Connecting the Dots: Improving Unit Safety Culture to Stop HAI (October 8, 2013)

Webinar Transcript

Paul Tedrick
American Hospital Association - Chicago
October National Content Call
October 8, 2013
11:00 AM Central Time

Operator: The following is a recording for Paul Tedrick with the American Hospital Association of Chicago on Tuesday, October 8, 2013 at 11:00AM Central Time. This is the October National Content Call. Excuse me, everyone. We now have our speakers in conference. Please be aware that each of your lines is in a "listen only" mode. At the conclusion of the presentation, we will open the floor for questions. At that time, instructions will be given if you would like to ask a question. I would now like to turn the conference over to Mr. Paul Tedrick. Sir, please begin.

Paul Tedrick: Thank you very much. I would like to thank everyone for participating today in the October National Content Call, "Connecting the Dots: Improving Unit Safety Culture to Stop HAI." If you have not received a copy of today's presentation, you can find them by going on to www.onthecusp - C-U-S-P - calls. You can also send me an email at P, as in Paul, Tedrick; that's Ted and Rick put together, T-E-D-R-I-C-K at Today's presenter is Dr. Katherine Jones, Associate Professor at the Department of Physical Therapy Education at the School of Allied Health at the University of Nebraska Medical Center. Without further adieu, it is my absolute pleasure to introduce Dr. Jones to you.

Dr. Katharine Jones: Thank you, Paul. Welcome to this webinar, which is intended to help you improve unit safety culture to decrease hospital-acquired infections. And I'm a Health Services researcher and physical therapist and I focus primarily on the role of eamwork. As you can probably tell, I'm a bit under the weather, so I would really encourage you to ask questions at the appropriate time for clarification and, if I'm not coming across clearly, please let the Operator know.

So, moving on to slide two, I just wanted to share with everybody that my work interpreting results from the hospital survey on patient safety culture has been supported by federal, state, and local organizations.

Slide three - In the next 45 minutes, we will define safety culture and explore theoretical frameworks that can help you to interpret results so that you can prioritize change. We will also explore the relationship between safety culture and patient safety interventions so that you can appreciate the interaction that exists between safety culture and the specific patient safety interventions that you want to implement to decrease healthcare-acquired infections.

Slide four - To fully understand the concept of safety culture, we will define it, explore the role of culture in general in organizations, and then break down safety culture into categories, levels, and specific components. The more that you can categorize something, it's easier for us, as human beings, to see patterns and understand how to improve things.

Slide five - The concept of safety culture tend to be defined as enduring, shared, and learned beliefs and behaviors that reflect your organization's willingness to learn from its errors, its experience. If you think about it, all culture, whether it's our national beliefs and behaviors, the culture of our specific ancestry, whether its culture in an organization; all of it is learned by what we see and hear others do, and that's the same in regards to the organizations that we work in and the same in regards to safety culture. You can use four beliefs that are found in a safe, informed culture to engage frontline workers about healthcare-acquired infections. For example, you can ask: Do we believe that your processes are designed to prevent HAI? Do we believe that we are committed to detect and learn from every HAI? Do we believe that we discipline behaviors that increase the risk of HAI? So, the behaviors that increase the risk, not just the outcome of HAI. Do we believe that working in teams will result in fewer healthcare-acquired infections? So, these are some questions that you can pose that reflect those four beliefs that are present in a safe, informed culture.

Slide six - From an industrial organizational psychology and management perspective, organizational culture, in general, is made up of the enduring, shared, and learned beliefs and behaviors that allow us to make sense of what's going on in the environment around us. If, when you start a new job, maybe you just transferred to a new unit or you start working a new organization, that sense of lack of familiarity, of being uncomfortable, or just being unsure of yourself for the first six months to a year comes from not understanding the culture. The culture reflects the common language. That's why I'm such a proponent of the TeamSTEPPS curriculum because it gives us a common language to talk about behavior. The NICK MERCK taxonomy for categorizing errors gives us a common language for talking about error. What's important to understand is that it is leaders that create and teach culture, based on how they share information, how they reward behavior, what behaviors they provide feedback about, and most importantly, leaders create and teach culture by how and when they hold people accountable for what they're supposed to do. Safety culture then is a specific subset of an organization's culture that reflects its willingness to learn from experience. Safety culture can be considered a cross-cutting contextual factor that moderates the effectiveness of patient safety intervention. So, a positive safety culture makes the work of implementing interventions to stop HAI's easier and more likely to lead to success, while a less positive safety culture makes us work harder and less likely to be successful. Think of safety culture like the soil that you plant your garden in. If you can pay attention to the quality of the safety culture foundation in which you are implementing your stop HAI interventions, it pays attention to address the foundation before you go forward with the specific interventions.

