, unique opportunities and obligations exist for vascular surgeons to contribute to quality assessment of their interventions, to evaluate long-term outcomes and to devise strategies for comprehensive cost-effective care for the conditions affecting patients with vascular disease.

Hypoalbuminemia and long-term survival after coronary artery bypass: a propensity score analysis

BACKGROUND

Hypoalbuminemia is associated with increased morbidity in surgical patients. The impact of low albumin level on survival in cardiac surgical patients is unknown. We hypothesized that a low preoperative albumin level negatively affects long-term survival after coronary artery bypass graft (CABG) surgery.

METHODS

We reviewed prospectively gathered data from the records of 1,164 consecutive patients who underwent primary isolated CABG at our institution between 1997 and 2007. Propensity score analysis of 18 preoperative and intraoperative variables balanced potential confounding factors between the two groups of patients, so that the final study cohort consisted of 588 patients: 294 with a preoperative albumin level less than 3.5 g/dL (ie, hypoalbuminemia) and 294 patients with a preoperative albumin level of 3.5 g/dL or greater. We assessed long-term survival by using Kaplan-Meier curves generated by log rank tests.

RESULTS

The two groups of patients were well matched in terms of preoperative and intraoperative covariates. Both groups had similar early outcomes, including 30-day mortality rates (2.0% versus 1.7%; p = 0. 76) and the incidence of major adverse cardiac events (2.7% versus 2.7%; p = 1.0). However, patients with hypoalbuminemia had a significantly worse 8-year survival rate (65% ± 7% versus 86% ± 3%; hazard ratio 2.2; 95% confidence interval: 1.4 to 3.6; p = 0.001) than patients without hypoalbuminemia.

CONCLUSIONS

Although preoperative hypoalbuminemia did not predict increased early postoperative mortality or morbidity in CABG patients, it did independently predict poor long-term survival after CABG. Identifying the mechanism that underlies this relationship is essential in improving overall survival among patients with low serum albumin levels who are undergoing surgical myocardial revascularization.

Alcohol screening and risk of postoperative complications in male VA patients undergoing major non-cardiac surgery

BACKGROUND

Patients who misuse alcohol are at increased risk for surgical complications. Four weeks of preoperative abstinence decreases the risk of complications, but practical approaches for early preoperative identification of alcohol misuse are needed.

OBJECTIVE

To evaluate whether results of alcohol screening with the Alcohol Use Disorders Identification Test - Consumption (AUDIT-C) questionnaire-up to a year before surgery-were associated with the risk of postoperative complications.

DESIGN

This is a cohort study.

SETTING AND PARTICIPANTS

Male Veterans Affairs (VA) patients were eligible if they had major noncardiac surgery assessed by the VA's Surgical Quality Improvement Program (VASQIP) in fiscal years 2004-2006, and completed the AUDIT-C alcohol screening questionnaire (0-12 points) on a mailed survey within 1 year before surgery.

MAIN OUTCOME MEASURE

One or more postoperative complication(s) within 30 days of surgery based on VASQIP nurse medical record reviews.

RESULTS

Among 9,176 eligible men, 16.3% screened positive for alcohol misuse with AUDIT-C scores ≥ 5, and 7.8% had postoperative complications. Patients with AUDIT-C scores ≥ 5 were at significantly increased risk for postoperative complications, compared to patients who drank less. In analyses adjusted for age, smoking, and days from screening to surgery, the estimated prevalence of postoperative complications increased from 5.6% (95% CI 4.8-6.6%) in patients with AUDIT-C scores 1-4, to 7.9% (6.3-9.7%) in patients with AUDIT-Cs 5-8, 9.7% (6.6-14.1%) in patients with AUDIT-Cs 9-10 and 14.0% (8.9-21.3%) in patients with AUDIT-Cs 11-12. In fully-adjusted analyses that included preoperative covariates potentially in the causal pathway between alcohol misuse and complications, the estimated prevalence of postoperative complications increased significantly from 4.8% (4.1-5.7%) in patients with AUDIT-C scores 1-4, to 6.9% (5.5-8.7%) in patients with AUDIT-Cs 5-8 and 7.5% (5.0-11.3%) among those with AUDIT-Cs 9-10.

CONCLUSIONS

AUDIT-C scores of 5 or more up to a year before surgery were associated with increased postoperative complications.

The impact of obesity on long-term survival after coronary artery bypass grafting

BACKGROUND

Obesity is a well-known risk factor for coronary artery disease. The objective of our study was to examine the impact of obesity on long-term survival after coronary artery bypass grafting (CABG).

MATERIALS AND METHODS

Using prospectively gathered data, we reviewed records of 1163 consecutive patients who underwent isolated primary CABG between 1997 and 2007. We compared outcomes of obese patients (body mass index [BMI] > or = 30 kg/m(2); n = 472) and non-obese patients (BMI < 30 kg/m(2); n = 691). Long-term survival was assessed by using Kaplan-Meier curves generated by log-rank tests and adjusted for confounding factors with Cox logistic regression analysis.

