Thank you Dr. Edwards. The Surgical Care Improvement Project, or SCIP, is significant due both to its practical and conceptual impacts. Practically, participation in the SCIP program is related to public perceptions of quality for institutions and to financial reimbursement. More importantly, SCIP embodies the concept of a broad-based consortium to standardize and modify surgical practice.
The objectives for our discussion today include: Defining the goals and organization of SCIP Discussing the conceptual basis for SCIP elements Reviewing the performance measures and current national and VA data Exploring the general impact of process and performance measures for quality improvement in surgery. Due to time limitations, I will not be able to discuss in detail the scientific basis for each SCIP performance measure.
SCIP is a unique consortium of key healthcare agencies. Those listed here are on the SCIP Steering Committee and have worked collaboratively to define, discuss and debate various elements of surgical processes. Importantly, SCIP is intended to unite surgeons, anesthesiologists, perioperative nurses, critical care specialists, and hospital administration. SCIP has a national expert panel that is comprised by many important societies including the Surgical Infection Society, the American Academy of Orthopedic surgeons, the American Society of Anesthesiologists, the American Hospital Association, the American College of Surgeons, and AORN. SCIP is not just SIP. As many of you will recall, SIP, or the Surgical Infection Prevention Program, was started in August of 2002 with a focus on antibiotic usage for operative procedures. The underlying concept of SCIP is that surgical care is not the sole responsibility of a surgeon, but rather a system of clinical and administrative personnel that must work in concert to enable the delivery of high quality care.
When considering SCIP, it is critical that we understand the reasons that it exists. The well known IOM reports indicated that 69% of adverse events and deaths are related to “errors” in care. Lucian Leape has described these types of errors, which include diagnostic failures, treatment errors, errors in prevention, and others including communication failure and equipment failure. Due to the rate of surgical complications and their impact on patients and finances, surgery has been considered a key area for targeted improvement in errors.
As we review specific performance measures, keep in mind that the stated goal of SCIP is to reduce surgical mortality and preventable surgical morbidity 25% within US healthcare by 2010. As recently as 2006, SCIP stated that compliance to its performance measures for Medicare beneficiaries could prevent over 13,000 deaths and tens of thousands of complications. The concept that compliance with evidenced-based processes will decrease complications and deaths is the central hypothesis for SCIP. Let me stop here and ask for your thoughts and opinions.
Although hospital participation in SCIP was initially quite limited, it is much more widespread subsequent to the end of 2005. A key motivator was the proposed IPPS rule that suggested that hospitals should start reporting SIP measures in January of 2006 to avoid a 2% reduction of their Medicare annual payment update. Although reporting was not required until July of 2006, just the possibility of a payment reduction was sufficient motivation to increase participation. And as many clinicians are aware, CMS seems to be moving in the direction of increased financial implications for noncompliance with SCIP.
SCIP is best considered by its four preventable complication modules. These include: Prevention of surgical infection, Prevention of cardiovascular complications, Venous thromboembolism prophylaxis, and Prevention of postoperative respiratory complications. Within the VA, the prevention of ventilator associated pneumonia has been assigned primarily to critical care centers instead of surgery so we will not include a detailed discussion of it today.
As of 2006, 7 performance measures were proposed related to prevention of surgical infections. The first three deal with antibiotic usage.
And the final four address other elements that might decrease surgical infections. The fifth, which was an outcome, related to surgical site infections diagnosed during the index hospitalization was removed. Control of blood sugar for cardiac surgery patients, appropriate hair removal, and normothermia for colectomy patients are the others. Each of these is endorsed by NQF with the exception of element 7. In fact, for the VA, this element is not a current performance measure.
