Figure 18-1. Systems failures.
Figure 18-2. Overview of the reporting system.
A second way we use the safety culture survey data is to identify specific hospital units that are struggling with a dispirited or complacent attitude toward the safety effort. Such problems often result from poor nurse leadership, disruptive physicians, or a lack of perceived resources. Armed with information about the safety culture (or lack of it), the institution can implement a focused strategy customized to the problem. Figure 18-3 shows results from our survey arranged by individual nursing unit, demonstrating certain units in need of help with error reporting and, conversely, units where the culture is good and institutional resources are not needed. A third source of information derived from the survey comes from narrative comments entered by individual caregivers. This is a very rich source of information and, since it is anonymous, often draws a fine line under issues that are hard to talk about in any other forum. Issues of suspected physician impairment, abusive behavior, or lack of leadership skills are sometimes identified.
Two important questions regarding safety culture and our efforts to improve it remain. First, using the AHRQ tool, have we seen an aggregate improvement over the past two survey intervals? Many safety initiatives have been instituted over this period of time, and yet the aggregate culture data have not shown much change. We interpret this information to mean that much more work needs to be done, and that changing a culture is a hard thing to do, akin to changing direction of an aircraft carrier. Also, over the period of time we have been studying our culture, our institutional activity has gone up dramatically, the complexity of our patients has increased, and nursing turnover has been high. Under these circumstances, no change in the safety culture data might be viewed more optimistically.
A second question is, “Are there individual strategies we have used that influence the safety culture positively?” If so, we could use these strategies more broadly. The answer to this question is yes. Patient safety rounds have been an important strategy that has improved the safety culture. Over the course of the past several years, we have made safety rounds on over 200 occasions, at 2-week intervals. Safety rounds are carried out by leadership (chief medical officer, chief of nursing, CEO, etc.) and a pointed 45-minute discussion ensues with the unit caregivers, including nurses, aides, clerks, and transporters. The culture effects of this endeavor are profound. When caregivers believe that the leadership is willing to listen, takes safety very seriously, and will put resources behind the articulated concerns, an overall feeling of confidence and support of the safety effort follows. Figure 18-4 demonstrates that caregivers having participated in patient safety rounds viewed the patient safety environment much more positively than those who had not.7
But, is a positive safety culture actually associated with improved safety? The assumption is yes, but data were hard to come by until recently. In Michigan, a multihospital collaborative was initiated (the Keystone Project), the objective of which was to implement evidence-based practices known to decrease the incidence of bloodstream infections (BSIs).8 A total of 107 hospitals were involved, and caregivers responded to the Safety Attitudes Questionnaire (SAQ), similar to the AHRQ tool described previously. Results (incidence of BSIs) were correlated with answers to the SAQ. The results are seen in Figure 18-5. There was an important association noted between the best results (percent reduction in BSIs) and the most positive answers to SAQ questions. This is only an association (and subject to the usual caveats about associations vs. cause and effect), but important nonetheless. These results support the underlying hypothesis that when leadership prioritizes safety and implements actions to support safety, caregivers reflect this in their answers to the SAQ and this is associated with improved patient safety.
Figure 18-3. Nonpunitive response to error.
THE EVIDENCE BASE FOR PATIENT SAFETY
As important as it is to lay a strong foundation in safety culture, the energy and enthusiasm of caregivers to provide safe patient care must be rooted in activities that have been found to be effective. Unfortunately, there does not exist at this point a large base of evidence in patient safety, largely because the field is relatively young. The World Health Organization (WHO) convened a group to carefully analyze existing studies and highlight effective strategies with an evidence base.9 These are listed in Table 18-1. Comments about specific evidence-based actions follow.