RESULTS

Obese patients were slightly younger (60 +/- 8 versus 63 +/- 9y; P < 0.0001), were less likely to be current tobacco smokers (30% versus 41%; P < 0.0001), had a higher incidence of diabetes (51% versus 33%; P < 0.0001), and had a lower incidence of cerebral vascular disease (18% versus 24%; P = 0.009) than non-obese patients. The two groups of patients had similar 30-d rates of mortality (1.3% versus 1.5%; P = 0.8) and major adverse cardiac events (2.3% versus 2.5%; P = 0.9). Adjusted Cox regression survival curves were also similar between the two groups of patients (adjusted hazard ratio, 1.2; 95% confidence interval, 0.8-1.8; P = 0.28).

CONCLUSIONS

Obese patients who underwent CABG had 30-d mortality rates and early outcomes similar to those of non-obese patients. Long-term survival was also similar between these two groups of patients after adjustment for confounding variables.

The impact of placing multiple grafts to each myocardial territory on long-term survival after coronary artery bypass grafting

OBJECTIVE

Incomplete myocardial revascularization decreases survival for patients undergoing coronary artery bypass grafting. The effects of constructing multiple grafts to each major diseased artery territory are unknown. We aimed to determine the impact on long-term survival after coronary artery bypass grafting of placing multiple grafts to each myocardial territory.

METHODS

We reviewed data from 1129 consecutive patients who underwent coronary artery bypass grafting at our institution between 1997 and 2007 and compared outcomes between patients who received multiple grafts to each major diseased artery territory (n = 549) with those of patients who received single grafts to each territory (n = 580). We assessed long-term survival with Kaplan-Meier curves generated by log-rank tests, adjusting for confounding factors with Cox proportional hazards regression analysis.

RESULTS

Patients who received multiple grafts to each major diseased artery territory had longer cardiopulmonary bypass and aortic crossclamp times than patients who received single grafts per territory. Patient groups had similar early outcomes, including 30-day mortalities (1.3% vs 1.4%, P > .999) and incidences of major adverse cardiac events (2.9% vs 2.2%, P = .57). Cox regression 10-year survival curves were also similar between groups (adjusted hazard ratio 0.94, 95% confidence interval 0.67-1.34, P = .74).

CONCLUSION

Patients who received multiple grafts to each major diseased artery territory had early outcomes similar to those who received single grafts per territory. Constructing multiple grafts to each major diseased artery territory increases operative time and does not improve long-term survival.

Ranking of trauma center performance: the bare essentials

BACKGROUND

Evaluation of trauma center performance has been limited to comparisons of observed versus expected mortality using trauma and injury severity score methodology. Few studies have focused on identifying top performers. In part, this is due to the perceived need for extensive data required to adequately risk adjust. We set out to identify the patient and injury-related factors that most affect case-mix across centers and thus are most likely to alter assessments of hospital performance.

METHODS

One hundred ninety trauma centers contributing data to the National Trauma Databank (NTDB) during 2004 to 2005 were used for hospital rankings (n = 169,929 patients). Trauma centers were ranked by crude mortality. We then added variables [injury severity score {ISS}, systolic blood pressure {SBP}, mechanism, age, gender, comorbidities, body region abbreviated injury scale {AIS}] singly to a risk-adjustment model to obtain adjusted probability of death. Trauma centers were then ranked again. The variable that affected rankings the greatest was kept and the process was repeated in an iterative fashion until the incremental change in ranks was minimal.

RESULTS

ISS accounted for the most variation in mortality rates across trauma centers, shown by the large rank change with addition of ISS to the model. Specifically, when ISS was taken into consideration, 92% of trauma centers changed their rank by >/=3 and almost half their quartile rank by at least 1. In lesser order of importance, age, SBP, head AIS, mechanism, gender, and abdominal AIS were relevant to adjust for case mix.

CONCLUSIONS

Trauma center rankings are affected by few parameters, reflecting their relationship to mortality and their relative frequencies. Complex risk adjustment methodology is not required to address differences in case mix. Data abstraction for the purpose of comparing trauma center performance should focus on ensuring that at minimum, these variables are collected with a high degree of accuracy.

Relationship Between Preventability of Death After Coronary Artery Bypass Graft Surgery and All-Cause Risk-Adjusted Mortality Rates

Background— The goal of this study was to determine the relationship between all-cause, risk-adjusted, in-hospital mortality after coronary artery bypass graft surgery and the proportion of preventable in-hospital deaths as a measure of quality of care at an institution level.