Slide seven - So, as an example about the - to illustrate the cross-cutting contextual role of safety culture, recent research supports this concept that safety culture creates the foundation for our patient safety interventions. And data from 179 hospitals, Martin and colleagues found that higher HSOPS scores as measured by the composite average on the X axis, these scores were correlated with a lower composite measure of eight patient safety indicators as measured on the Y axis. So, there's a negative correlation between safety culture and patient safety incidents. The more positive the safety culture, the fewer the patient safety indicators. So, this finding in this survey validates the fact that patient safety culture is one indicator of patient safety.

Moving on to slide eight, then similarly reflecting this overarching pervasiveness of safety culture, patients here, what we see in what we do. Thora and colleagues found a positive relationship between the comparative average and the consumer assessment of health plans survey. So, the CAHPS survey of patients' satisfaction. Hospitals with a more positive safety culture are likely to have more positive patient satisfaction scores.

So, moving on to slide nine - This work by Edgar Shine I found to be just very helpful. And Edgar Shine indicates that we can think about culture in four distinct categories. There's a macro culture of our nation, which right now, our macro culture is that our politics is not working. There are the cultures that we have within all of our organizations, our organizational culture. Each set of professions and disciplines have its own separate sub-culture, and then within our organizations we have micro culture. And it is these micro cultures of units and departments that are what we need to use as the unit of analysis when we conduct safety culture surveys. And the reason for this is that the strongest influence on culture is the local leader; your unit director, your shift supervisor, the person to whom the frontline worker is reporting, is the biggest determinate of safety culture. And this is why the resolved in this project are reported by unit and typically by ICU versus non-ICU and by nurse versus non-nurse within the unit.

So, slide ten - Once we recognize that culture varies at the unit level and by profession, we can think about the assumptions and the values that each of us have that give rise to our behaviors. Our behaviors are supposed to reflect our underlying assumptions and our values. So, let's take a look at an example. Let's assume that our underlying assumption is that safety is a system property where someone educated in regards to safety systems and the nature of human error. So, we believe that safety is a system property. This assumption then gives rise to the value of teamwork. We recognize that valuable human beings make fewer errors when they work in teams. And then the behavior that's consistent with that assumption and that value of teamwork is that we would participate in rounds to assess, say, catheter, urinary catheter appropriateness. Let's take the opposite approach. Let's say that we're more hierarchical and we haven't had any training in safety systems and we assume that safety is a result of individual competency. We value our own autonomy, and so, therefore, our behavior is that we do not participate in rounds to assess catheter appropriateness. If there are any questions or concerns about that example, please jot them down and let's discuss it at the end. Otherwise, if this is an example that sort of rings true with you, I'd also appreciate that comment as well.

So, moving on to slide 11 - James Reason is an industrial organizational psychologist that's taught us a huge amount about error. And his book, "Managing the Risk of Organizational Accidents," if it's not something that you've looked at, I really encourage you to take a look at it. And in that book, James Reason lays out for us that there are really four components to a culture of safety. We can think about reporting, just culture, flexible culture, and learning culture. These four components represent four areas of beliefs and behaviors that interact to produce an organization that is informed about risk and hazards, and all of safety culture is ultimately related to: How well can the people that are charged with minimizing risk and hazards cap in to the knowledge of frontline workers? It's all about engaging the frontline so that they report and tell you what they know and you, who are responsible for implementing change, can make use of what they know. There's an article by Sarah Singer that I'm going to reference later on about engaging frontline expertise that was just published in the Joint Commission Journal that I really also encourage you to tap into. But reasons for components of safety culture are all about how we leverage frontline information by reporting, that people feel free to report their errors and near misses. Of course, they won't report if we punish them for telling us what's not working in their system. So, James Reason also introduced this concept of just culture. What he described as an atmosphere of trust in which people are encouraged, even rewarded, for providing essential safety-related information. But there is also a very clear line that is drawn for people between what is acceptable and what is unacceptable behavior. So, many of us have been involved in David Marks' work on just culture and ultimately the theoretical foundation for what David Marks is doing and all of his algorithms are based in James Reason's original work. Then finally, if people report and we don't punish them for reporting, the organization has to be flexible enough to change based on that information. So, a flexible culture is one that adopts team behaviors, such as those taught in the TeamSTEPPS curriculum, and these behaviors increase the likelihood that frontline workers share information, that we can manage changing workloads, and that the organization as a whole is able to adapt and learn from experience. Teamwork is really the essence of what enables an organization to be flexible and adaptable. Reporting and just culture and being flexible and adaptable is all about learning. Learning means not just that you know what's not working, but that you have the will to do something about it. So, a learning culture correctly interprets information from its safety systems and has the will to implement change. An informed culture is more likely to be a safe culture. And informed, safe cultures are what it takes to be an HRO or a high reliability organization.