1. Most of us are aware of the work of Dr. Polk and others in the 1960’s that demonstrated the value of prophylactic antibiotics for decreasing surgical site infection rates. Importantly, an antibiotic with an appropriate spectrum of coverage for organisms anticipated to be exposed to the wound must be selected; and administration must be timed so as to ensure adequate tissue levels of the agent at the time of skin incision and throughout the procedure. 2. Studies generally do not demonstrate any additional benefit of prophylaxis when extended after closure of the incision, and several studies have indicated a higher infection rate when antibiotics are continued beyond 24 hours. While there are frequent questions about the role of antibiotics when drains are left in proximity to prosthetics during vascular or orthopedic procedures, expert consensus is to not continue antibiotics beyond 24 hours. Cardiac surgery patients are permitted to receive antibiotics until 48 hours postoperatively. 3. Hyperglycemia, particularly in cardiac surgery patients has been recognized as a risk factor for postoperative infection and is associated with higher in-hospital deaths. In fact, the Portland Protocol of glucose management has demonstrated a reduction in surgical site infections after cardiac surgery, and other reports have been confirmatory. 4. Shaving greater than 12 hours before surgery has been accepted as a risk factor for SSI, but shaving immediately prior to the procedure has also been suggested to increase SSI rates in some trials. 5. Data regarding the importance of normothermia for SSI are more controversial although the potential consequences of hypothermia for MI rates, transfusion frequency, and other perioperative occurrences are also recognized.
For the first 3 elements of SCIP that relate to antibiotics, there are 7 types of cases that are included. Within the VA, these are analyzed separately, but are grouped for final reporting. Thus, there is a single score for each institution for each element and not national monitoring of performance for each case type.
Nearly all knowledgeable surgeons agree with the concepts for SIP 1, 2, and 3. And many of us believe that our practices are consistent with the literature. However, a review of 39,000 Medicare patients undergoing surgery in US hospitals during 2001 demonstrated several interesting patterns. In general, practitioners selected appropriate antibiotics. But only 50% of the time were they administered at an appropriate time; and less than 50% of the time were they stopped within 24 hours. Note that the expansion of SCIP in 2005 resulted in improvement for these latter two factors which has generally continued.
And while the average performance is improved, as of the third quarter of 2007, there were significant deviations in compliance rates among facilities. National averages are presented here in yellow. Benchmark rates in green are determined by data submitted to the HQA from the highest 10% performing hospitals. The low 10% performing hospitals are represented by orange bars. Remember that a goal of SCIP is to standardize certain elements of surgical practice. Many facilities are doing well as evidenced by the average approaching the benchmark; however, it is also clear that there are subgroups of facilities that have not
How does the VA system performance compare to the overall national average? Quite well in most of these categories. For selection of an appropriate antibiotic, the VA is currently exceeding both the national average and the target level of 95%. Note that for this and each of the following slides with VA data, that the “national average” refers to all reporting hospitals and not just those within the VA. I should note that these measures continue to be refined based upon new literature. An excellent example is the inclusion of ertapenem as an acceptable, and probably preferred agent, for colorectal surgery in consequence to Dr. Itani’s New England Journal of Medicine study.
For appropriate timeliness of initial administration, the VA initially evidenced a rate of about 60% in early 2005. Recall that the baseline national average in 2001 was 56%, and then 78% by January 2005. Although we were a initially behind, there was rapid improvement such that we now exceed national averages and are meeting the target level of 95%. For many of our institutions, significant improvement occurred when anesthesia services agreed to participate in SCIP and accepted responsibility for timing of administration. The intent of SCIP to enlist the support of non-surgeons in surgical processes has probably been most successful with this element.
Trends for discontinuation of antibiotics within 48 hours for cardiac surgery and within 24 hours for all other cases are similar. In 2001, the national average for from the Bratzler study was 41% and improved to 65% in early 2005. At the start of 2005, the VA had slightly lower compliance than the national average, but quickly improved. The VA system now exceeds both the national average of about 76% and the target of 88%.
For the three final Infection Prevention elements, these national, non-VA data indicate that hair removal technique is very good, but glucose control and normothermia are compliant in 80-85% of cases. Note again that the low 10% of hospitals perform very inconsistently.