Methods and Results— We conducted a retrospective analysis of 347 randomly selected in-hospital deaths after isolated coronary artery bypass graft surgery at 9 institutions in Ontario over the period of 1998 to 2003. Nurse-abstracted chart summaries were reviewed by 2 experienced cardiac surgeons who were blinded to patient, surgeon, and hospital and used a standardized implicit tool to identify preventable death. A third reviewer reassessed all cases in which the first 2 reviewers disagreed. Rates of preventable deaths were estimated for each hospital and compared with all-cause mortality rates. A structured adverse event audit completed by each surgeon-reviewer was used to identify quality improvement opportunities for the preventable deaths. A total of 111 of 347 deaths (32%) were judged preventable despite a low risk-adjusted mortality range (1.3% to 3.1%) across hospitals. No significant correlation was found between all-cause, risk-adjusted in-hospital mortality rates and the proportion of preventable deaths at the hospital level (Spearman coefficient, −0.42; P =0.26). A large proportion of preventable deaths were related to problems in the operating room (86%) and intensive care unit (61%). Many deaths were associated with deviations in perioperative care (32% based on concurrence of 2 reviewers, and another 42% in cases in which 1 reviewer reached that opinion).

Conclusions— Approximately one third of in-hospital coronary artery bypass graft deaths were judged preventable by surgeon reviewers. All-cause risk-adjusted mortality rates are convenient measures of institutional quality of care but were not correlated with preventable mortality in our jurisdiction. Providers should conduct detailed adverse event audits to drive meaningful improvements in quality.

Mortality of Department of Veterans Affairs patients undergoing coronary revascularization in private sector hospitals

OBJECTIVE

A limitation of studies comparing outcomes of Veterans Affairs (VA) and private sector hospitals is uncertainty about the methods of accounting for risk factors in VA populations. This study estimates whether use of VA services is a marker for increased risk by comparing outcomes of VA users and other patients undergoing coronary revascularization in private sector hospitals.

DATA SOURCES

Males 67 years and older undergoing coronary artery bypass graft (CABG; n=687,936) surgery or percutaneous coronary intervention (PCI; n=664,124) during 1996-2002 were identified from Medicare administrative data. Patients using VA services during the 2 years preceding the Medicare admission were identified using VA administrative files.

STUDY DESIGN

Thirty-, 90-, and 365-day mortality were compared in patients who did and did not use VA services, adjusting for demographic and clinical risk factors using generalized estimating equations and propensity score analysis.

RESULTS

Adjusted mortality after CABG was higher (p<.001) in VA users compared with nonusers at 30, 90, and 365 days: odds ratio (OR)=1.07 (95 percent confidence interval [CI], 1.03-1.11), 1.07 (95 percent CI, 1.04-1.10), and 1.09 (95 percent CI, 1.06-1.12), respectively. For PCI, mortality at 30 and 90 days was similar (p>.05) for VA users and nonusers, but was higher at 365 days (OR=1.09; 95 percent CI, 1.06-1.12). The increased risk of death in VA users was limited to patients with service-connected disabilities or low incomes. Odds of death for VA users were slightly lower using samples matched by propensity scores.

CONCLUSIONS

A small difference in risk-adjusted outcomes for VA users and nonusers undergoing revascularization in private sector hospitals was found. This difference reflects unmeasured severity in VA users undergoing revascularization in private sector hospitals.

Mortality of Department of Veterans Affairs patients admitted to private sector hospitals for 5 common medical conditions

Whether prior use of Veterans Affairs services is a marker for increased mortality was evaluated by using Medicare data for men aged 67 years and older admitted for acute myocardial infarction, chronic heart failure, chronic obstructive pulmonary disease, pneumonia, or stroke during 1996 to 2002. Patients using Veterans Affairs services during the 2 years preceding hospital admission were identified using Veterans Affairs encounter data, and 30-day mortality was compared in patients who did and did not use Veterans Affairs services, adjusting for patient risk factors. For most Veterans Affairs users, the odds of death were similar or slightly less than the odds of death for nonusers. For acute myocardial infarction, pneumonia, and stroke, the risk of death was slightly higher for Veterans Affairs users with low income. Results using propensity-matched samples were similar. The use of Veterans Affairs services is not a strong marker of unmeasured severity among patients in private sector hospitals.

The Postoperative Morbidity Survey was validated and used to describe morbidity after major surgery

OBJECTIVES

To describe the reliability and validity of the Postoperative Morbidity Survey (POMS). To describe the level and pattern of short-term postoperative morbidity after major elective surgery using the POMS.

STUDY DESIGN AND SETTING

This was a prospective cohort study of 439 adults undergoing major elective surgery in a UK teaching hospital. The POMS, an 18-item survey that address nine domains of postoperative morbidity, was recorded on postoperative days 3, 5, 8, and 15.

RESULTS

Inter-rater reliability was perfect for 11/18 items (Kappa=1.0), with Kappa=0.94 for 6/18 items. A priori hypotheses that the POMS would discriminate between patients with known measures of morbidity risk, and predict length of stay were generally supported through observation of data trends, and there was statistically significant evidence of construct validity for all but the wound and neurological domains. POMS-defined morbidity was present in 325 of 433 patients (75.1%) remaining in hospital on postoperative day 3 after surgery, 231 of 407 patients (56.8%) on day 5, 138 of 299 patients (46.2%) on day 8, and 70 of 111 patients (63.1%) on day 15. Gastrointestinal (47.4%), infectious (46.5%), pain-related (40.3%), pulmonary (39.4%), and renal problems (33.3%) were the most common forms of morbidity.