So, moving on to slide 12 - Just an acknowledgment that we cannot change what we do not measure and that high reliability organizations do engage in this continuous cycle of measurement, action planning, and implementation of change.

So, moving on to slide 13 then - Safety culture can be measured both qualitatively and quantitatively. Qualitatively refers to focus groups and interviews, but it can also refer to the open-ended comments that we collect on the hospital survey on safety culture. We are going to focus obviously most of our attention on the quantitative tool, the hospital survey on patient safety culture.

I'm going to stop just for a second and grab a drink of water.


Sorry for that. Moving on to slide 14 then - When you conduct the hospital survey on patient safety culture, you are really achieving six different goals. You are identifying areas of culture in need of improvement. You are increasing awareness of patient safety concepts. Conducting the survey in and of itself is a patient safety intervention. You are raising expectations that you will take action, based on the information that you receive from people on the survey. The third goal that you're achieving is that you are evaluating the effectiveness of patient safety interventions over time. You're conducting internal benchmarking by looking at variations by job titles or professions and variations by units or departments. You're conducting external benchmarking when you compare your results to aggregate data at the state level or at the national level. You're meeting regulatory requirements at the Joint Commission requirement to evaluate safety culture. And consistent with what we've learned from Edgar Shine's work, you are identifying gaps between beliefs and observed behaviors within those professional subcultures and micro cultures.

Moving on to slide 15 then - You should measure safety culture as a baseline prior to implementing patient safety interventions that you want to monitor. And then you should monitor periodically, just as to a way to keep a pulse to assess change over time. I don't advocate really effecting safety culture really any oftener than every 18 to 24 months, because it is something that changes slowly over time. Behavior changes slowly.

Moving on to slide 16 then - Just a quick reminder that the hospital survey on patient safety culture does consist of 42 items that are categorized into 12 composites or dimensions. Most importantly, I want you to recognize that because safety culture occurs primarily at the unit or department level, nine of the dimensions of the – are measuring culture at the unit level, while just three measure a culture at the hospital as a whole level. But there are two outcome measures and these are also at the unit or department level. Patient safety grade and number of events reported are considered outcome measures, but these, again, are at the department or unit level. And then pay attention to the comments that you receive. We have developed a coding hierarchy that we use to code the open-ended comments, and we see that these comments track very closely with the quantitative results. What you must do at a minimum with your comments is at least acknowledge that you have heard and are addressing any specific patient safety concerns. People will tell you very specific patient safety concerns that they have and you need to acknowledge those. Whether or not you can address those directly at that time may be a different question, but you have to acknowledge that they've been heard.

Moving on to slide 17 - When we interpret the results then, we want to start by cross-linking the different dimensions of the survey to Reason's components of safety culture. Then we want to identify unit-wide areas in need of improvement. We'll move on to looking at how we can use external benchmarks to state averages and the national database ISTU; how you can use internal benchmarking. We have to stop before we look at the item level results. We have to understand reverse worded items, and then we have to take a look at the beliefs and behaviors and really tear apart each composite and look at the beliefs and the behaviors. And then sometimes what we're looking at is maybe not beliefs versus behaviors, but less structured behaviors as compared to more structured behaviors.

So, moving on to slide 18 - We can see that there are two HSOPS dimensions that cross-walk to regions reporting culture. Now, why am I asking you to link the dimensions of this survey to Reason's four key components? It's pretty simple. It's much easier to think about four big areas that we need to address; reporting culture, just culture, teamwork culture, and learning culture, as opposed to 12 different dimensions. And these dimensions then, you need to think about them as giving you information about one of the four key components of safety culture. With just culture then, we just have the non-punitive response to error dimension.

Moving to slide 19 - We have five dimensions that reflect flexible or teamwork culture. These dimensions measure the multiple aspect of teamwork. Perceptions of staffing are influenced by the effectiveness of teamwork on a unit. There are six dimensions that reflect learning culture. The attitudes and behaviors of unit managers are included in learning culture because whether or not people are able to learn from experience depends on how unit leaders and supervisors, managers, use information and provide feedback to frontline workers.