With regard to the administration of perioperative beta-blockers to prevent postoperative myocardial ischemia, it is very clear that patients already on such medications must be given them postoperatively. However, an initial interest in beta-blocker administration for patients never having received them previously evaporated as the result of the POISE trial.10 This was an international randomized controlled trial focusing on this specific issue, and the result, after analyzing many thousands of patients, was that giving beta-blockers perioperatively to naive patients caused more harm than good. Specifically, treated patients developed more troublesome bradycardia and hypotension than controls, and this resulted in a higher incidence of stroke, obviating the potential benefit of the drug in preventing myocardial ischemia.
Figure 18-4. Comparison of 2007 Agency for Healthcare Research and Quality (AHRQ) data participants versus nonparticipants in patient safety rounds.
Figure 18-5. Teamwork climate across Michigan intensive care units.
Using maximum sterile barriers for central venous pressure (CVP) catheter insertion to prevent BSI may seem obvious, and the implementation of this protocol has resulted, in many studies, in a dramatic fall in the incidence of this complication. This strategy is complemented by the use of antibiotic-impregnated CVP catheter lines and the use of chlorhexidine in the daily maintenance of the insertion site. The use of ultrasound to help guide CVP catheter line insertion is clearly effective.
Prevention of the feared complication of ventilator-associated pneumonia is a very important consideration, since this development has a high fatality rate and is very expensive to treat. There is some evidence that the continuous aspiration of subglottic secretions is important. Our institution has been successful in decreasing the ventilator-associated pneumonia rate dramatically (Fig. 18-6) using the multipronged strategy described in Table 18-2.
Table 18-1 Evidence-based Interventions for Safe Patient Care
Given the paucity of real evidence to achieve what we think of as safe patient care, and because there is an urgent need to act, another strategy has been to develop consensus guidelines. Although such guidelines are not based on randomized trials, there is value in getting the best and most experienced minds together to synthesize what all would agree to be best practices. Trials may come later to support or refute consensus.
A very influential group that develops consensus guidelines is the National Quality Forum (NQF), an organization of a wide variety of experts, consumers, government officials, and corporate directors. The NQF several years ago published its list of “30 Safe Practices” recommended for implementation. Because this chapter is oriented toward surgery, Table 18-3 lists a selection of the 30 Safe Practices germane to the inpatient setting.
Consensus and Accreditation
The NQF works closely with the Joint Commission. When the NQF has reached consensus on a specific safety practice, this is often translated into a Joint Commission requirement for hospital accreditation, in this setting recognized as Joint Commission “patient safety goals.” These goals are more specific than consensus guidelines and have more “bite” to them, in that all hospitals need to fulfill these requirements in order to be accredited. For example, goal 7, “Reduce the risk of health care–associated infections,” stipulates, among other things, that hospitals have specific strategies in place to prevent surgical site infections (SSIs); that they provide regular feedback about SSI rates to caregivers, with follow-up for 30 days; that they discontinue the use of shaving as a method of preoperative hair removal; and so on. Goal 8, “Reduce the risk of patient harm resulting from falls,” stipulates that each hospital implement a specific falls reduction program.
Table 18-2 University of Michigan Protocol for Prevention of Ventilator-Associated Pneumonia (VAP)
Figure 18-6. Ventilator-associated pneumonia – University Hospital intensive care units.
Consensus, Accreditation, and Payment
The Centers for Medicare and Medicaid Services (CMS) has a lot to say about implementation of safety practices in that they pay a large fraction of the nation’s health care bill. As it regards safety in surgery and hospital care, CMS has embarked on two important strategies. The first is the Surgical Care Improvement Project (SCIP), initiated in 2006, whose stated goal was to reduce the incidence of selected surgical complications by 25% by the year 2010. To do this, CMS produced several specific process measures required to be met by hospitals before receiving full payment for services. Private insurers soon followed suit. SCIP measures are listed in Table 18-4. In contrast to the more broadly defined consensus guidelines, agreed-upon SCIP measures had a strong evidence base. Whether this effort will be translated into national improvement in results remains to be determined. The Joint Commission has also integrated certain SCIP measures into its evaluation requirements.
Table 18-3 National Quality Forum Safe Practices