CONCLUSION

The POMS is a reliable and valid survey of short-term postoperative morbidity in major elective surgery. Many patients remain in hospital without any morbidity as recorded by the POMS.

A study design to assess the safety and efficacy of on-pump versus off-pump coronary bypass grafting: the ROOBY trial

Background Since the late 1960s, coronary artery bypass graft (CABG-only) procedures were traditionally performed using a heart-lung machine on an arrested heart (on-pump). Over the past decade, an increasing number CABG-only procedures were performed on a beating heart (off-pump). Advocates of the off-pump approach expect to reduce many of the adverse side effects related to using the heart-lung machine, while advocates for the on-pump procedure raise concerns related to graft patency rates and long-term event-free survival for the off-pump technique.

Purpose The U.S. Department of Veteran Affairs (VA) Cooperative Studies Program funded a randomized, multicenter clinical trial comparing the clinical and resourcerelated outcomes following on-pump versus off-pump techniques for veterans undergoing a non-emergent CABG-only procedure. The planning committee was faced with several critically important challenges to assure feasibility of study costs and required sample size; generalizability to non-VA surgical practices; and comparability of clinically meaningful results. These challenges are discussed.

Methods This study is a prospective, randomized, multicenter, single blinded (patient) clinical trial that compares on-pump and off-pump techniques for veterans requiring non-emergent CABG-only procedures. There will be 2200 patients randomized at 17 VA Medical Centers when the five-year recruitment period ends on 15 April 2007. There are two primary objectives: a short-term objective to assess the immediate impact of the two techniques on 30-day mortality/morbidity and a long-term objective to assess one-year mortality/morbidity. Major secondary outcomes are one-year graft patency rates and change in neuropsychological assessments from baseline to one year. All patients are assessed at 30 days post-surgery or discharge from the hospital, whichever is latest, and at one-year post-surgery.

Results During planning, several key issues had to be decided. These included 1) choosing primary objectives: a short-term (30-day) and a long-term (one-year) objective were chosen; 2) choosing primary outcome measures: composite measures were selected to ensure sufficient end-points; 3) standardization of surgical techniques: minimal standardization required but guidelines and continuing discussions on both techniques provided; 4) establishing criteria for surgeons and residents for participation: surgeons required to have completed 20 off-pump procedures prior to doing study procedures and residents, in presence of study surgeon, capable of doing either procedure; 5) identifying metrics of cognitive dysfunction sensitive to treatment: a neuropshychologist hired who centrally monitors cognitive functioning testing; and 6) blinding participants of surgical procedure: attempt to blind participants.

Limitations Areas of concern are whether all surgeons sufficiently experienced on the off-pump procedure, should residents have been allowed to do study surgeries, should techniques have been standardized more and were the best neurocognitive tests selected.

Conclusion The study design presented allows for a balanced and fair assessment of the on-pump and off-pump CABG procedures across a diversity of clinical outcomes and resource use metrics. Its results have the potential to influence clinical cardiac surgical practice in the future.

Has the publication of cardiac surgery outcome data been associated with changes in practice in northwest England: an analysis of 25,730 patients undergoing CABG surgery under 30 surgeons over eight years

OBJECTIVES

To study changes in coronary artery surgery practice in the years spanning publication of cardiac surgery mortality data in the UK.

METHODS

A retrospective analysis of prospectively collected data from all National Health Service centres undertaking adult cardiac surgery in northwest England was carried out. Patients undergoing coronary artery surgery for the first time between April 1997 and March 2005 were included. Changes in observed, predicted and risk adjusted mortality (EuroSCORE) were studied. Evidence of risk-averse behaviour was looked for by examining the number of patients at low risk (EuroSCORE 0-5), high risk (6-10), and very high risk (11 or more), before and after public disclosure.

RESULTS

25,730 patients underwent coronary artery surgery during the study period. The observed mortality decreased from 2.4% in 1997-8 to 1.8% in 2004-5 (p = 0.014). The expected mortality (EuroSCORE) increased from 3.0 to 3.5 (p<0.001). The observed to expected mortality ratio decreased from 0.8 to 0.51 (p<0.05). The total number and percentage of patients who were at low risk, high risk and very high risk was 2694 (84.6%), 449 (14.1%) and 41 (1.3%) before and 2654 (81.7%), 547 (16.8%) and 47 (1.4%) after public disclosure, respectively, demonstrating a significant increase in the number and proportion of high risk patients undergoing surgery (p<0.001).

CONCLUSIONS

Publication of cardiac surgery mortality data in the UK has been associated with decreased risk adjusted mortality on retrospective analysis of a large patient database. There is no evidence that fewer high risk patients are undergoing surgery because mortality rates are published.