Glucose control to less than 200 for the first two postoperative days has been of interest to many within cardiac surgery nationally even before SCIP. VA performance has been good and exceeds national averages. Despite significant efforts by many cardiac programs, there are still up to 10% of patients for whom control is very challenging, and research to define optimal management protocols is ongoing.
VA has done very well with not shaving for hair removal. Hair removal is not required, but when desired, either a depilatory or clippers are acceptable. In 1992, Cliff Ko studied nearly 2000 patients undergoing cardiac surgery and demonstrated that shaving on the morning of surgery was associated with a higher SSI rate. Kjonniksen et al in 2002 performed a metanalysis and noted that hair removal per se did not increase SSI unless shaving was used.
Published research has correlated impaired wound healing, adverse cardiac events, altered drug metabolism, and coagulopathies with unplanned perioperative hypothermia. A study by Kurtz, et al in 1996 found that the incidence of culture-positive surgical site infections among those with mild perioperative hypothermia was three times higher than normothermic perioperative patients. He also noted an association of mild perioperative hypothermia with delayed wound closure and prolonged hospitalization. In a meta-analysis of outcomes and costs, Mahoney and Odom (1999), demonstrated that hypothermia is associated with a significant increase in adverse outcomes, including an increased incidence of infections. The authors also concluded that hypothermia is associated with an increased chance of blood products administration, myocardial infarction, and mechanical ventilation. These adverse outcomes resulted in prolonged hospital stays and increased healthcare expenditures. A multinational group led by Dan Sessler performed a randomized trial that also suggested that SSI rates were lower when mild hypothermia was avoided during the procedure. Controversies regarding the quality of data to support this measure and regarding acceptable techniques for measuring temperature have resulted in this performance measure being removed from the SCIP Measure Set by Joint Commission. It is now an optional measure for Joint Commission, but OQP collects these data as a supporting indicator. As a consequence, there is no current target that is defined. Before we move to the next category, let me ask if there are any brief questions or comments.
The second of the four preventable complication modules in SCIP relates to prevention of cardiovascular complications.
In 2006 there were three proposed measures related to prevention of cardiovascular complications. Of these, only the second is active. This indicator requires that patients who have been receiving a beta blocker preoperatively should have dosing continued during the 24 hour perioperative period.
Beta blocker continuation during the immediate perioperative period has improved from about 83% in October 2006 to about 91% during the first quarter of the current fiscal year. It is acceptable for the beta blocker to be given orally preoperatively or by anesthesia in the holding area or in the OR. Some interesting system issues have arisen regarding this measure. For instance, inpatients may miss their medication dose on the morning of surgery because the default time for morning medications is after the time that the patient leaves the inpatient ward to go to the holding area. In other programs, cardiac surgeons are less aggressive about use of beta blockers for rate control in some patients with ventricular dysfunction and prefer to use amniodarone. Data for the second quarter of the fiscal year has been recently released and is indicative of continued improvement.
The recent publication of results of the POISE trial has prompted significant discussion and controversy regarding appropriate usage of beta blockers. Lindenauer and other authors’ publications have previously suggested that beta blocker usage for patients with known or significant risk of cardiac disease reduces cardiac events and deaths. The ACC/AHA guidelines for management of cardiac disease in non-cardiac surgery patients were updated within the past year and are based on such studies. A smaller set of studies have not identified the same degree of benefit for beta blocker usuage, however.
The POISE Trial was a large multinational trial funded primarily by the Canadian Institute of Health Research. It involved nearly 200 hospitals in 23 countries and randomized over 8000 patients to receive either extended release metoprolol or placebo. Note that medication was started in the immediate preoperative period and that there was no adjustment permitted for tachycardia, hypotension, or neurologic symptoms postoperatively.