So, moving to slide 20 then - There are three kind of rules of thumb that we can use to identify unit-wide areas in need of improvement. If there are dimensions that are below the state or national average, if there are dimensions that are less than 75 percent positive, and when you see large gaps between beliefs and behaviors or between less structured behaviors and more structured behaviors within the composites, then these are areas that you should probably pay attention to to begin with.

So, moving to slide 21 then, let's take a look at how external benchmarking can help to identify strengths in areas in need of improvement. In this graph then, you can see that the 12 dimensions of the survey are on the X axis, and I'll just ask you to look at the nine dimensions that are to the left from overall perceptions of safety to staffing. Those nine dimensions are the dimensions that are measured at the department level. So, see the variability there. There's much more variability between a unit's results and the benchmarks. And then look at how all the points come together for the last three dimensions; hospital management support for patient safety, teamwork cross hospital units, and hospital hand-off and transitions. And you can see then, these are the three dimensions that are measured at the hospital as a whole level, and these tend to converge because these averages do tend to converge. So, you can see then for your unit, let's say, I've given you templates that allow you to create this graph. Once you open up the Excel template, you can put your unit level results into the appropriate column and then your unit's results will appear as a blue line. So, you can see then that this graph obviously shows you hat event reporting is clearly a strength for this unit. Some areas of improvement might need to be - you can see where things are below the orange line, which is the ICU state average, and the black line, which is the 2012 national database for ICU's, for reporting ICU's. So, the supervisor/manager actions for your unit are below that. Also, your unit is below teamwork within units. So, those would be two concerns as well as non-punitive response to error. Now, remember what I was saying about unit leaders setting the tone for whether or not frontline workers feel free to report, for whether or not frontline workers feel that they're learning from information. So, I see a relationship between the fact that people in this unit feel that perhaps just culture could use some attention, that perhaps people maybe do feel like they are being reported as opposed to the information being reported. Who makes them feel that way? Somebody does. And it's likely that may be the frontline supervisor or the manager. It's interesting to see here that despite feeling like the atmosphere may be punitive, this sample unit does feel like they are likely to report their events as well.

So, moving to slide 22 then, this shows you an example of the internal benchmarking. So, in this project, they are benchmarking internally nurse results versus non-nurse results. So, you need to take into account the fact that non-nurse results may include the ward clerk as well as any physicians that are operating on your unit. At any rate, what you can see then is where the blue and the orange line diverge, those are places where nurses and non-nurses do not have similar opinions. The need to, as far as supervisors and managers at the hospital as a whole for senior leaders to really engage frontline workers, you can see that gap between what nurses think about hospital management as a whole and what non-nurses think. Also, there is some divergence in organizational learning, communication openness, and feedback in communication about error. So, nurses and non-nurses do not have similar opinions about things, and this is helpful to figure out what is it that nurses know that non-nurses don't know and vice versa.

Moving to slide 23 - Before we dive into the item level results, we really have to make sure that everybody is on the same page in regards to the reverse worded items. The score that's reported in the  is the percent positive, and this percent positive score is the percentage of responses that are rated a 4 or a 5 and agree or strongly agree or, if it's a frequency question, it's most of the time or always. That's for positively worded items. If an item is reverse worded, then the positive thing will be to disagree or strongly disagree or, say, rarely or never. And the  has labeled their reverse worded items with an "R." So, for example, the item A14R is reverse worded. We work in crisis mode trying to do too much too quickly. Well, you would want people to respond positively to that. You would want them to disagree. So, the percent positive score is the percent that disagreed or strongly disagreed with that item. Eight of the 12 composites or dimensions in the survey have at least one reverse worded item in them. In two of the composites, all of the items are reverse worded; that's hand-off in transitions and non-punitive response to error.

Moving to slide 24 then, let's take a look at what the survey tells us about reporting culture. So, here, for example, we're looking at the first item. When a mistake is made, but it's caught and corrected before affecting the patient, how often is this reported? Sixty percent are telling us most of the time or always that this is reported.