Risk adjustment and outcome research. Part I

OBJECTIVE

The increasing demand for comparative evaluation of outcomes requires the development and diffusion of epidemiologic research, the ability to correctly formulate hypotheses, to conduct analyses and to interpret the results. The purpose of this paper is to provide a detailed but easy-reading review of epidemiologic methods to compare healthcare outcomes, particularly risk-adjustment methods.

METHODS

The paper is divided into three parts. Part I describes confounding in observational studies, the ways confounding is identified and controlled (propensity adjustment and risk adjustment), and the methods for constructing the severity measures in risk-adjustment procedures.

CONCLUSIONS

It is becoming increasingly important for policy makers and planners to identify which factors may improve or worsen the effectiveness of treatments and services and to compare the performances of providers. Politicians, managers, epidemiologists, and clinicians should make their decisions based on the validity and precision of study results, by using the best scientific knowledge available. The statistical methods described in this review cannot measure 'reality' as it 'truly' is, but can produce 'images' of it, defining limits and uncertainties in terms of validity and precision. Studies that use credible risk-adjustment strategies are more likely to yield reliable and applicable findings.

Effects of a statewide physician-led quality-improvement program on the quality of cardiac care

BACKGROUND

Several states have implemented mandatory public reporting of outcomes of cardiac revascularization procedures. Washington is the first to develop a nonmandatory, physician-led reporting program with public accountability and universal hospital participation. The purpose of this study was to determine whether quality improvement interventions resulted in the correction of data deficiencies and performance outliers for cardiac revascularization procedures.

METHODS

From 1999 through 2003, there were 18 hospitals with coronary bypass surgery and interventional cardiology programs and 12 with only the latter. All patients > or =18 years undergoing 24372 isolated coronary bypass surgeries and 59,656 percutaneous coronary interventions were included. After 1999 to 2001 data were analyzed in early 2002, the Clinical Outcomes Assessment Program implemented a 6-step quality-improvement intervention to measure and remeasure data quality, process compliance, and performance.

RESULTS

In 2003, 4 of the 18 surgery programs had 1 statistical outlier with respect to 4 performance measures, whereas 2 of 30 coronary intervention programs were mortality outliers. For bypass surgery, all programs maintained full compliance with program standards by adhering to timely and reliable submission of data, developing plans to address performance outliers, and demonstrating that outlier status did not persist from baseline to remeasurement. For coronary interventions, 1 program was a persistent outlier for mortality in 2002 and 2003.

CONCLUSIONS

The Clinical Outcomes Assessment Program has successfully monitored cardiac care patterns in Washington State over a 5-year period. Most hospitals that perform coronary revascularization procedures meet acceptable performance standards.

Evaluating Surgical Outcomes

The study of outcomes has become essential for guiding quality of-care assessment and for clinical research. In this article, the properties and process of patient outcomes measurement are described. The limitations of traditional outcomes are discussed and contrasted with the emerging concept of "patient-centered"outcomes, measured by validated instruments to assess the effects of surgical interventions on health-related quality of life, functional status, pain, and patient satisfaction. The strengths and weaknesses of several measurement tools used in the surgical literature are evaluated. Finally, the authors introduce "composite outcomes" as a reflection of the multidimensional nature of modern patient care.

The identification and development of Canadian coronary artery bypass graft surgery quality indicators

OBJECTIVE

The study objective was to develop quality indicators for coronary artery bypass graft surgery that relate to quality of care, associate with preventable death, and could be reported on performance reports.

METHODS

A comprehensive list of quality indicators was collected from quality improvement organizations including the Society For Thoracic Surgery, Northern New England Cardiovascular Disease Study Group, and Veteran's Affairs System. Indicators were collated from practice guidelines from the American College of Cardiology and the American Heart Association. A MEDLINE search using the keywords "quality indicators" and "coronary bypass" was completed. A 17-member multidisciplinary international expert panel was assembled, who voted using a 2-step Delphi process regarding association with quality of care, risk adjustment, association with preventable death, and inclusion on performance reports.

RESULTS

A total of 149 quality indicators were examined. This list was distilled to 33 indicators related to quality of care, 10 indicators that could be adequately risk adjusted, 34 indicators related to preventable death, and 18 indicators to be included on performance reports. These selected indicators consisted of 19 outcome variables, 23 process of care variables, and 4 structure variables. The quality indicators believed to be useful on a Canadian institutional coronary artery bypass graft surgery report card included the following: 30-day mortality, in-hospital mortality, electrocardiographic myocardial infarction, red cell transfusion, allogeneic blood product transfusion, deep sternal wound infection, postoperative stroke, postoperative dialysis, intensive care unit readmission, intensive care unit length of stay, ventilation time, repeat cardiac operation, repeat surgery with cardiopulmonary bypass, repeat revascularization, waiting time to surgery, completion of surgery within a recommended waiting time, use of left internal thoracic artery graft, and institutional volume.

CONCLUSIONS

This set of consensus quality indicators can be used as a standard list to be monitored by providers of coronary artery bypass graft surgery in an effort to continuously evaluate and improve their performance.

