The Impact of the Surgical Care and Outcomes Assessment Program (SCOAP)

Transcription

1 The Impact of the Surgical Care and Outcomes Assessment Program (SCOAP) David R. Flum, MD, MPH Sean D. Sullivan, PhD Rebecca Gaston Symons, MPH This project is funded by Life Sciences Discovery Fund Award No The Surgical Care and Outcomes Assessment Program is a program of the Foundation for Healthcare Quality operated under Washington State s Department of Health as Coordinated Quality Improvement Program (CQIP). University of Washington s Surgical Outcomes Research Center serves as the academic home for research. Primary collaborators in this technical report include: David Flum MD, Allison Devlin Rhodes MS, Rebecca Gaston Symons MPH, Sarah O. Lawrence MA (Surgical Outcomes Research Center, University of Washington); Sean Sullivan PhD (Pharmaceutical Outcomes Research and Policy Program, University of Washington); William O Brien (Department of Epidemiology, University of Washington); Justine Norwitz (JSN Consulting, Inc); and Terry Rogers MD, Rosa Johnson ARNP, MN, CPHQ (Foundation for Health Care Quality).

2 SCOAP Technical Report (PI: Flum) Page 2 of 13 Introduction The Surgical Care and Outcomes Assessment Program (SCOAP) is a state-wide, multi-hospital quality improvement program initiated in 2006 and expanded in 2008 through a grant from the State of Washington Life Sciences Discovery Fund (LSDF). The purpose of SCOAP is to improve the safety, quality and efficiency of surgical care across statewide hospitals. As part of our commitment to evaluating the impact of SCOAP in Washington state, we have undertaken a number of studies to estimate the impact of surgical quality improvement programs undertaken in SCOAP hospitals on measures of quality, outcomes and hospital resource use and costs. The purpose of this specific evaluation is to determine the resource use and cost implications of quality improvement initiatives in SCOAP hospitals. Three surgical procedures were selected for this evaluation (elective colectomy, gastric bypass, and non-elective appendectomy). Methodology Data Source We used two data sources for this research: the SCOAP hospital database and the Washington State Comprehensive Hospital Abstract Reporting System (CHARS) database. CHARS includes discharge abstract data from all public and private acute-care hospitals in Washington State (excluding Veterans Affairs and military hospitals). Each record of the dataset contains demographic variables, admission and discharge administrative details, International Classification of Diseases, Ninth Revision (ICD-9) procedure and diagnosis codes and coded hospital identifiers. We identified all cases of colectomy, gastric bypass for severe obesity and non-elective appendectomy in the CHARS database for 2006 through 2008 along with all subsequent readmissions. Index cases were identified based on ICD-9 procedure codes (Appendix A). The charlson index, a summary value of comorbid conditions derived from ICD-9 procedure codes, was used to characterize the analytic cohort.. Readmissions were probability matched to SCOAP index procedures based on available patient identifiers. The SCOAP hospital data includes information about how the procedures were done and the results of those procedures. SCOAP data were not used in this economic impact assessment, but hospitals that were using SCOAP were characterized as SCOAP hospitals. Variable Definitions SCOAP procedures Cases were categorized as SCOAP procedures if they occurred during and after a hospital s first quarter of SCOAP participation. All other procedure records were categorized as Non-SCOAP. Cost The CHARS dataset provides a value for total hospital charges for each inpatient record. Costs in dollars were computed by adjusting charges with the year-specific WA average ratio of charge to cost (RCC). Then, all costs were expressed in 2008 monetary values using the year-specific CPI (hospital services) conversion factor. Length of Stay (LOS) The length of stay for each record is provided as the discharge date minus the admit date. The resulting length of stay is reported in whole days.

3 SCOAP Technical Report (PI: Flum) Page 3 of 13 Cost per day (C/D) The total cost divided by the number of days for each record is reported as the average cost per day. Adverse events (AE) Both index admissions and 30-day readmissions were evaluated for adverse outcomes. Adverse outcomes were identified based on ICD-9 diagnosis and procedure codes (Appendix B) Readmissions Admissions that occurred within 30 days of discharge from the index admission were considered for evaluation. Readmission hospital may or may not be the same as the index hospital. The costs of the readmissions were added to the index admission cost to calculate the total 30-day cost of care. Analysis Means of continuous variables are compared across groups using t-test and ANOVA, depending on the set of independent variables being considered. Relationships between categorical variables are evaluated using Pearson s X 2. Ordinary least squares regression was used to evaluate the relationship between multiple independent variables and total cost, average length of stay and cost per day. Independent variables included a Year specific dummy variable, a measure of hospital teaching status and geographic location, and an indicator of whether or not an adverse event occurred as part of the procedure.