Results were just published in the May edition of Lancet. Note that MI rates were decreased by metoprolol consistent with several prior trials. What has been of interest with the POISE trial is the higher rate of CVA, hypotension, and death in the beta blocker arm. The authors concluded, “Our results highlight the risk in assuming a perioperative β-blocker regimen has benefit without substantial harm, and the importance and need for large randomised trials in the perioperative setting. Patients are unlikely to accept the risks associated with perioperative extended-release metoprolol. The trial has generated significant discussion and even frank contention. Some caveats are important to note. Due to the blinded design of the trial, dosage adjustment could not be permitted in the metoprolol group. Critics have questioned that the CVA and associated death rates might have been lower if knowledgeable clinicians were able to titrate or even discontinue beta blockers when appropriate. While discussion about the implications to clinical practice are ongoing, note that the current SCIP performance measure is not impacted directly by this trial.
The final of the preventable complication modules in SCIP which we will discuss today relates to prevention of venous thromboembolism. Many clinicians tend to minimize the importance of VTE in their personal practices. However, there are more than 200,000 new cases per year, with the vast majority occurring in hospitalized patients. In fact for every one episode in an outpatient that was not recently hospitalized, there are nearly 1000 cases in hospitalized or recently discharged patients.
Leapfrog has stated that VTE is “the most common preventable cause of hospital death in the United States.” And AHRQ has stated that “thromboprophylaxis is the number one patient safety practice.” Yet the disconnect between evidence for DVT prophylaxis and practice has led the American Public Health Association to term it a “public health crisis.” Although it is beyond the scope of our discussion today, please recognize the importance of familiarity with national recommendations for appropriate VTE prophylaxis. Key principles include standardization of approaches and tying appropriate intervention to risk assessment. Note that both the Joint Commission and VACO are likely to introduce a requirement for risk assessment for all inpatients as a safety goal within the next year.
As with some of the other modules, only a subset of the originally proposed measures are active at present. Currently, each patient should be ordered VTE prophylaxis and should receive it within 24 hours of surgery. Outcomes related to 30 day incidence of PE and DVT are not currently included.
SCIP measures for VTE prevention are based solely upon class 1A data from the American College of Chest Physicians consensus statement. As has been the pattern for other measures within SCIP, pre-intervention data were obtained by Medicare. For VTE measures, about 70% of patients received VTE prophylaxis prior to SCIP based upon a national sample of over 19,000 Medicare patients during the first quarter of 2005. Despite over 2 years of these measures, national compliance remains in the 80 to 85% range.
VA utilization of guidelines for ordering VTE prophylaxis have been good overall and currently exceed the national average in private sector and target levels.
Similarly, VA rates of compliance with patients receiving appropriate VTE prophylaxis is now above both the target level and the national non-VA average.
VHA has been an important contributor to the development of surgical process measures for SCIP. Further, VA surgery programs have also been highly effective in compliance with SCIP measures. The recently released VHA OIG report on SCIP had no recommendations for improvement. For those of you that are familiar with the VHA OIG, this is a tremendous compliment and testifies to the excellent work that has been done by so many. I would like to particularly highlight Dr. DePalma’s role as the VHA representative to SCIP and liaison with CMS and Dr. Itani who has been the VA chair for the SIP measures.
Despite the enthusiasm from the OIG and excellent success overall with SCIP within the VA, we need to stop and ask about the impact of SCIP specifically and the relevance of process measures more generally. Patch Dellinger looked at early results from the implementation of SIP elements in 2002 and 2003 and noted a decrease of about 27% in overall surgical infections.
What about mortality? Since 2004 there has been a decrease in mortality for Medicare patients undergoing SCIP-related operations. But this is an actual percentage of 0.37% or a relative percent improvement of about 9.5%. Recall that the goal of SCIP was a 25% decrease in deaths and morbidities. So, how are we doing? I guess that it is a matter of perspective. But is this improvement a consequence of SCIP or of changes in more global aspects of patient care? If related to SCIP, is it the change in processes per se or a general phenomenon of more attention being given to these patients because we are being watched? At the least, time will be needed to determine whether improvement continues.
