Kwan Hur

Risk adjustment of the postoperative morbidity rate for the comparative assessment of the quality of surgical care : Results of the National Veterans Affairs Surgical Risk Study

BACKGROUND The National Veterans Affairs Surgical Risk Study was designed to collect reliable, valid data on patient risk and outcomes for major surgery in the Veterans Health Administration and to report comparative risk-adjusted postoperative mortality rates for surgical services in Veterans Health Administration. STUDY DESIGN This cohort study was conducted in 44 Veterans Affairs Medical Centers. Included were 87,078 major noncardiac operations performed under general, spinal, or epidural anesthesia between October 1, 1991, and December 31, 1993. The main outcomes measure was all-cause mortality within 30 days after the index procedure. Multivariable logistic regression risk-adjustment models for all operations and for eight surgical subspecialties were developed. Risk-adjusted surgical mortality rates were expressed as observed-to-expected ratios and were compared with unadjusted 30-day postoperative mortality rates. RESULTS Patient risk factors predictive of postoperative mortality included serum albumin level, American Society of Anesthesia class, emergency operation, and 31 additional preoperative variables. Considerable variability in unadjusted mortality rates for all operations was observed across the 44 hospitals (1.2-5.4%). After risk adjustment, observed-to-expected ratios ranged from 0.49 to 1.53. Rank order correlation of the hospitals by unadjusted and risk-adjusted mortality rates for all operations was 0.64. Ninety-three percent of the hospitals changed rank after risk adjustment, 50% by more than 5 and 25% by more than 10. CONCLUSIONS The Department of Veterans Affairs has successfully implemented a system for the prospective collection and comparative reporting of risk-adjusted postoperative mortality rates after major noncardiac operations. Risk adjustment had an appreciable impact on the rank ordering of the hospitals and provided a means for monitoring and potentially improving the quality of surgical care.

Relation of Surgical Volume to Outcome in Eight Common Operations : Results From the VA National Surgical Quality Improvement Program

OBJECTIVE To examine, in the Veterans Health Administration (VHA), the relation between surgical volume and outcome in eight commonly performed operations of intermediate complexity. SUMMARY BACKGROUND DATA In multihospital health care systems such as VHA, consideration is often given to closing low-volume surgical services, with the assumption that better surgical outcomes are achieved in hospitals with larger surgical volumes. Literature data to support this assumption in intermediate-complexity operations are either limited or controversial. METHODS The VHA National Surgical Quality Improvement Program data on nonruptured abdominal aortic aneurysmectomy, vascular infrainguinal reconstruction, carotid endarterectomy (CEA), lung lobectomy/pneumonectomy, open and laparoscopic cholecystectomy, partial colectomy, and total hip arthroplasty were used. Pearson correlation, analysis of variance, mixed effects hierarchical logistic regression, and automatic interaction detection analysis were used to assess the association of annual procedure/specialty volume with risk-adjusted 30-day death (and stroke in CEA). RESULTS Eight major surgical procedures (68,631 operations) were analyzed. No statistically significant associations between procedure or specialty volume and 30-day mortality rate (or 30-day stroke rate in CEA) were found. CONCLUSIONS In VHA hospitals, the procedure and surgical specialty volume in eight prevalent operations of intermediate complexity are not associated with risk-adjusted 30-day mortality rate from these operations, or with the risk-adjusted 30-day stroke rate from CEA. Volume of surgery in these operations should not be used as a surrogate for quality of surgical care.

Prognostic models of thirty-day mortality and morbidity after major pulmonary resection ☆ ☆☆ ★

