Management Practices, Work Force Capability, Work Processes, and Organizational Culture
Ada Sue Hinshaw, Ph.D., R.N., F.A.A.N.
It is my pleasure this evening to be able to talk with you about the recommendations that came from the evidence and from the areas we had considered in terms of the blueprint for health-care organizations.
First of all, let me share with you the blueprint itself. Don was the mastermind of this blueprint, and we were all delighted with it because it brings it all together in one place, which is very helpful to everyone. Essentially, we are looking at threats that can arise to patient safety in four areas and strategies that we have looked at and recommendations that we have made in the same four areas: management practices, work force capability, work processes, and organizational culture.
Altogether, there are 18 recommendations, and 10 of them are primarily for the health-care organizations in the country.
Let us look at each of these individually but, first of all, let us take a look at management practices. Visible leadership, an interactive style, and decentralized decision making are all management practices that build strong and positive work environments.
We have seen this in a number of existing health-care management models in the country that have enacted this particular style of interactive leadership, which involves everyone in the organization—and particularly the staff nurses—in the decision making that is made with regard to patient care and staffing. Also, if leadership is not committed to the safety issues and to building the culture of safety, which we will talk more about later, then it is very difficult to improve safety in any way. In all cases, improvements should be made on the basis of evidence-based manage-
ment, which develops strategies by relying on management research on how to form trust, manage change, involve workers, etc.
We have seen in the last several years some real differences and some real changes in hospital nursing leadership, and so it is important to summarize those. To say the least, they are familiar to many of the nursing leaders who are here.
The chief nursing officer has been given expanded responsibilities, but we don’t know all the implications of this. There needs to be research to look at this particular issue. It could be seen as an opportunity for expanded control. It could also mean that there is less attention to be really given to nursing leadership within an organization.
As you look at the evidence of what has happened in the last several years, let me tell you that in one major survey in 1998, which focused on the chief nurse officers at university teaching hospitals, it was found that at 82 percent of the hospitals, the chief nursing officer has had expanded responsibility. In only 24 of those hospitals have we seen a shift in the nursing in their title, and only 24 percent now have nursing in their title. In some university teaching hospitals, there is no longer a separate department of nursing, and in 91 percent there was an obvious decrease in the number of midlevel managers who were available.
So, we not only have some major changes at the top leadership level; we have major changes at the midlevel management area as well.
The concern then is what happens with potential loss for a voice of nursing within those levels of management. Also, what happens in terms of the weakening of clerical leadership? Do we have in fact a number of people, a number of staff nurses particularly, who feel that there are no longer individuals available to them for intellectual and resource support?
Therefore, one of the first sets of recommendations is related to what health-care organizations need to consider in terms of acquiring nurse leaders for all management levels. From the top level of participation on the boards of health-care organizations to decentralized decision making with the staff nurses who are on the units, it is essential that nurses be able to provide consistent input and communication about patient care decisions and necessary health-care resources.
We also looked at five different management practices, and we particularly draw your attention to these because they are ones that are very helpful in terms of increasing safety for patients: first of all, balancing efficiency and reliability; secondly, creating and sustaining trust. This is trust in two directions. In other words, nursing leaders need to trust the competency, the decision-making capability of the individuals who are at the point of care. And individuals who are at the point of care need to know that their nursing leaders are in fact available to them, are concerned
about their resource base, and have the ability to make their decisions move with patients, etc. Actively managing change is a very important part of this. We talked about involving workers and then creating a learning organization, an organization where the people share information; there is open communication; they are able to try new innovative ideas, study those, evaluate them and then have the feedback processes that allow the nurses and the workers in the organization to actually use that information.
With nurse work environments, there is evidence in the past decade and one-half of several areas that we are concerned about. Obviously, we are concerned about the increased emphasis that was on efficiency for a number of years rather than on the balance of efficiency with safety or with quality. There are also some issues of concern related to poor change management, limited nurse involvement, and limited use of knowledge management practices. As you can, there is an ample evidence base suggesting that these are some of the concerns in the work environment of nurses.
The next recommendation focuses on leadership at the top and the importance of educating all of the leaders, the board members, and the managers and about the idea of a culture of safety. We will talk more about that later, but the culture of safety is the link between management practices and safety, so that everyone is involved with the issues and the strategic planning around safety. For example, in a culture of safety, when a board member gets reports on financial affairs of the agency or the health-care organization, they also get a report on safety indicators and the outcome processes for safety.
The next recommendation deals with professional associations and philanthropic organizations, and looks for collaborations that will help health-care organizations to advance their evidence-based management practices. This involves putting together academicians and managers and nurses and the multi-disciplinary kind of team that was evident in our particular team, in order to think through together what we can do with management practices that will increase safety and make for strong and positive environments.