A new score for predicting neonatal very low birth weight mortality risk in the NEOCOSUR South American Network

OBJECTIVE

To develop and validate a model for very low birth weight (VLBW) neonatal mortality prediction, based on commonly available data at birth, in 16 neonatal intensive care units (NICUs) from five South American countries.

STUDY DESIGN

Prospectively collected biodemographic data from the Neonatal del Cono Sur (NEOCOSUR) Network between October 2000 and May 2003 in infants with birth weight 500 to 1500 g were employed. A testing sample and crossvalidation techniques were used to validate a statistical model for risk of in-hospital mortality. The new risk score was compared with two existing scores by using area under the receiver operating characteristic curve (AUC).

RESULTS

The new NEOCOSUR score was highly predictive for in-hospital mortality (AUC=0.85) and performed better than the Clinical Risk Index for Babies (CRIB) and the NICHD risk models when used in the NEOCOSUR Network. The new score is also well calibrated - it had good predictive capability for in-hospital mortality at all levels of risk (HL test=11.9, p=0.85). The new score also performed well when used to predict in hospital neurological and respiratory complications.

CONCLUSIONS

A new and relatively simple VLBW mortality risk score had a good prediction performance in a South American network population. This is an important tool for comparison purposes among NICUs. This score may prove to be a better model for application in developing countries.

Cardiac Surgery Risk Models: A Position Article

Differences in medical outcomes may result from disease severity, treatment effectiveness, or chance. Because most outcome studies are observational rather than randomized, risk adjustment is necessary to account for case mix. This has usually been accomplished through the use of standard logistic regression models, although Bayesian models, hierarchical linear models, and machine-learning techniques such as neural networks have also been used. Many factors are essential to insuring the accuracy and usefulness of such models, including selection of an appropriate clinical database, inclusion of critical core variables, precise definitions for predictor variables and endpoints, proper model development, validation, and audit. Risk models may be used to assess the impact of specific predictors on outcome, to aid in patient counseling and treatment selection, to profile provider quality, and to serve as the basis of continuous quality improvement activities.

Impact of renal insufficiency on short- and long-term outcomes after cardiac surgery

BACKGROUND

Renal insufficiency is highly prevalent in North America and has been established as a nontraditional risk factor for cardiovascular disease. Cardiovascular disease remains the primary cause of mortality in the general population and is often treated with coronary artery bypass surgery (CABG). This population-based study aimed to determine the risk of nondialysis dependent renal insufficiency (RI) on the long-term outcomes of patients who undergo CABG.

METHODS

Prospectively collected data from 26,506 patients were abstracted from the Cardiac Care Network database from 9 revascularization hospitals in Ontario, Canada. Multivariate regression analysis examined associations between preoperative RI and inhospital, 30-day, and 1-year mortality according to 3 levels of serum creatinine: <120 micromol/L (normal), 120 to 180 micromol/L (mild RI), and >180 micromil/L (moderate-severe RI) and 5 levels of creatinine clearance (Cockcroft-Gault): >100 mL/min (normal), 80 to 99 mL/min (mild impairment), 60 to 79 mL/min (mild-moderate impairment), 40 to 59 mL/min (moderate impairment), and <40 mL/min (severe impairment).

RESULTS

The overall inhospital, 30-day, and 1-year mortality rates were 1.90%, 2.0%, and 4.5%, respectively. Patients with RI had greater overall comorbidity. After adjustment for confounding factors, RI was associated with the greater risk of both 30-day (odds ratio 3.7, 95% CI 2.3-5.8, P <.0001) and 1-year mortality (odds ratio 4.6, 95% CI 3.3-6.4, P <.0001).

CONCLUSION

Preoperative renal impairment should be recognized as a significant risk factor for mortality after CABG. A trend of increasing risk with severity of renal impairment was demonstrated for both 30-day and 1-year mortality in this large, population-based study.

Breast cancer surgery trends and outcomes: results from a national department of veterans affairs study

BACKGROUND

This study examined trends and outcomes for breast cancer surgery performed at Department of Veterans Affairs (VA) hospitals.

STUDY DESIGN

We examined breast cancer operations performed in VA hospitals from October 1991 to September 1997. Data from the VA National Surgical Quality Improvement Program, surgical pathology reports, discharge data, and outpatient data were used. Surgical outcomes included postoperative length of stay, 30-day morbidity rates, 1-year surgery-related readmission rates, and mortality. An expert panel of breast cancer clinicians identified surgery-related hospital readmissions. Hierarchical regression analysis was used to identify patient, provider, and hospital characteristics associated with postoperative length of stay, and 30-day morbidity.

RESULTS

From October 1991 to September 1997 1,333 breast operations were performed, ranging from 1 to 38 on average per hospital; 478 operations were for breast cancer. Among breast cancer surgery patients, 25% were men. Thirty-day morbidity rates, 1-year hospital readmission rates, and mortality were very low for both men and women. Postoperative length of stay averaged 6.8 days. Lower income, longer operation times, and older age increased the likelihood of 30-day morbidity. Lower functional status, older age, longer operation time, and lower average annual volume of procedures increased postoperative length of stay. Documentation of the extent of disease and surgical margin in pathology reports was poor in medical records.