BACKGROUND A part of the prospective, multi-institutional National Veterans Affairs Surgical Quality Improvement Program was developed to predict 30-day mortality and morbidity for patients undergoing a major pulmonary resection. METHODS Perioperative data were acquired from 194,319 noncardiac surgical operations at 123 Veterans Affairs Medical Centers between October 1, 1991, and August 31, 1995. Current Procedural Terminology code-based analysis was undertaken for major pulmonary resections (lobectomy and pneumonectomy). Preoperative, intraoperative, and outcome variables were collected. The 30-day mortality and morbidity models were developed by means of multivariable stepwise logistic regression with the preoperative and intraoperative variables used as independent predictors of outcome. RESULTS A total of 3516 patients (mean age 64 9 years) underwent either lobectomy (n = 2949) or pneumonectomy (n = 567). Thirty-day mortality was 4.0% for lobectomy (119/2949) and 11.5% for pneumonectomy (65/567). The preoperative predictors of 30-day mortality were albumin, do not resuscitate status, transfusion of more than 4 units, age, disseminated cancer, impaired sensorium, prothrombin time more than 12 seconds, type of operation, and dyspnea. When the intraoperative variables were considered, intraoperative blood loss was added to the preoperative model. In the presence of these intraoperative variables in the model, do not resuscitate status and prothrombin time more than 12 seconds were only marginally significant. Thirty-day morbidity, defined as the presence of 1 or more of the 21 predefined complications, was 23.8% for lobectomy (703/2949) and 25.7% for pneumonectomy (146/567). In multivariable models, independent preoperative predictors (P <.05) of 30-day morbidity were age, weight loss greater than 10% in the 6 months before surgery, history of chronic obstructive pulmonary disease, transfusion of more than 4 units, albumin, hemiplegia, smoking, and dyspnea. When intraoperative variables were added to the preoperative model, the duration of operation time and intraoperative transfusions were included in the model and albumin became marginally significant. CONCLUSIONS This analysis identifies independent patient risk factors that are associated with 30-day mortality and morbidity for patients undergoing a major pulmonary resection. This series provides an initial risk-adjustment model for major pulmonary resections. Future refinements will allow comparative assessment of surgical outcomes and quality of care at many institutions.

Validating risk-adjusted surgical outcomes: site visit assessment of process and structure1

Abstract Background: Risk-adjusted mortality and morbidity rates are often used as measures of the quality of surgical care. This study was conducted to determine the validity of risk-adjusted surgical morbidity and mortality rates as measures of quality of care by assessing the process and structure of care in surgical services with higher-than-expected and lower-than-expected risk-adjusted 30-day mortality and morbidity rates. Study Design: A structural survey of 44 Veterans Affairs Medical Center surgical services and site visits to 20 surgical services with higher-than-expected and lower-than-expected risk-adjusted outcomes were conducted. Main outcome measures included assessment of technology and equipment, technical competence of staff, leadership, relationship with other services, monitoring of quality of care, coordination of work, relationship with affiliated institutions, and overall quality of care. Results: Surgical services with lower-than-expected risk-adjusted surgical morbidity and mortality rates had significantly more equipment available in surgical intensive care units than did services with higher-than-expected outcomes (4.3 versus 2.9, p Conclusions: Significant differences in several dimensions of process and structure of the delivery of surgical care are associated with differences in risk-adjusted surgical morbidity and mortality rates among 44 Veterans Affairs Medical Centers.

Association of perioperative β-blockade with mortality and cardiovascular morbidity following major noncardiac surgery.