While it seems intuitive that implementation of evidence-based processes should improve outcomes, not all data support that such focused interventions are sufficient. A key study was published last month in JACS by Mary Hawn, Dr. Itani and their colleagues. They reviewed over 9000 VA patients who underwent SCIP procedures and for whom data were available in NSQIP and regarding compliance with SCIP-1 for timely administration of prophylactic antibiotics. A variety of patient and facility analyses were performed.
Unadjusted surgical site infection rates were higher in patients that did not receive timely antibiotics, but risk adjusting for SSI risk resulted in no correlation between SIP compliance and SSI rate. Thus, the single factor of timely antibiotic administration did not contribute to either patient or facility SSI rates.
In conclusion, some observations. Focus on surgical outcomes will continue and almost certainly will increase, largely related to the cost of complications. Surgeons will not really have a choice regarding compliance due to evolving financial implications of non-participation and/or non-compliance. Surgical process measures are also here to stay….whether motivated by the opportunity to quantify elements of care for financial reasons or by an underlying belief that standardization will improve outcomes. Tremendous resources are required to operationalize process measures. We will need to become thoughtful about the balance between the reasonable potential for improvement compared to resource availability. I hope that you have gained an understanding of the mechanics of SCIP today. More importantly, I trust that you will also appreciate the complexity of the underlying premises and the challenges related more generally to quality improvement in surgery. I’d be happy to entertain questions but hope that some of you will comment about the role of process measures in surgical quality improvement.
Surgical Care Improvement Project Mark A. Wilson, MD, PhD Vice-President, Surgery VA Pittsburgh Healthcare System
Objectives <ul><li>Define goals and organization of SCIP </li></ul><ul><li>Discuss conceptual basis for SCIP elements </li></ul><ul><li>Review performance measures and current data </li></ul><ul><li>Explore the relevance of process measures to quality improvement in surgery </li></ul>
What is SCIP? <ul><li>American College of Surgeons </li></ul><ul><li>American Hospital Association </li></ul><ul><li>American Society of Anesthesiologists </li></ul><ul><li>Association of peri-Operative Registered Nurses </li></ul><ul><li>Agency for Healthcare Research and Quality </li></ul><ul><li>Centers for Medicare & Medicaid Services </li></ul><ul><li>Centers for Disease Control and Prevention </li></ul><ul><li>Department of Veteran’s Affairs </li></ul><ul><li>Institute for Healthcare Improvement </li></ul><ul><li>Joint Commission on Accreditation of Healthcare Organizations </li></ul>
Why SCIP? <ul><li>69% of adverse events and deaths in healthcare are related to “errors” and are thus potentially preventable. (IOM) </li></ul><ul><li>2.6% of ~ 30 million operations in the US -> SSI; significant impact on LOS, finances, etc. http://www.ihi.org/ihi/Topics/PatientSafety/SurgicalSiteInfections/SurgicalSiteInfectionsCaseForImprovement </li></ul><ul><li>7-8 million operated patients per year with significant cardiac risk factors and at least 1 million cardiac events annually </li></ul><ul><li>Significant risks for perioperative venous thromboembolism </li></ul>
SCIP Goal <ul><li>To reduce preventable surgical morbidity and mortality by 25% by 2010 </li></ul><ul><li>SCIP constituents believe that Medicare could annually prevent up to 13,027 perioperative deaths and 271,055 surgical complications in major surgical cases by a high level of compliance with evidence-based processes for surgical care. </li></ul>
Voluntary Reporting Hospitals Deficit Reduction Act of 2005 0 500 1000 1500 2000 2500 3000 3500 4000 # Hospitals 2002 2003 2004 2005 2006 2007 Number of Reporting Hospitals