CONCLUSIONS

Hospital stays were longer, and morbidity and readmission rates for patients having breast cancer operations at VA hospitals were comparable to those reported for private sector hospitals.

Comparison of short-term mortality risk factors for valve replacement versus coronary artery bypass graft surgery

BACKGROUND

Risk factors for 30-day operative (short-term) mortality following coronary artery bypass graft (CABG only) procedures are well established. However, little is known about how the risk factors for short-term mortality following valve replacement procedures (with or without a CABG procedure performed) compare with CABG only risk factors.

METHODS

Department of Veterans Affairs (VA) records (65,585 records) were collected from October 1991 through March 2001 and analyzed. Risk factors for short-term mortality were compared across three subgroups of patients: CABG only surgery (n = 56,318), aortic valve replacement (AVR) with or without CABG (n = 7450), and mitral valve replacement (MVR) with or without CABG (n = 1817). Multivariable logistic regression analyses were used to compare the relative magnitude of risk for 19 candidate predictor variables across subgroups.

RESULTS

Only three patient baseline characteristics differed significantly in magnitude of risk between the procedure groups. Partially or totally dependent functional status significantly increased the risk of short-term mortality for AVR patients (odds ratio [OR] 1.64, 95% confidence interval [CI] 1.29-2.09) and MVR patients (OR 2.21, 95% CI 1.48-3.30), but not for CABG only patients (OR 1.04, 95% CI 0.93-1.16). Conversely, previous heart surgery and New York Heart Association functional class III or IV symptoms conferred greater magnitude of risk for CABG only patients compared with the valve subgroups.

CONCLUSIONS

Overall, the risk factors for short-term mortality following valve replacement and CABG surgery appear to be relatively consistent. However, clinicians should be aware of the importance of preoperative functional status as a unique predictor of mortality following valve surgery.

Relationship Between Processes of Care and Coronary Bypass Operative Mortality and Morbidity

BACKGROUND

Information is limited regarding the effects of processes of care on cardiac surgical outcomes. Correspondingly, many recommended cardiac surgical processes of care are derived from animal experiments or clinical judgment. This report from the VA Cooperative Study in Health Services, "Processes, Structures, and Outcomes of Cardiac Surgery," focuses on the relationships between 3 process groups (preoperative evaluation, intraoperative care, and supervision by senior physicians) and a composite outcome, perioperative mortality and morbidity.

METHODS

Data on 734 risk, process, and structure variables were collected prospectively on 3,988 patients who underwent coronary artery bypass grafting at 14 VA medical centers between 1992 and 1996. Data reduction was accomplished by examining data completeness and variation across sites and surgeon, using previously published data and clinical judgment. We then applied multivariable logistic regression to the 39 remaining processes of care to determine which were related to the composite outcome after adjusting for 17 patient-related risk factors and controlling for intraoperative complications.

RESULTS

Our first analysis showed several measures of operative duration, the use of inotropic agents, transesophageal echo, lowest systemic temperature, and hemoconcentration/ultrafiltration, to be powerful predictors of the composite outcome. Because the use of inotropic agents and operative duration may be related to an intermediate outcome (eg, intraoperative complications), we performed a second analysis omitting these processes. The use of intraoperative transesophageal echo and hemoconcentration/ultrafiltration remained significantly associated with an increased risk of an event (odds ratios 1.60 and 1.36, respectively).

CONCLUSIONS

Our results viewed in the context of past studies suggest the possibility that inotropic use, TEE, and hemoconcentration/ultrafiltration may have adverse effects on operative outcome. Further evaluation of these processes of care using observational data, as well as randomized trials when feasible, would be of interest.

Preoperative risks and outcomes of hip and knee arthroplasty in the Veterans Health Administration

The relationship between patient characteristics and outcomes of total joint arthroplasty (TJA) was examined in a population of veterans treated in VA hospitals. Outcomes included 30-day mortality and morbidity, postoperative length of stay, and readmission caused by surgical complications. A larger proportion of women then men were functionally impaired before surgery in both the hip (22% vs. 14%) and knee samples (14% vs. 7%; all P<.01). Rates of adverse outcomes in this population were very low. Preoperative comorbid conditions, abnormal laboratory values, and being nonwhite were related to poor outcomes of TJA. Gender was a significant independent predictor of morbidity and length of stay for total knee arthroplasty.

Differences in length of stay in Veterans Health Administration and other United States hospitals: is the gap closing?

OBJECTIVES

Compare risk-adjusted length of stay (LOS) in VA and other United States (non-VA) hospitals and determine if relative differences in LOS have changed in recent years.

RESEARCH DESIGN

Retrospective cohort study.