IMPORTANCE The effectiveness of perioperative β-blockade in patients undergoing noncardiac surgery remains controversial. OBJECTIVE To determine the associations of early perioperative exposure to β-blockers with 30-day postoperative outcome in patients undergoing noncardiac surgery. DESIGN, SETTING, AND PATIENTS A retrospective cohort analysis evaluating exposure to β-blockers on the day of or following major noncardiac surgery among a population-based sample of 136,745 patients who were 1:1 matched on propensity scores (37,805 matched pairs) treated at 104 VA medical centers from January 2005 through August 2010. MAIN OUTCOMES AND MEASURES All cause 30-day mortality and cardiac morbidity (cardiac arrest or Q-wave myocardial infarction). RESULTS Overall 55,138 patients (40.3%) were exposed to β-blockers. Exposure was higher in the 66.7% of 13,863 patients undergoing vascular surgery (95% CI, 65.9%-67.5%) than in the 37.4% of 122,882 patients undergoing nonvascular surgery (95% CI, 37.1%-37.6%; P < .001). Exposure increased as Revised Cardiac Risk Index factors increased, with 25.3% (95% CI, 24.9%-25.6%) of those with no risk vs 71.3% (95% CI, 69.5%-73.2%) of those with 4 risk factors or more exposed to β-blockers (P < .001). Death occurred among 1.1% (95% CI, 1.1%-1.2%) and cardiac morbidity occurred among 0.9% (95% CI, 0.8%-0.9%) of patients. In the propensity matched cohort, exposure was associated with lower mortality (relative risk [RR], 0.73; 95% CI, 0.65-0.83; P < .001; number need to treat [NNT], 241; 95% CI, 173-397). When stratified by cumulative numbers of Revised Cardiac Risk Index factors, β-blocker exposure was associated with significantly lower mortality among patients with 2 factors (RR, 0.63 [95% CI, 0.50-0.80]; P < .001; NNT, 105 [95% CI, 69-212]), 3 factors (RR, 0.54 [95% CI, 0.39-0.73]; P < .001; NNT, 41 [95% CI, 28-80]), or 4 factors or more (RR, 0.40 [95% CI, 0.25-0.73]; P < .001; NNT, 18 [95% CI, 12-34]). This association was limited to patients undergoing nonvascular surgery. β-Blocker exposure was also associated with a lower rate of nonfatal Q-wave infarction or cardiac arrest (RR, 0.67 [95% CI, 0.57-0.79]; P < .001; NNT, 339 [95% CI, 240-582]), again limited to patients undergoing nonvascular surgery. CONCLUSIONS AND RELEVANCE Among propensity-matched patients undergoing noncardiac, nonvascular surgery, perioperative β-blocker exposure was associated with lower rates of 30-day all-cause mortality in patients with 2 or more Revised Cardiac Risk Index factors. Our findings support use of a cumulative number of Revised Cardiac Risk Index predictors in decision making regarding institution and continuation of perioperative β-blockade. A multicenter randomized trial involving patients at a low to intermediate risk by these factors would be of interest to validate these observational findings.

Modeling Clustered Count Data with Excess Zeros in Health Care Outcomes Research

In health research, count outcomes are fairly common and often these counts have a large number of zeros. In order to adjust for these extra zero counts, various modifications of the Poisson regression model have been proposed. Lambert (Lambert, D., Technometrics 34, 1–14, 1992) described a zero-inflated Poisson (ZIP) model that is based on a mixture of a binary distribution (πi) degenerated at zero with a Poisson distribution (λi). Depending on the relationship between πi and λi, she described two variants: a ZIP and a ZIP (τ) model. In this paper, we extend these models for the case of clustered data (e.g., patients observed within hospitals) and describe random-effects ZIP and ZIP (τ) models. These models are appropriate for the analysis of clustered extra-zero Poisson count data. The distribution of the random effects is assumed to be normal and a maximum marginal likelihood estimation method is used to estimate the model parameters. We applied these models to data from patients who underwent colon operations from 123 Veterans Affairs Medical Centers in the National VA Surgical Quality Improvement Program.

Outcome after pancreaticoduodenectomy for periampullary cancer: An analysis from the veterans affairs national surgical quality improvement program

The aim of this study is to define the risk factors that predict adverse outcomes for patients undergoing pancreaticoduodenectomy for periampullary cancer in the Department of Veterans Affairs Healthcare System (VA). The VA National Surgical Quality Improvement Program prospectively collected data on 462 patients undergoing pancreaticoduodenectomy in 123 VA medical centers from 1990 to 2000. Independent variables included 68 preoperative and 12 intraoperative variables. The main outcome measures were 30-day postoperative mortality and morbidity, as measured by a set of 20 pre-defined complications. Predictive models for 30-day morbidity and mortality were constructed using logistic regression analysis. The 30-day morbidity rate was 45.9% (212/462). The 30-day postoperative mortality rate was 9.3% (43/462). Significant predictors of mortality included: preoperative serum albumin, American Society of Anesthesiologists classification, preoperative bilirubin >20mg/dl, and operative time. The use of preoperative biliary tract instrumentation did not predict postoperative death or septic complications. This study provides a set of preoperative risk factors that are predictive of adverse outcome following pancreaticoduodenectomy. These factors may assist in patient selection for this procedure and are likely to facilitate risk-adjusted comparison of pancreaticoduodenectomy outcomes between different health care systems.

Association of Perioperative Statin Use With Mortality and Morbidity After Major Noncardiac Surgery.