4 SCOAP Technical Report (PI: Flum) Page 4 of 13 Results Washington State Hospitals & SCOAP 82 Washington hospitals were identified as having done at least one of the surgical procedures of interest during the years 2006 to In 2006, the average number of SCOAP participating hospitals was 13. In 2007 that number increased to 18 and increased again in 2008 to 27 (as of January 2010 the number of participating institutions is 56). By the fourth quarter of 2008, 30 hospitals were participating in SCOAP; (36% of the Washington hospitals performing the surgical procedures under consideration). Figure 1: Number of SCOAP hospitals by Quarter Number of SCOAP Hospitals Q Q2 Q3 Q4 Q Q2 Q3 Q4 Q Q2 Q3 Q4 Between 2006 and 2008, 9,440 cases of elective colectomy, 3,016 cases of gastric bypass and 14,445 cases of non-elective appendectomy were identified in the CHARS database. Cases were excluded if the patient was less than 18 years old. In 2006, SCOAP procedures accounted for 30% of elective colectomy, 39% of gastric bypass and 19% of appendectomy. In 2008, SCOAP covered 50%, 65% and 42% of colectomy, bypass and appendectomy, respectively (Figure 2). Figure2: SCOAP Penetration in Washington State (SCOAP cases by year) 70% 60% 50% 40% 30% 20% 10% 0% Colectomy Bypass Appendectomy

5 SCOAP Technical Report (PI: Flum) Page 5 of 13 The SCOAP Population To account for differences in patients and hospitals that may account for cost differentials we describe these characteristics across SCOAP and non-scoap hospitals (Table 1). Statistically significant differences in characteristics were accounted for in adjusted estimates. Table 1. Procedure characteristics by SCOAP participation and procedure type. Elective Colectomy Gastric Bypass Non-Elective Appendectomy Non- SCOAP SCOAP Total p Non- SCOAP SCOAP Total p Non- SCOAP SCOAP Total p N Patient characteristics Male Age (mean) Medicaid Charlson Hospital Characteristics Teaching Hospital Rural Hospital

6 SCOAP Technical Report (PI: Flum) Page 6 of 13 Unadjusted Procedure costs For all procedures the unadjusted average cost at a SCOAP hospital was significantly less than the cost at a Non-SCOAP hospital. Among colectomy cases, the average costs were $19,591 and $17,191 for Non- SCOAP and SCOAP, respectively (p < 0.000). For bypass procedures, the Non-SCOAP and SCOAP costs were $22,192 and $14,797 (p < 0.000), respectively. For appendectomy, the costs were $9,853 and $8,083 (p < 0.000). SCOAP hospital appear to be bending the cost curve by keeping costs of care low while costs of care remain the same or increase in non-scoap hospitals. Figure3a: Unadjusted Average Cost: Colectomy $22,000 $20,000 $18,000 $16,000 $14,000 $12,000 $10, Non SCOAP SCOAP Figure3b: Unadjusted Average Cost: Bypass $24,000 $22,000 $20,000 $18,000 $16,000 $14,000 $12,000 $10, Non SCOAP SCOAP

7 SCOAP Technical Report (PI: Flum) Page 7 of 13 Figure3c: Unadjusted Average Cost: Appendectomy $19,000 $17,000 $15,000 $13,000 $11,000 $9,000 $7,000 $5, Non SCOAP SCOAP Estimated Cost Difference The adjusted cost difference is the product of the adjusted cost per day difference between SCOAP and Non-SCOAP procedures, the adjusted LOS and volume of SCOAP procedures. Cost and LOS values have been adjusted for year, adverse event, hospital location & type. (Appendix C: Regression Models) Table 2: Select coefficients from the OLS regression models Adjusted LOS Adjusted Cost/Day Non-SCOAP SCOAP Non-SCOAP SCOAP Colectomy Bypass Appendectomy Cost/Day difference 95% Conf. Interval Lower Upper Limit Limit Total Cost difference 95% Conf. Interval Lower Upper Limit Limit Colectomy ,219,000 13,472,000 16,967,000 Bypass ,167,000 9,283,000 15,051,000 Appendectomy ,425,000 9,230,000 11,620,000 The estimated total adjusted cost difference (all procedures combined): $37,812,000 (95% CI: [$31,985,000, $43,638,000])