Surgical Infection Prevention <ul><li>SCIP INF 1: Prophylactic antibiotic received within one hour prior to surgical incision </li></ul><ul><li>SCIP INF 2: Prophylactic antibiotic selection for surgical patients </li></ul><ul><li>SCIP INF 3: Prophylactic antibiotics discontinued within 24 hours after surgery end time (48 hours for cardiac patients) </li></ul>
Surgical Infection Prevention - 2 <ul><li>SCIP INF 4: Cardiac surgery patients with controlled 6 a.m. postoperative serum glucose </li></ul><ul><li>SCIP INF 5: Postoperative wound infection diagnosed during index hospitalization (Outcome) </li></ul><ul><li>SCIP INF 6: Surgery patients with appropriate hair removal </li></ul><ul><li>SCIP INF 7: Colorectal surgery patients with immediate postoperative normothermia </li></ul>
Rationale <ul><li>Reduction of SSI when tissue levels of antibiotics are appropriate at time of surgery </li></ul><ul><li>No demonstrated benefit to prophylaxis postoperatively, and higher infection rates if antibiotics are continued beyond 24 hours </li></ul><ul><li>Hyperglycemia contributes to SSI risk </li></ul><ul><li>Shaving pre-operatively increases SSI </li></ul><ul><li>Data to support a reduction of SSI rates when normothermia is maintained are controversial. </li></ul>
SCIP 1-3 <ul><li>7 case types that are included: </li></ul><ul><li>CABG </li></ul><ul><li>Other cardiac </li></ul><ul><li>Colon surgery </li></ul><ul><li>Hip arthroplasty </li></ul><ul><li>Knee arthroplasty </li></ul><ul><li>Hysterectomy </li></ul><ul><li>Vascular surgery </li></ul>
Evolution of National Performance *National sample of 39,000 Medicare patients undergoing surgery in US hospitals during 2001. Bratzler DW, Houck PM, et al. Arch Surg . 2005; 140:174-182 . //
National Performance Q3 2007 0 20 40 60 80 100 Antibiotics w/i 1 hour Correct Antibiotic Antibiotic DC’d w/i 24 hours Percent National Average Benchmark Low 10%
Correct Prophylactic Antibiotic Selection National VA Data Note : Nationwide Average among all JC-accredited healthcare organizations for the July 06-June 07 period: 93% Jun 06 through Nov 07 0% 20% 40% 60% 80% 100% Jun-06 Jul-06 Aug-06 Mar-07 Apr-07 May-07 Jun-07 Jul-07 Aug-07 Oct-07 Nov-07 Score Average Target FY08 Target = 95% National Average (VA/non-VA) = 94.2%
Prophylactic Antibiotic Started Timely National VA Data SIP – Inpt – Prophylactic Antibiotics Started Timely (sip 1a) Note : Nationwide Average among all JC-accredited healthcare organizations for the July 06-June 07 period: 89% 0% 20% 40% 60% 80% 100% Jan 05 Jan 06 Mar 07 Nov 07 FY08 Target = 95% Average = 85.2% Score Average Target
Prophylactic Antibiotics DC’d Timely National VA Data Note : Nationwide Average among all JC-accredited healthcare organizations for the July 06-June 07 period: 82% Note : Nationwide Average among all JC-accredited healthcare organizations for the July 06-June 07 period: 82% 91.0% 0% 20% 40% 60% 80% 100% Oct 05 Jan 06 Mar 07 Nov 07 Average = 75.8% FY08 Target = 88% Score Average Target
National Performance Q3 2007 0 20 40 60 80 100 Percent Glucose Control (cardiac) No Razor Normothermia National Average Benchmark Low 10%
Glucose Levels – Cardiac Surgery National VA Data 93.2% 0% 20% 40% 60% 80% 100% Oct 05 Jan 06 Jan 07 Nov 07 FY08 Target = 95.0% Average = 87.0% Score Average Target
Hair Removal By Acceptable Method National VA Data Score Average Target 99.9% 0% 20% 40% 60% 80% 100% Oct 06 Jan 07 Nov 07 FY08 Target = 95% Average = 99.2% Score Average Target
First Temp in Range – Colon Surgery National VA Data Score Average Target 82.3% 0% 20% 40% 60% 80% 100% Oct 05 Jan 06 Jan 07 Mar-07 Apr-07 May-07 Jun-07 Jul-07 Aug-07 Oct-07 Nov-07 66.8% = Average Score Average Target
Colon Surgery – Normothermia Postop National VA Data 0% 20% 40% 60% 80% 100% Oct 06 Jan 07 Aug 07 FY07 Target = 70% Average = 69.8% Score Average Target Note: This measure was discontinued at the end of FY2007.