Moving to slide 25 then, let's think about some things that we can do that improve reporting. The most important thing that we do to improve reporting is that the frontline worker feels as if they get feedback about what happens to event report. So often on the survey, we'll get responses from people that say, "I report something and no one ever hears about. All reports just go into a black hole." But successful reporting systems are responsive. They are timely. They focus on systems and not people, and they analyze and engage frontline workers in action planning to make improvements. We use taxonomies when we report events. The NICK MERCK taxonomy is especially helpful in reporting adverse events where we're interested in focusing on what's the difference between a near miss, an error that reaches the patient that does not result in harm, and errors that do result in harm. So, ways to improve reporting, you can do it on the front end by making sure that people get feedback about the report, excuse me, you can do it on the front end by making it easy to report, and you can do it on the back end by closing a loop and making sure people get feedback about reporting. That's why leadership walk rounds, leverage frontline expertise, is a robust form of leadership walk rounds. All of these activities that provide feedback also improve reporting.

Moving to slide 26 then, about just culture, I want to just point out that there's three items in this dimension and the item that is always the least positive is this item; staff worried that mistakes they make are kept in the personnel files. In this example, only 9 percent of people agreed - excuse me, disagreed or strongly disagreed with this item. You can see that it is R3, so it's reverse worded.

Obviously this is not the time or place to, on slide 27, I just touch on a few of the things that you've got to think about when you're implementing policies and procedures to support just culture, but understanding human error and understanding of human factors and how human beings interact with systems. David Marks' approach to behavior, whether it's human error, negligence, or reckless behavior and decision-making based on outcomes versus behaviors. We want to make our decisions on discipline based on behavior and not outcomes. The most important thing is that all managers and leaders share the same knowledge about what it takes to implement the just culture, and they use the same decision-making process when it comes to deciding whether or not to hold an individual accountable or whether you're going to look to assist them. So, some type of algorithm, such as David Marks' work or even at a very basic level, the James Reason's unsafe algorithm, must be used by all managers, whether it's the manager in Environmental Services or the manager of ICU. Everybody has to be on the same page as to how they decide whether it's an individual or the system that you're going to look toward.

Slide 28, I've given you an example of James Reason's unsafe act algorithm. If you have questions about that, we can take a look at that at the end.

Slide 29 - We'll take a look then at some of the teamwork items. And this is where I want to show you an example of a gap between beliefs and behavior. Item 1 in this dimension is people support one another in this department. What does that mean? You listen to me and you feel sorry for me because I'm not feeling well today. That's great. But what I really need you to do is to pitch in and maybe help me out with a few tasks. And so item four, when one area in this department gets really busy, others help out; that's the behavior. And you can see that there is a gap between believing that people support one another, 88 percent, and helping each other out when it gets busy, which is only 57 percent. So, we need tools to bridge those gaps. And those tools can be found in the TeamSTEPPS curriculum. When you conduct a brief huddle or de-brief so that people can share the product of their situation monitoring and I can say, you know, "I'm really not feeling well today. Could somebody help me out by helping me admit this patient." We can share that information and ask for help and seek help in those brief huddles and de-briefs.

Slide 30 is talking about communication openness. And I find this slide, this dimension, communication openness, to be very informative and true to what we know about human nature. If you ask somebody, "Will you speak up if you see something that negatively affects patient care?" Well, that is really seeking a socially desirable response. Of course, we will, and I bet you'd like to see that higher than 78 percent. But the reality is that quite often if we're going to speak up about something that may negatively affect patient care, we often have to speak up to somebody with more authority. So, item two asks, "Do staff feel free to question the decisions or actions of those with more authority?" And you can see the 28 percent gap there between 78 percent, "Yeah, we'll speak up about something that negatively affects patient care," and 50 percent will speak up to those with more authority. This item gets at how psychologically safe people feel free to voice their concerns. This concept of psychological safety is huge and it stems from the reactions that we get when we speak up. If somebody minimizes our concern or they flat out ridicule what we said or we're told that that information is not helpful, we're not going to speak up again. And it is leaders and managers that typically establish the sense of psychological safety that we have. So, again, we have to have tools that bridge that gap between our belief that we'll speak up and our behavior that we'll speak up to those with more authority. And, again, the TeamSTEPPS curriculum gives us tools, such as advocacy and assertion, the CUSP tool; I'm concerned, I'm uncomfortable, and the two challenge rule.

Moving on to slide 31, we've got hand-off in transition and we have tools from the TeamSTEPPS curriculum as well, but structure communication and improve our ability to structure hand-offs in transition.

Slide 32 summarizes for you the key components of the TeamSTEPPS curriculum, which I feel is the key tool to improving the component of flexible [skips out 00:39:37] to go the AHRQ website and find out more about it.

Slide 33 and 34 give you examples of tools that you can use; the CUSP tool of graded assertiveness, and the SBAR tool, and these are specific examples around urinary catheters and you can take a look at those.