The next area in the blueprint is that of work force capability, and we looked at three different areas. Safe staffing levels are one area. As you know, there are a number of studies now that really help us to understand the relationship between nurse staffing and patient outcomes. It is very clear that when there is inadequate staffing there are negative patient outcomes, and we can talk in terms of an ongoing series of studies, from Kovner’s earlier work in the late nineties through to Aiken’s more recent work that was published in JAMA.
We also know more research must be done in the area of hospital
studies. As we began to look at what we could do in the way of staffing recommendations, we ran into difficulty immediately because the studies, as well as they are done by Linda Aiken and others, are at the aggregated level of all units in a hospital or all nurses in a hospital. From this research, we could not sort out the separate medical-surgical component, and as Bill says, “What is a medical-surgical unit these days?” Research work really does need to be done to help us to better pinpoint what the staffing ratios need to look like or what staffing in general needs to look like in each area of the hospital.
We have better data in nursing home studies. For example, the data that came out of the U.S. Department of Health and Human Services (DHHS) minimum nurse staffing ratios in nursing homes was reported in 2001. Andy Kramer was a major part of that study, and so any questions we will refer to Andy. He has got this at his fingertips.
This is a very important study because it was very extensive, looked at a number of nursing homes and many patients, had very strong staffing data, and then found consistent associations between staffing levels and quality of care.
So, whether we are talking hospitals or we are talking nursing homes, the relationship between staffing and quality of care or staffing and errors in patient safety is very consistent and very strong. Essentially, this study suggests that there are persistent associations. He was even able to show that as you increase the staffing a certain amount you will get a certain increment increase in the outcome for residents, and if you increase the staffing again you will get a certain increment in the outcome for patients. So, it was very predictable in that sense.
There was of course a level at which the staffing was raised and there were no higher outcomes for patients, and so that provides very clear information about optimal levels of staffing.
It is important to note that more than 75 percent of nursing homes studied were below that optimal level. So, we have a long way to go in terms of the staffing issues with this particular area.
We took three different approaches in this area with work force capability, regulatory approaches that we are recommending, several internal staffing practices for the health-care organizations themselves, and a marketplace consumer-driven approach that we really liked.
It is going to be interesting to look at how we might move that particular set as well. First of all, we are recommending that the DHHS should update the 1990 regulations that specify minimum nursing home staffing standards. It was over a decade ago that these particular nursing home staffing standards were established, and at this point those standards called for a registered nurse (RN) in nursing homes only eight hours a day. We all know what has happened to the acuity of patients in the last
decade and one-half. We all know that patients are going home or into nursing homes much more acute than they were, and the complexity of care has grown tremendously, and so it is really important now to look at and we are recommending that we require at least one RN within the facility at all times.
Currently, the standard for staffing levels and numbers is that there is one standard for staffing, and whether there are 60 residents or 300 residents in that facility that staffing standard doesn’t change. That is the reason for the second staffing recommendation that we are making: the staffing levels really need to be changed as the number of patients is increased.
We then looked at staffing levels for nurse assistants who provide the majority of care in nursing home facilities. They are currently carrying about an average of 11 patients each, which is a very high number and does have some real consequences in terms of patient safety.
Let us now look both at hospitals and at nursing homes and some of the recommendations that we are making here. First of all, I don’t know how many of you are familiar with hospitals in the sense of when they take their census, but usually it is at midnight. But during the day patients are admitted. Patients are discharged. Patients are transferred. All of this makes for extreme activity that the nurse must handle, and at each point, particularly transfers, the safety data show a higher risk for patients. We need to become much more aware of how many patients we are actually working with.
One study that looked at a medical-surgical unit found that during the midnight census they had an average of 23 patients, but when they looked at the full cadre of patients it was 35 on an average for that unit, 12 patients higher. You can believe that those patients were the ones being admitted, transferred, and discharged, patients who require high care.
With each of these you can see that we are moving to really look at getting direct care nursing staff much more involved in the decisions around staffing. Bill has a beautiful example he will use to show you that, and so I am not going to look at it much. We are also talking about getting elasticity into the system. It is going to be critical in the future for us to have some ability to have more staff or less staff as those numbers go up and down.
We also talked about direct care staff involvement and empowering nursing staff to regulate unit flow in that sense.
The other recommendation is for hospitals themselves and this is with respect to hospital intensive care units (ICUs). We were only able to make recommendations for the ICUs for the very reason that I cited earlier. In ICUs the research shows very clearly that once you have more than two patients the error rate does climb and complications occur. Take particu-
lar note of the way this particular recommendation is stated because what it says is not that this is minimal staffing or this is optimal staffing. Instead it says, “At any point the staffing gets to this level, you are at higher risk for patients having difficulty or having errors committed with them.” So, if you have an ICU that has one nurse for every two patients, you need to become much more vigilant about what is happening with the error outcomes on that particular unit. It is the same for the nursing home recommendations that we have made.