Cardiovascular Complication Prevention <ul><li>SCIP Card 1: Non-cardiac vascular surgery patients with evidence of coronary artery disease who received beta-blockers during the perioperative period </li></ul><ul><li>SCIP Card 2: Surgery patients on a beta-blocker prior to arrival that received a beta-blocker during the perioperative period </li></ul><ul><li>SCIP Card 3: Intra- or postoperative acute myocardial infarction (AMI) diagnosed during index hospitalization and within 30 days of surgery (Outcome) </li></ul>
Beta Blocker Usage National VA Data 90.8% 0% 20% 40% 60% 80% 100% Oct 06 Jan 07 Nov 07 FY08 Target = 92% Average = 84.7% Score Average Target Surgery Pts on Beta-Blocker Therapy Prior to Admission Who Received a Beta-Blocker During the Perioperative Period
Role of Beta Blockers ??? The Lancet 2008; 371 :1839-1847 Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial POISE Study Group
POISE Trial <ul><li>190 hospitals, 23 countries </li></ul><ul><li>8351 patients with, or at risk of ASHD undergoing non-cardiac surgery </li></ul><ul><li>Randomized to double-blinded receipt of extended release metoprolol or placebo </li></ul><ul><li>Started 2-4 hours preop and continued for 30 days postop </li></ul><ul><li>No dosage adjustment </li></ul>
POISE Trial - Results MI findings are consistent with prior trials Hypotension was more common in metoprolol group….? contributor to stroke and death Would titration by experienced clinicians decrease CVA and/or death rates for beta blockers? Patient criteria for beta blockers and time of initiation continue to be discussed
Venous Thromboembolism <ul><li>Leapfrog: VTE is “the most common preventable cause of hospital death in the United States.” </li></ul><ul><li>AHRQ: “Thromboprophylaxis is the number one patient safety practice.” </li></ul><ul><li>American Public Health Association: VTE prophylaxis is a “public health crisis.” </li></ul><ul><li>Guidelines: </li></ul><ul><ul><li>American College of Chest Physicians </li></ul></ul><ul><ul><li>Intervention must be tied to risk assessment </li></ul></ul><ul><ul><li>Evolving Joint Commission patient safety goal </li></ul></ul>
Thromboembolism Prevention <ul><li>SCIP VTE 1: Surgery patients with recommended venous thromboembolism prophylaxis ordered </li></ul><ul><li>SCIP VTE 2: Surgery patients who received appropriate venous thromboembolism prophylaxis within 24 hours prior to surgery to 24 hours after surgery </li></ul><ul><li>SCIP VTE 3: Intra- or postoperative pulmonary embolism (PE) diagnosed during index hospitalization and within 30 days of surgery (Outcome) </li></ul><ul><li>SCIP VTE 4: Intra- or postoperative deep vein thrombosis (DVT) diagnosed during index hospitalization and within 30 days of surgery (Outcome) </li></ul>
Evolution in National Performance National Data for All Hospitals // 82.4 84.1 84.8 86.3 77.8 79.6 80.5 82.1 71.9 69.7 60 70 80 90 100 Q1, 2005* Q2 2006 Q3 2006 Q4 2006 Q1 2007 Q2 2007 Q3 2007 Percent Recommended VTE prophylaxis VTE prophylaxis received
Appropriate VTE Prophylaxis Ordered National VA Data 95.1% 0% 20% 40% 60% 80% 100% Oct 06 Jan 07 Nov-07 FY08 Target = 92% Average = 86.8% Score Average Target
Received Appropriate VTE Prophylaxis National VA Data 92.4% 0% 20% 40% 60% 80% 100% Oct 06 Nov 06 Dec 06 Jan 07 Feb 07 Mar-07 Apr-07 May-07 Jun-07 Jul-07 Aug-07 Oct-07 Nov-07 FY08 Target = 90% Average = 83.4% Score Average Target
VHA “Report Card” on SCIP <ul><li>“ We concluded that all facilities evaluated during the CAP reviews implemented strategies to prevent or reduce the incidence of surgical infections. ….For those measures that were below VHA’s established goals, managers implemented appropriate action plans to improve performance. </li></ul><ul><li>We made no recommendations . “ </li></ul><ul><li>VHA OIG, Healthcare Inspection, Surgical Quality Improvement Program, March 2008 </li></ul><ul><li>http://www.va.gov/oig/54/reports/VAOIG-07-00773-106.pdf </li></ul>
SCIP Impacts Outcomes from SIP… Overall surgical infection rate decreased 27%, from 2.28% in the first 3 months to 1.65% in the last 3 reporting months. Dellinger EP, et al. Am J Surg . 2005;190:9–15.