Moving on to slide 35 is results from the survey about learning culture. And these really reflect supervisors and managers then. This gives us a window into what that unit manager - how they function. And you can see that about two-thirds of people feel like a supervisor/manager says a good word and gives feedback. Seventy-one percent disagree that my supervisor/manager overlooks patient safety problems that happen over and over. Well, remember our rule of thumb; 71 percent, hmmm. You know, then that means that there's almost 30 percent of people that believe that the supervisor/manager is overlooking patient safety problems over and over. I would want to know a little bit more about what that's about.

Slide number 36 - This is, again, a slide where the pattern that I see is actually the opposite of what we typically see, so that the largest percent positive is we are given feedback about changes put into place based on the event report. Sixty-two percent positive, you'd like to see that higher, but what we typically see is we are informed about errors that happen in this department. That's typically much higher than we are given feedback about changes put into place based on event report. So, I urge you to take a look at that in your own results and look at that pattern, because usually people are told about errors more often than they are given feedback about changes put into place based on event reports. So, take a look at that and see what your pattern is. But think about, again, event reporting as a form of close loop communication and you will dry up reporting if you do not give people feedback about what happens with event reports, and if you do not engage them in the process of understanding those event reports.

Slide 37 then does hold true to the pattern that we typically see. I consider item number two the gold standard of a high reliability organization. Mistakes have led to positive changes here. So, what that means is, first of all, people know mistakes have happened and then they know what positive change has happened as a result of that mistake. And you can see item number one, 87 percent of people say, "Hey, yeah. We're actively doing things to improve patient safety, but have mistakes actually led to positive changes?" Sixty-three percent is a little bit less. And then there is this 91 percent that indicates that after we make changes to improve patient safety, we evaluate their effectiveness. So, there's a strong culture of evaluation in this unit, which is consistent with what we saw on the previous item.

Item 38 then, I want to just - these are some tools that support learning culture; a robust culture of root cause analysis at the individual event level and the aggregate level. Safety briefings, leveraging frontline expertise, leadership walk rounds; all of these things are really ways in which we are engaging the frontline and understanding our events and deciding what to do about them. The article that I would refer you to on leveraging frontline expertise is by Sarah Singer and colleagues and it is in the August, 2013 Joint Commission Journal. And she describes, Sarah Singer describes, in this article this cycle of leveraging frontline expertise where you begin by gathering information and you do that through your reporting systems, you do that to the leadership briefings and walk rounds, but this is much, much more than just standing around and talking. It's more than management walking around. You also conduct forums, you track events in a database, and you engage frontline workers. It's a two-way flow of information. What Sarah Singer found was that when managers took in this information and then they decided what to do without frontline improvement, there was a backlash, and people thought that culture was actually worse. So, the four depth of this leveraging frontline expertise include information gathering, prioritizing what you're going to do, following up, and then providing feedback communication. And if you think about that, those four general cycles are involved in everything that we do to improve learning in our organizations, whether it's root cause analysis, safety briefings. It's all about closing that loop with the frontline.

So, moving on to slide 39 and trying to wrap up here so we have time for questions, I want you to understand the relationship between On the CUSP, Stop CAUTI, or On the CUSP Stop all of HAIs. What's the relationship between this patient safety intervention and safety culture as a whole? And they are interrelated. They affect each other. If you look at the cog, safety culture, you can see that safety culture is affected by leadership and, in turn, safety culture than affects whether or not your patient safety interventions in Stop HAI are going to be effective. So, it makes sense then to pay attention to how leaders interact with frontline and drive culture before and during your patient safety interventions for Stop HAI. So then, we can take a look at leaders and what your role is in transformational change, because you can see that leaders are driving safety culture, and safety culture is driving the effectiveness of the patient safety interventions. And this comes from Edgar Shine's work on slide 40 that leaders have to create that compelling positive vision. You don't go out and say, "We're going to change the culture." You go out and set the goal as a performance problem. This is behaviors that we are or are not doing. You ensure that the new behaviors that you want to see are formally taught. You reinforce those behaviors. You walk the walk, you talk the talk; you're out there engaging in rounds and providing data and interpreting data and championing the efforts. You provide opportunities for people to practice. You coach; you provide feedback. You are a positive role model. You create the structures that make it possible for people to think and work and behave in a way that will allow them to stop HAIs.

So, in summary then, safety culture, on slide 41, is our beliefs, our behaviors that reflect whether our organization learns from its experience. Safety culture is associated with the frequency of adverse events and patient satisfaction. You can use different theoretical frameworks to ensure that the results, such as Reason's four key components of safety culture, Edgar Shine's concept of, "We have gaps between beliefs and behaviors because we are frail human beings." His concept that culture varies by profession, by unit, and this idea that leadership must drive culture.

