Special Topics

2007 Journal of the American College of Cardiology  
GWTG is an initiative to improve guidelines adherence in hospitalized patients, in 3 modules (CAD, stroke, HF). The use of composite measures to compare all 3 GWTG modules has not been done. Methods: GWTG uses a collaborative model with web-based data collection, decision support, and on-demand reporting. Data were collected by hospitals at baseline and subsequent quarters (Q) for pre-specified performance measures (PM). These data were used to construct for each module a defect free measure (%
more » ... patients who receive all eligible PM) and a composite measure (CM) (% successful execution of each opportunity across all PM for that module). Time trend is assessed with repeat measure analysis to evaluate improvement in sites with at least 30 pts per Q from at least 30 sites for each Q. This resulted in 15 Q data for CAD, 12 Q data for stroke and 3 Q data for HF. Results: Changes in defect free care were similar in all modules from baseline to Q3: CAD 8.7% Stroke 6.6%, and HF 9%; CM: CAD 4.9%, Stroke 4.4% and HF 5.7%. In the longer running CAD and stroke programs continued improvement was seen with sites achieving defect free scores of 65.3% for CAD at Q15 and 73.9% at Q12 for Stroke and respective composite scores of 85.4% and 88.7%. Conclusions: GWTG hospitals produce comparable improvement in defect free and CM performance over time in CAD, Stroke or HF patients. Improvement is sustained and continues for up to 15 quarters (CAD). Despite this trend of improving care, more work is needed to implement high reliability strategies. Background: GWTG-HF is an initiative to characterize and improve guidelines use in hospitalized heart failure (HF) pts. While therapies for HF are effective and recommended across etiologies, it is unknown whether quality of care as indexed by specific performance measures (PM) differs by HF etiology. Methods: GWTG-HF uses a collaborative model, and a web-based tool for data collection, decision support, and on-demand reporting. Data were collected by 146 hospitals on 17,416 pts for 5 pre-specified PM: discharge instructions, LV function measurement, ACEI/ARB, beta blocker (BB), smoking cessation counseling and a defect free measure (DFM) of % pts who receive all of the 5 PM for which they were eligible. PM were analyzed based on ischemic (I) or non-ischemic (NI) etiology and adjusted for pt characteristics and comorbidities. Results: Pts with I-HF were older (74.5 + 11.9 vs 70.5 + 16.4 years), with more renal dysfunction (22.2% vs. 16.2%, p<0.0001), lower % women (44.1% vs. 57.6%, p<0.0001), and more whites (77.4% vs. 64.3%, p<0.0001) compared to the NI-HF group. Adjusted data demonstrate that BB use was more likely to occur in I (OR 1.20); all other PM and the DFM were similar between the two groups. Conclusions: I compared to NI etiology of HF is not an independent predictor of quality of care for HF, except for the use of BBs. Higher BB use in the I group may be related to indications for both HF and ischemic heart disease. Further education and process changes should be designed to improve care irrespective of HF etiology. Adjusted Analysis for PM in Ischemic vs. Non-ischemic HF Ischemic 10,312 Non-Ischemic 7,102 Odds Ration Ischemic vs. Non-ischemic Measures Percent Adherence Percent Adherence Unadjusted OR (95% CI) Adjusted OR* (95% CI) Discharge Instructions 73.8% 74.1% 1.05 (0.99-1.12) 1.02 (0.97-1.08) LV Function Assessment^91.6% 91.1% 1.02 (0.88-1.17) 1.04 (0.90-1.20) ACEI/ARB Use 80.4% 84.0% 0.77 (0.67-0.89) 0.91 (0.78-1.06) Smoking Cessation Counseling 78.2% 82.3% 0.87 (0.75-1.01) 0.94 (0.79-1.11) Beta Blocker Use 88.5% 86.6% 1.08 (0.94-1.24) 1.20 (1.02-1.40) Composite 63.9% 64.8% 0.94 (0.88-1.00) 1.02 (0.95-1.10) Background: Previous studies have reported widely varying smoking abstinence rates in populations receiving behavioral interventions for smoking cessation. Methods: We undertook a meta-analysis of randomized clinical trials (RCTs) evaluating each of 4 behavioral interventions to gain more accurate estimates of their efficacy. The behavioral interventions studied were minimal clinical intervention (brief advice) as well as more resource-intensive interventions including individual, group, and telephone counseling. We searched for all RCTs reporting biochemically-validated smoking abstinence rates at 6 and/or 12 months. Smoking abstinence was examined in a randomeffects meta-analysis using the most conservative measure of abstinence reported. Results: We identified 51 RCTs randomizing 28,927 patients (minimal clinical intervention: 9 RCTs, n=6,456; individual counseling: 23 RCTs, n=10,646; group counseling: 12 RCTs, n=3,600; telephone counseling: 10 RCTs, n=8,225). All 4 behavioral interventions were efficacious at increasing smoking abstinence rates (figure). Of note, all point estimates of ORs were similar for each of the 4 interventions. However, the confidence interval for minimal clinical intervention included unity. Conclusion: The use of behavioral interventions results in a modest increase in smoking abstinence rates. Minimal clinical intervention may be as