National 30-day All Cause Mortality Non-cardiac surgery SCIP Impacts Limited to all Medicare patients undergoing those operations included in SCIP. 3.89 3.78 3.52 3.88 0 1 2 3 4 5 2004 2005 2006 2007 30-day mortality % 75,940 deaths 1,951,669 operations 75,167 deaths 1,938,962 operations 71,312 deaths 61,577 deaths 1,748,860 operations 1,887,105 operations GOAL
Self Analysis <ul><li>J Am Coll Surg. 2008 May;206(5):814-9 </li></ul>Association of timely administration of prophylactic antibiotics for major surgical procedures and surgical site infection. Hawn MT , Itani KM , Gray SH, et al. Patients with EPRP SCIP-1 and NSQIP data were studied Patient and facility level analyses comparing SCIP-1 and SSI were performed Adjustment for clustering effects within hospitals, validation of SSI risk score and procedure type (percentage of colon, vascular, orthopedic) 9,195 elective procedures (5,981 orthopedic, 1,966 colon, and 1,248 vascular) in 95 VA hospitals.
<ul><li>Timely antibiotic administration occurred in 86.4% of patients who had an SSI rate of 4.6%; untimely administration was associated with SSI rate of 5.8% in unadjusted analysis </li></ul><ul><li>Patient level risk-adjusted multivariable generalized estimating equation modeling found the SSI risk score was predictive of SSI (p < 0.001) and SIP-1 was not associated with SSI. </li></ul><ul><li>Hospital level multivariable linear modeling found procedure mix (p < 0.0001), but not SIP-1 rate or facility volume, to be associated with facility SSI rate. </li></ul><ul><li>The study had 80% power to detect a 1.75% difference for patient level SSI rates. </li></ul><ul><li>Timely antibiotic administration did not markedly contribute to overall patient or facility SSI rates. </li></ul>
Observations <ul><li>Focus on surgical outcomes will continue </li></ul><ul><ul><li>driven largely by financial issues (payer cost and provider compensation) </li></ul></ul><ul><ul><li>the right thing for all of us anyway </li></ul></ul><ul><li>Surgical process measures are increasingly accepted….data to assess efficacy are needed. </li></ul><ul><li>Implications regarding P4P are significant! </li></ul><ul><li>Definition of the processes that are of sufficient clinical importance to warrant resource commitment for standardization is critical </li></ul>
Some helps… An excellent summary of the background for SCIP elements: http://vaww.visn1.med.va.gov/Estrada.config?resource=52620 VA SCIP data: http://vaww.oqp.med.va.gov SCIP sites: http://www.qualitynet.org http://www.medqic.org/scip/ Special thanks to Dale Bratzler, DO; Chair, SCIP Steering Committee, Oklahoma OIFO for sharing national SCIP data