8 SCOAP Technical Report (PI: Flum) Page 8 of 13 Projected Cost differences Based on the 2008 volume and distribution of surgical cases, adjusted LOS, adjusted cost differences & 2009 SCOAP hospital enrollment, we have projected cost differences for 2009 of $24,585,000 but estimates are limited by projection from The total projected cost difference for 2006 to 2009 would be $62,396,000. Figure 4: Total Adjusted Cost Differences by Year (Projected for 2009) 30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000, Conclusions The estimated unadjusted total cost difference (all procedures and years, combined) between SCOAP and non-scoap hospitals was $30,183,000 (after adjustment for patient and center characteristics $37,811,000). This difference represents expenses that may have been incurred had the SCOAP program not existed during this time period and may be considered as costs not incurred (savings) by the system. SCOAP hospitals nearly doubled in number after the close of this data window and based on projections extended from 2008 data it may be that there have been over $60 million in cost differences by the end of The differences in costs are the result of decreased intensity of care (cost/day) at SCOAP hospitals but length of stay is only lower amongst selected procedures. A more refined analysis using robust 2009 and 2010 data, and pre- and post-scoap, by hospital analyses will be forthcoming when those CHARS data are released.

9 SCOAP Technical Report (PI: Flum) Page 9 of 13 Appendices Appendix A: ICD-9 procedure codes for index case identification Colectomy Bypass Appendectomy

10 SCOAP Technical Report (PI: Flum) Page 10 of 13 Appendix B: ICD-9 diagnoses and procedure codes for included adverse events Complications included during index admission Diagnosis or procedure code* Cardiac arrest Pulmonary Temporary tracheostomy 31.1* Continuous mechanical ventilation for 96 consecutive hours or more 96.71* Pulmonary embolism and infarction 415.1, Iatrogenic pneumothorax (postoperative) Pulmonary edema, postoperative Pulmonary insufficiency following trauma and surgery Urinary complications, specified as due to a procedure Infectious Respiratory complications, pneumonia, resulting from a procedure Postoperative infection 998.5, , Percutaneous drainage or abdominal wall incision for abscess 54.0, 54.91* Postoperative shock Wound related Disruption of operation wound 998.3, , Non-healing surgical wound Persistent postoperative fistula Procedure related Reopening of recent thoracotomy site 34.03* Hemorrhage, hematoma, or seroma complicating a procedure , , Foreign body accidentally left during a procedure Additional complications included if hospital readmission occurred* Cardiac

11 SCOAP Technical Report (PI: Flum) Page 11 of 13 Myocardial infarction 410.0, 410.1, 410.2, 410.3, 410.4, 410.5, 410.6, 410.7, 410.8, 410.9, 411.1, , Heart failure, acute , , , , , Acute respiratory failure , , Acute renal failure 584.5, 584.6, 584.7, 584.8, Infectious Pneumococcl pneumonia 481 Other bacterial pneumonia 482.0, 482.1, 482.2, 482.3, 482.4, 482.8, Urinary tract infection Infections of the kidney 590.1, , , 590.2, 590.8, , , Bacteremia Septicemia 38.0, 38.1, 38.10, 38.11, 38.19, 38.2, 038.3, 38.4, 38.41, 38.42, 38.43, 38.44, 38.49, 38.8, 38.9 Drainage of intraperitoneal abscess 54.19* Delirium, acute and subacute 293.0, Procedure related Incision of chest wall and pleura, insertion of intercostal catheter 34.01, 34.09, 34.04* Reopening of recent laparotomy 54.12* Esophagostomy, surture of laceration of esophagus 42.10, 42.11, 42.19, 42.82* Revision of gastric anastamosis, suture of laceration of stomach 44.5, 44.61* Revision of anastamosis of intestine 46.93, 46.94* Other repair of intestine or rectum, colostomy 46.71, 46.72, 46.73, 46.74, 46.75,

12 SCOAP Technical Report (PI: Flum) Page 12 of , 46.79, 48.71, 46.10, 46.11, 46.13, 46.14* Anastamosis or revision of gallbladder or bile duct 51.36, 51.37, * Repair of bile ducts, liver, pancreas 51.71, 51.79, 50.61, 50.69, 52.95* Suture of laceration of kidney, ureter, bladder 55.81, 56.82, 57.81* * indicates procedure codes

13 SCOAP Technical Report (PI: Flum) Page 13 of 13 Appendix C: Regression Models for Adjusted Procedure Costs and Length of Stay Colectomy Cost Length of stay Cost per day Coef. Std. Err. p Coef. Std. Err. p Coef. Std. Err. p Year Year Rural Hospital Teaching Hospital Adverse Event SCOAP procedure Intercept Bypass Cost Length of stay Cost per day Coef. Std. Err. p Coef. Std. Err. p Coef. Std. Err. p Year Year Rural Hospital Teaching Hospital Adverse Event SCOAP procedure Intercept Appendectomy Cost Length of stay Cost per day Coef. Std. Err. p Coef. Std. Err. p Coef. Std. Err. p Year Year Rural Hospital Teaching Hospital Adverse Event SCOAP procedure Intercept