efficacious as more resourceintensive behavioral interventions. Background: While extended-release (ER) niacin is a highly effective agent for raising HDL-C and is proven to reduce CVD risk, it may not be optimally utilized due to niacin induced flushing (NIF), leading to poor patient compliance. We examined NIF and its relationship with ER niacin discontinuation and nonadherence in actual clinical practice. Methods: We used a pilot tested questionnaire to conduct a telephone interview of 500 patients identified from administrative claims data to have newly initiated ER niacin (mean time between initiation and interview: 9.26 mos [SD=3.29]). Results: About 27% of the sample reported having discontinued taking niacin at the time of interview ("discontinuers", n=136) with a mean duration of use of 3.2 mos (SD=3.1). About 91% of the discontinuers reported experiencing flushing symptoms and 54% reported 'severe' or 'extreme' flushing (on a scale of none, mild, moderate, severe, extreme). Taking into account the possible benefits of ER niacin, 87% of the discontinuers who had flushing (n=124) reported that the greatest degree of flushing they could tolerate was 'mild' to 'moderate'. Overall, 66% of the discontinuers reported that it would be easier to take ER niacin as recommended if the flushing side effects were reduced. Even among ER niacin "continuers" (n=364) at the time of interview, 82% reported experiencing flushing symptoms and 21% reported 'severe' to 'extreme' flushing. Among the continuers who had flushing (n=299) , 21% reported skipping/ delaying doses and 19% reported stopping niacin for > 7 days and then restarting. Overall, 21% of the continuers reported that ER niacin side effects have made them want to stop taking it. Multivariate analyses indicated that flushing symptom severity was a strong predictor of discontinuation ['severe' OR: 3.2, 95% CI 1.4-7.2; 'extreme' OR: 11.3, 95% CI 4.2-30.4] and skipping/delaying of doses ['moderate' OR: 2.4, 95% CI 1.0-5.9; 'severe' OR: 9.4, 95% CI 3.6-24.7; 'extreme'OR: 8.0, 95% CI 1.9-33.5] of niacin. Conclusion: Since long-term continuous treatment is generally necessary in persons with dyslipidemia, flushing side effects with ER niacin appear to limit the acceptability of this otherwise highly effective therapy. 9:00 a.m. Background: Clinical outcomes of compensated patients with heart failure (HF) due to impaired or preserved ejection fraction (EF) undergoing elective major non-cardiac surgery are not well described. Methods: We reviewed data for consecutive patients with HF (mean age 69 ± 12 years, 54 % men) who underwent a systematic preoperative optimization by hospitalists from 2003 to 2006. Patients were stratified into impaired (EF ≤ 40%) or preserved ejection fraction (EF>40%). Propensity matching with multivariate logistical analysis was used to compare outcomes between patients in the two strata. Results: 562 HF patients (192 EF≤40% and 370 EF>40%) and 10,693 controls without HF were followed for a median of 1.9 years post-operatively. In propensity analysis of matched cohorts (Table 1) , the presence of HF was not associated with significantly increased 30-day post-operative mortality risk. 30-day readmission rate and length of stay (LOS) were higher in HF, particularly in preserved ejection fraction (p=0.04). Conclusions: In patients with compensated HF undergoing elective major non-cardiac surgery, operative and 30-day mortality were comparable to that of propensity-matched controls. However, patients with HF (particularly in those with preserved rather than impaired EF) were more likely to have longer length of stay and be readmitted at 30-days post-operatively. P≤ 0.05 vs. appropriate propensity-matched control group, †control matched for impaired EF; ‡control matched for preserved EF. 9:00 a.m. Background: The ACC/AHA 2006 guideline target for ST-elevation myocardial infarction (STEMI) is door-to-balloon time ≤ 90 minutes (min). We hypothesize that quality improvement (QI) methods can decrease door-to-balloon (D2B) time and improve care delivery. Methods: We studied all STEMI patients presenting to Vanderbilt University Medical Center from July 2005-June 2006. We created a process flow chart and analyzed all D2B process steps including all subinterval time periods. In February 2006, we implemented targeted changes in the process including emergency department activation of the catheterization lab, insistence of 30 min arrival time for the catheterization team, new role assignments for personnel, and use of a diagnostic catheter for non-culprit lesions and use of an interventional catheter for suspected culprit lesions. We used statistical process control (SPC) methods to monitor D2B in real-time, show changes resulted in improvement, and create a dashboard with control limits to identify common and special cause variation. Results: Our changes led to a 59 min decrease (p=0.0005) in median D2B time from 123 min (interquartile range = 96.5-156 min) to 64 min (interquartile range = 56-94 min). Process flow charting coupled with subinterval time measurement allowed us to identify opportunities to eliminate waste and reduce