PATIENTS

Patients with ten common medical diagnoses admitted to all VA hospitals and to non-VA hospitals included in the National Hospital Discharge Survey (NHDS) during 1996 through 1999.

DATA

Comparable data elements were obtained from VA administrative databases and the NHDS. LOS was adjusted for age, gender, marital status, and comorbidity. Comorbidity was assessed using a validated methodology that considers 30 conditions.

RESULTS

Unadjusted mean LOS was longer in VA than non-VA patient for all 4 years, in aggregate (7.1 vs. 4.9 days, respectively; P < 0.001), and for each year individually. However, the difference in mean LOS in VA and non-VA patients declined from 2.9 days in 1996 to 1.6 days in 1999. LOS in VA patients remained longer (P < 0.001) in linear regression analyses, adjusting for demographics and comorbidity. However, the difference in LOS declined from 28.5% (95% CI, 28.1%-29.0%) in 1996 to 17.0% (95% CI, 16.6%-17.4%) in 1999. These results were similar in analyses of individual geographic regions.

CONCLUSIONS

Risk-adjusted LOS was longer in VA hospitals than in other United States hospitals. However, differences in LOS narrowed between 1996 and 1999. These findings suggest that changes in the organization and delivery of VA health care in the mid-1990s may be closing the gap between the VA and other healthcare systems in hospital utilization.

Mortality and length of stay in a veterans affairs hospital and private sector hospitals serving a common market

OBJECTIVE

To compare severity-adjusted in-hospital mortality and length of stay (LOS) in a Veterans Administration (VA) hospital and private sector hospitals serving the same health care market.

DESIGN

Retrospective cohort study.

SETTING

A large VA hospital and 27 private sector hospitals in the same metropolitan area.

PATIENTS

Consecutive VA (N = 1,960) and private sector (N = 157,147) admissions in 1994 to 1995 with 9 high-volume diagnoses.

MEASUREMENTS

Severity of illness was measured using validated multivariable models that were based on data abstracted from medical records. Outcomes were adjusted for severity and compared in VA and private sector patients using multiple logistic or linear regression analysis.

MAIN RESULTS

Unadjusted mortality was similar in VA and private sector patients (5.0% vs 5.6%, respectively; P =.26), although mean LOS was longer in VA patients (12.7 vs 7.0 days; P <.001). Adjusting for severity, the odds of death in VA patients was similar (odds ratio [OR] 1.07; 95% confidence interval [95% CI], 0.74 to 1.54; P =.73). However, a larger proportion of deaths in VA patients occurred later during hospitalization (P <.001), and the odds of death in VA patients were actually lower (P <.05) in analyses limited to deaths during the first 7 (OR, 0.56) or 14 (OR, 0.63) days. Adjusted LOS was longer (P <.001) in VA patients for all 9 diagnoses.

CONCLUSIONS

If the current findings generalizable to other markets, hospital mortality, a widely used performance measure, may be similar or lower in VA and private sector hospitals serving the same markets. The longer LOS of VA patients may reflect differences in practice patterns and may be an important source of bias in comparisons of VA and private sector hospitals.

Evaluating outcomes of hospital care following coronary artery bypass surgery in Rome, Italy

OBJECTIVE

Monitoring health outcomes across hospitals has become a growing interest as a potential means to promote quality of care, but in Italy it is at the beginning stage. We aimed at comparing the performance of different cardiac surgery units and testing the utility of routinely collected data in this respect.

METHODS

From the Lazio region hospital information system (HIS), we selected a cohort of 1603 individuals (84% males; mean age 63 years, SD 8) residing in Rome (2,685,890 inhabitants), who underwent isolated coronary artery bypass surgery (CABG, ICD-9-CM code: 36.1) in the period 1996-97 in seven major cardiac surgery units in the city. They were identified as A, B, C (teaching), D and E (non-teaching) units. Information on vital status at 30 days after the CABG surgery was obtained through an automatic record linkage with the Municipal Registry of Rome. The association between cardiac surgery units and outcome was evaluated through logistic regression taking into account the following a priori risk factors in different models: gender, age, socio-economic status, type of ischaemic heart disease and comorbidities.

RESULTS

The overall mortality was 5.4% (range 2.1-11.4%). Statistically significant predictors of outcome included: age (OR=7.5 for age> or =70 vs. 35-49 years), acute myocardial infarction (OR=32.7 vs. acute-subacute forms/angina), chronic myocardial ischaemia (OR=4.2 vs. acute-subacute forms/angina), other heart diseases (OR=4.8), chronic renal disease (OR=16.0) and peripheral arterial disease (OR=2.9). Statistically significant variability in mortality was observed across hospitals; taking hospital A as reference, hospital D showed the highest risk (OR=5.7, 95% CI=1.9-17.3, in the fully adjusted model).

CONCLUSIONS

We suggest that a true variation in quality of care play a role in the observed differences across hospitals, although chance and inaccurately measured risk factors cannot be excluded. Despite some limitations, the HIS is a valid tool for screening cardiac surgery units with poor performance.