Importance The efficacy of statins in reducing perioperative cardiovascular and other organ system complications in patients undergoing noncardiac surgery remains controversial. Owing to a paucity of randomized clinical trials, analyses of large databases may facilitate informed hypothesis generation and more efficient trial design. Objective To evaluate associations of early perioperative statin use with outcomes in a national cohort of veterans undergoing noncardiac surgery. Design, Setting, and Participants This retrospective, observational cohort analysis included 180 478 veterans undergoing elective or emergent noncardiac surgery (including vascular, general, neurosurgery, orthopedic, thoracic, urologic, and otolaryngologic) who were admitted within 7 days of surgery and sampled by the Veterans Affairs Surgical Quality Improvement Program (VASQIP). Patients were admitted to Department of Veterans Affairs hospitals and underwent 30-day postoperative follow-up. Data were collected from October 1, 2005, to September 30, 2010, and analyzed from November 28, 2013, to October 31, 2016. Exposure Statin use on the day of or the day after surgery. Main Outcomes and Measures All-cause 30-day mortality (primary outcome) and standardized 30-day cardiovascular and noncardiovascular outcomes captured by VASQIP. Use of statins and other perioperative cardiovascular medications was ascertained from the Veterans Affairs Pharmacy Benefits Management research database. Results A total of 180 478 eligible patients (95.6% men and 4.4% women; mean [SD] age, 63.8 [11.6] years) underwent analysis, and 96 486 were included in the propensity score–matched cohort (96.3% men; 3.7% women; mean [SD] age, 65.9 [10.6] years). At the time of hospital admission, 37.8% of patients had an active outpatient prescription for a statin, of whom 80.8% were prescribed simvastatin and 59.5% used moderate-intensity dosing. Exposure to a statin on the day of or the day after surgery based on an inpatient prescription was noted in 31.5% of the cohort. Among 48 243 propensity score–matched pairs of early perioperative statin-exposed and nonexposed patients, 30-day all-cause mortality was significantly reduced in exposed patients (relative risk, 0.82; 95% CI, 0.75-0.89; P < .001; number needed to treat, 244; 95% CI, 170-432). Of the secondary outcomes, a significant association with reduced risk of any complication was noted (relative risk, 0.82; 95% CI, 0.79-0.86; P < .001; number needed to treat, 67; 95% CI, 55-87); all were significant except for the central nervous system and thrombosis categories, with the greatest risk reduction (relative risk, 0.73; 95% CI, 0.64-0.83) for cardiac complications. Conclusions and Relevance Early perioperative exposure to a statin was associated with a significant reduction in all-cause perioperative mortality and several cardiovascular and noncardiovascular complications. However, the potential for selection biases in these results must be considered.

Risk Adjustment of the Postoperative Morbidity Rate for the Comparative Assessment of the Quality of Surgical Care: Results of the National Veterans Affairs Surgical Risk Study

BACKGROUND The National Veterans Affairs Surgical Risk Study was designed to collect reliable, valid data on patient risk and outcomes for major surgery in the Veterans Health Administration and to report comparative risk-adjusted postoperative mortality and morbidity rates for surgical services in the Veterans Health Administration. STUDY DESIGN This was a cohort study conducted at 44 Veterans Affairs Medical Centers closely affiliated with university medical centers. Included were 87,078 major noncardiac operations performed under general, spinal, or epidural anesthesia between October 1, 1991, and December 31, 1993. The main outcomes measures in this report are 21 postoperative adverse events (morbidities) occurring within 30 days after the index procedure. Multivariable logistic regression risk-adjustment models for all operations and for eight surgical subspecialties were developed. RESULTS Patient risk factors predictive of postoperative morbidity included serum albumin level, American Society of Anesthesia class, the complexity of the operation, and 17 other preoperative risk variables. Wide variation in the unadjusted rates of one or more postoperative morbidities for all operations was observed across the 44 hospitals (7.4-28.4%). Risk-adjusted observed-to-expected ratios ranged from 0.49 to 1.46. The Spearman rank order correlation between the ranking of the hospitals based on unadjusted morbidity rates and risk-adjusted observed-to-expected ratios for all operations was 0.87. There was little or no correlation between the rank order of the hospitals by risk-adjusted morbidity and risk-adjusted mortality. CONCLUSIONS The Department of Veterans Affairs has successfully implemented a system for the prospective collection and comparative reporting of postoperative mortality and morbidity rates after major noncardiac operations. Risk adjustment had only a modest effect on the rank order of the hospitals.