delays. We then implemented process changes which improved virtually every process step. After implementation, the new SPC charts revealed 1) only common cause variation, 2) a 67% decrease in the new upper control limit and 3) a 50% decrease in the central mean line. Conclusions: By using QI methods incorporating SPC, we mapped a process, identified bottlenecks and barriers, and implemented systematic process changes which significantly decreased D2B. We used SPC charts to confirm process improvement and provide ongoing control of a complex, inherently variable process. As quality measures align with financial incentives such as pay for performance, cardiologists and administrators will need to systematically utilize QI and SPC methods to analyze, improve and control complex processes of care. 9:00 a.m. Background: Recent studies have indicated that overweight patients may have better outcomes in the setting of chronic heart disease. We sought to determine if a similar relationship exists among patients with acute ST-elevation myocardial infarction (STEMI). Methods: 509 consecutive patients with STEMI at a single center were stratified into Body Mass Index categories according to the World Health Organization definitions of normal, overweight, and obese. Baseline characteristics (age, gender, diabetes, smoking, prior CABG, and prior PCI) and clinical outcomes were compared across BMI categories. Mulivariate analysis was used to assess the independant relationship between BMI category and in-hospital mortality. Results: Clinical outcomes are presented in Table 1. Normal BMI patients had a higher rate of in-hospital death than overweight and obese patients (p=.006). Patients with normal BMI's were older (65.9 ± 13.8 yrs) than overweight (61.2 ± 13.2) and obese (58.3 ± 12.6) patients (p<.05). However, on multi-variate analysis, overweight status remained an independant predictor of a lower rate of in-hospital death compared to normal BMI patients (OR 0.42, 95% CI 0.2-0.9, p=0.025). The odds of in-hospital deaths among obese patients was similar to that among normal BMI patients (OR .57, 95% CI 0.27-1.20, p=0.14). Conclusions: In the setting of STEMI, there appears to be a U-shaped relationship between BMI and in-hospital mortality, with overweight status associated with the lowest risk of death. Background: Post-operative infections among patients undergoing CABG surgery are concerning. This study examines the relationship between Risk-Adjusted Infection Rates (RAIR) and Evidence-Based Clinical Processes and Structures (EBCPS). Methods: This retrospective study uses data from the Heart Services Standards Database, a web-based survey of 158 HCA hospitals concerning EBCPS; and HCA Casemix Database, an administrative database of all admissions. The study population consists of 75 hospitals that performed > 52 CABG during 2005. Each hospital's RAIR was estimated from a logistic regression model controlling for 19 risk factors. Hospitals, based on their RAIR, were divided into four tiers. The EBCPS activities were classified into three similar groups: Education (10), Protocols (9), and Communications (4). Student-t tests were used to determine if top tier hospitals had more EBCPS than bottom tier hospitals. Results: The average RAIR for the top tier was 0.8% and 4.6% for the bottom tier. There was no significant difference in CABG volume between the two tiers (122 vs. 157, p=0.223). The top tier had, on average, more EBCPS implemented than bottom tier (13.6 vs. 10.4). This relationship also held for all three groups of activities, although the difference in the number of education activities was not significant (p=0.105). Conclusion: Top tier hospitals have significantly more evidence-based clinical processes and structures and lower risk-adjusted infection rates than bottom tier hospitals. Background: Retroperitoneal bleeding (RPB) is a rare complication of cardiac catheterization. We assessed differences in immediate complications and long term outcomes for patients with and without RPB. Methods: Of 17,650 consecutive patients who underwent PCI at a single center, we identified 86 RPB (.5% incidence). Fisher's exact test was used to analyze acute complications, and a multi-variate analysis was done to assess the significance of RPB on prognosis. Results: Significant associated hematologic and vascular adverse events are in Figure 1 . Short and long term prognosis are presented in Figure 2 . Conclusions: There is an increased incidence of associated acute complications in patients with an RPB. Development of an RPB was also an independent predictor for higher rate of MI and death at 1 month and 1 year. Special Topics % vs. 40%, p<0.05) when treated by female PCPs compared to male PCPs. Conclusions: When treated by female PCPs, female hypertensives 56 years or older have a more well controlled blood pressure compared to when treated by male PCPs. Patients of both gender had more well-controlled lipids when treated by female physicians compare to when treated by male physicians. Gender disparities should be minimised. 9:00 a.m.
doi:10.1016/j.jacc.2007.01.041 fatcat:4sbuu5mngfdn3moqqpqy3iwzpi