To action plan then, on slide 42, I want you to recognize that there are specific behaviors within each of the four key components. I've given you an inventory of safe practices that identify those safe practices within each of the four key components that you can take a look at that inventory. If there are practices on there that you're not clear about or that you want references for, the references are listed and you can contact me as well. And then finally, what is the relationship between Stop HAI and safety culture? They are interrelated. We must pay attention to your safety culture in order for your interventions, such as CUSP, Stop CAUTI, to be effective.

So, with that, I will stop and take time for questions. And I, again, thank you for being patient with my - I'm not quite myself today, so thank you for your patience and understanding.

Operator: Thank you.

Paul Tedrick: (overlapping) Operator, you can begin with the Q &A session by giving everyone instructions.

Operator: Thank you, ma'am. At this time, we will open the floor for questions. If you would like to ask a question, please press the "star" key followed by the "one" key. That is "star one" on your touchtone phone now. Questions will be taken in the order in which they are received. If at any time you would like to remove yourself from the questioning queue, please press "star two." Again, that is "star one" for questions. Our first question comes from Mary from California Department of Public Health.

Mary: Hi, Dr. Jones. Thank you so much for that very informative lecture. I have a few questions and I won't monopolize the questioning period, but I was wondering what do you consider a significant gap, a percentage gap, between belief and behavior? Or do you have something that you could use to, supply us a guideline for a significant gap, like 10 percent, 25 percent?

Dr. Katharine Jones: As a rule of thumb with the HSOPS is anything that's 5 percent or greater is considered significant for a change. When I'm looking at gaps between belief and behaviors, those usually jump out at you and those gaps greater than 10 percent, even 15 percent is - you usually see things - I've seen the gap between - on communication openness, the gaps between - people will freely speak up if they see something that negatively affects patient care. I've seen that as high as 85 percent, and then people feel free to speak up to those with more authority as low as 15 percent in the same unit. So, the gaps between beliefs and behaviors tend to be quite large, and so I'm going to say at least greater than 10 percent.

Mary: Thank you.

Operator: Thank you. Again, if you would like to ask a question, please press "star one" at this time. We are now holding for questions. We have a question from Kelly from Brookswood Rehabilitation Hospital.

Kelly: Yeah, hi. I missed the information on where we get the slides from in the beginning, so I was wondering if you could repeat that, because I don't have the slides and this is very good.

Paul Tedrick: Hi. This is Paul with HRET. You can find them in two places. The first place is you can just go directly to the website. It's www.onthecusp - C-U-S-P - (one word) contentcalls. And if you still are having trouble finding it, you can also send me an email at ptedrick - T-E-D-R-I-C-K - at

Kelly: Thank you.

Dr. Katharine Jones: (overlapping) Or – this is Katherine. I know that the slides that you've posted don't have my updated notes on them, so I can send you that updated. It's just that a lot of what I said is written on the notes section of the slides and I don't know if you're posting the PDF or if you're posting the slides.

Paul Tedrick: What is up there currently was the most recent addition. There was a problem with it the last time, but that one is actually up there this morning, the most current one.

Dr. Katharine Jones: Well, it's not the most current one -

Paul Tedrick: And it's a peg form.

Dr. Katharine Jones: (overlapping) I'm telling you, because, I mean, the slides are, but I'm telling you that a lot of what I said is in the notes section and - let me just say that I'm going to send you the slide again, because exactly what I said today is in the notes section on these slides. And if others want them, they can also contact me.

Paul Tedrick: Sounds great.

Operator: Thank you. Again, if you'd like to ask a question, it is "star one." We do have another question from Mary from the California Department of Public Health.

Mary: Hi. I would like some clarification. In the dimension, non-punitive response to error. This is like a reverse worded item, and I am wondering is it true that a low score here is better than a high score? Or am I misinterpreting that?

Dr. Katharine Jones: So, the way to think about the reverse worded items is that in the HSOPS you always want a high score. It is the percent positive that's reported. So, with reverse worded items, it is the percent that disagree. So, read the item as it's stated and, again, the non-punitive response to error is one of the two dimensions in which all of the items are reverse worded. You always want a higher score. So, if we take an example, when an event is reported, it feels like the person is being written up, not the problem. Would you want people to agree or disagree with that statement?

Mary: Disagree.