We also looked at the recommendation for a nationwide system for collecting staffing data and have been particularly cognizant of the fact that it will take some time to get these data streamlined. There is some precedent with Centers for Medicare and Medicaid Services current in the Medicare/Medicaid service data.
There is nursing staffing data and so there is some beginning here, but we are also recommending that one of the things that needs to be built into the forthcoming hospital report card is staffing data because the relationship between nurse staffing and outcomes and safety is so strong. We are also suggesting that more time may be required and should be allowed in order to obtain this staffing data for the hospital report card.
Remember, we are also talking about knowledge capability and acquisition of knowledge, and this is something related to the educational and training recommendations that we are making.
It interested us that in most safety-sensitive industries there is quite a bit of money spent on training. When we look at health-care safety training in relationship to those other industries, we come up short and so it is important to note that data.
We then made a series of recommendations about assisting nursing staff in ongoing acquisition and the maintenance of knowledge and skills. These are not recommendations that you might expect, but they are very important to have made explicit in this report, and are even more important because of the need to enhance knowledge about existing and future technologies.
We next had several recommendations around the interdisciplinary collaboration, and we looked at collaboration broadly. We were asked the question about RNs and doctors this morning on the webcast and during the media event, and I have to tell you that we looked very hard at the different studies. We found that most of the studies showed that over 85 percent of the nurses were reporting that they had very strong positive relationships with their physician colleagues, and so at this point one thing we know is that relationship is very important. It is crucial, but there is no data at this point except for anecdotal data to support the negativity that often surrounds that particular discussion.
Next in the blueprint is the work processes section. You can tell that we looked at two different areas. This particular slide shows you the first of the data concerning fatigue, looking at scheduled and actual shift durations. We go from 1.5 hours through to 22.5 hours in terms of range. You can tell the schedule data in the lighter color and the actual shifts are very different so that that is part of the unpredictability in what is happening with nurses. The 1.5 person, the person working 1.5 hours came in to work labor and delivery. They did not have any patients in labor and delivery, and so this person was sent home. That all makes for a problem in the work environment and a sense of instability on the part of the professional nurse.
So, I think it is something that we have to keep in mind. You can also see that about 84 percent of the nurses worked between 8 and 12 hours consistently.
The 12-hour shifts with limited rest are called sustained operations. That is when you do 3 to 5 of those in a row. The error rates increase rapidly after 12 hours of work, and also as we know fatigue has some very nasty effects on our ability to problem-solve and do critical analysis and to react quickly.
So, we have recommended that states should prohibit nursing staff from providing patient care in excess of 12 hours per day or 60 hours during a 7-day period. This is probably going to be one of the more interesting recommendations in the report, but almost every safety-sensitive industry that we know of and that we investigated has put limits on the number of hours that the individuals involved in that industry can work, everything from transportation to nuclear power plants. The early work that is being done now with fatigue and nursing shows us exactly the same thing in that sense.
Some of the other work processes are inherently dangerous, and we know that because they are both high risk and they are high frequency, which is what makes a difference, and so you can tell very quickly the difference here.
I should tell you—and I think I am being defensive, so I will say that up front—but when you look at medication administration, nurses commit a large number of the medication errors. I should also tell you that we intercept 86 percent of medication errors before they ever get to a patient.
So, there is a good side and a bad side to this, but nurses are fallible like all other health-care professionals. Hand washing: you see the data on that. You can see the inefficient work processes, what happens with documentation, time spent hunting and gathering. We are all aware of that particular phenomenon and also the time that is used while involved in a number of non-nursing activities.
So, we have a recommendation then countering some of that particu-
lar evidence for health-care organizations, and to really help with resources to redesign or to design the nursing work environment with these kinds of work processes in mind.
We are also suggesting that we need a multi-disciplinary group that will look at problems with documentation, because these problems arise from several sources and we need to be able to get a better grasp on streamlining it and deleting some. That last was my own opinion. I will tell you clearly.
The last blueprint area is that of the organizational culture and the culture of safety. We are very committed to the essential elements that go into that culture. Every individual in the organization is going to have to be committed to that culture of safety. It must be a long-term commitment by the organization itself. Some of the literature and the research suggest it takes at least 5 years, if not longer, to get such a culture into place, while maintaining the legally required confidentiality of data.
These are some very specific recommendations that we have offered through the report that have to do with the culture of safety. The next is the National Council of State Boards of Nursing working with other colleagues to really begin to look at how they can discriminate between latent errors, as Don defined them earlier, and errors that are actually human or errors that involve willful negligence or intentional misconduct.
Congress needs to pass legislation so that in fact people can report errors and near misses in errors and feel confident that they will not be either sued or maligned for providing information that will improve patient care.
So, we are back to the major blueprint, and that gives you a series of 18 recommendations that we had worked through.