Dr. Katharine Jones: Disagree, exactly. So, the percent positive is what's reported. In our example, 36 percent of people disagreed with that statement. So, that's really only a third that are disagreeing. Now, we'd have to look at the survey results to find out if the other two-thirds – typically there's about one-third that are agreeing, "Hey, yeah. I feel like it's me being written up, not the problem." They are agreeing that that's the case or there's a percentage of people that are afraid to have an opinion or don't have an opinion. The bottom line is only about a third of people feel like that's not true, that they disagree. So, with the survey, positive is always positive. It's the percent that think that things are good around here. So, ask yourself: What's the positive response to those reverse worded problems? And it is to respond, disagree, or rarely never.

Mary: Thank you.

Dr. Katharine Jones: You've got to spend time – take a look at the slide that I provided on that. You've got to spend time with people, because they'll look at that and they'll go, "Oh, only 9 percent of people think that." No, that's not it. Only 9 percent disagree, and that's not a good thing. You always want the score to be higher. And people will have difficulty wrapping their minds around the reverse worded items. And I've had people – do not take the items that are reverse worded and flip them around; that's not good to do, because then the next thing you know you've got your results backwards and nobody can keep track of what's going on and I've had people do that before, too.

Mary: Thank you.

Operator: Thank you. Again, if you would like to ask a question, please press "star one" at this time. We are now holding for questions. At this time, actually, I'm sorry – we do have another question from Mary from the California Department of Public Health.

Mary: Hi. Thank you. I'm looking at the graphs on slides 21 and 22, external and internal benchmarking, and I think it was said in the slides that a score above 75 percent is more favorable and might not call out an area that needs immediate attention. However, in the external benchmarking, as I look across 75 percent, is it true that most of the hospitals really need work in quite a few of these areas? That's what it appears to me - a lot of work to be done, because very few of them are above 75 percent.

Dr. Katharine Jones: The other thing that I would urge you to do is – I'm personally not a fan of using averages of benchmarks. I prefer to use the 10th and the 90th percentile, so you can see what the range is. And the reality is in the HSOPS that the range is rather narrow. So, it depends on whether you're aspiring to be average or whether you're aspiring to be excellent as to whether or not you want to set the 75 percent as a cut-off point for what you pay attention to.

Mary: Thank you.

Operator: Thank you. Again, if you would like to ask a question, please press "star one" at this time.

Dr. Katharine Jones: The last thought I would leave you with is that accountability is key in our health care system, and the pattern that you see, as Mary was talking about the benchmarks, where if you look at those graphs on slides 21 and 22, the pattern you see where non-punitive response to error tends to be the least positive dimension holds true all the time for all the results that I've seen in the seven or eight years that I've been doing this work with the HSOPS. And I think that this is reasonable and it's normal, because people have to have some sort of sense that they're being held accountable, and I think it's very difficult for people sometimes to sort out what is appropriately being held accountable and what's punitive. Human beings have a hard time with that and I think that is one of the challenges of management is to have a really open and transparent conversation about accountability. High quality people do not want to work in a place where people are not held accountable; therefore, you will always have some folks that think this is a punitive environment, and that's my take on the fact that it's not always non-punitive response to error. We need to have a just culture, but it needs to be fair and it needs to always have that stark line drawn between what's acceptable and what's not acceptable. And as long as you're holding people accountable for what's not acceptable, some people will think that's punitive, and that's my opinion on that.

Paul Tedrick: Hi, everyone. Well, I wanted to go ahead and wrap this up. I wanted to thank Dr. Jones for presenting this outstanding information with us today. And before we sign off, I want to remind everyone to please take a moment and fill out an evaluation of today's presentation. Your feedback is extremely important. A lot of our content call presentations are based on feedback we heard from you about the topics that you're interested in, and we continue to develop new content for future presentations all the time. You can find our evaluation as a slide. You'll see the link; it's at the end of the slide presentation. And once I get an updated presentation from Dr. Jones today that has the notes in it, that will be the one that will go up on the website and it should be up there this afternoon. So, once again, I wanted to thank Katherine for her great participation, or her great presentation today. I wanted to thank all of you for your participation and we wish everybody a happy and productive day.

Operator: Thank you. Ladies and gentlemen, that concludes today's conference. You may now disconnect.

# # # #

Return to CAUTI Toolkit Implementation Page

Page last reviewed December 2017
Page originally created November 2015
Internet Citation: Connecting the Dots: Improving Unit Safety Culture to Stop HAI (October 8, 2013). Content last reviewed December 2017. Agency for Healthcare Research and Quality, Rockville, MD.