Using Six Sigma to Lower Length of Stay

¶ … Sigma methodology to help lower patient length of stay while simultaneously improving financial and patient health outcomes. This paper first discusses the theoretical basis for this, outlining the Six Sigma methodology. Its history in health care and its use specifically to lower patient length of stay are both covered in the literature review section. The next section looks at the different measures that should be used to guide the Six Sigma process. Measurement is essential to Six Sigma, so it is critical that measures used are relevant to the problem at hand, and that the organization has means by which to measure the variables in question. A number of variables are proposed, and where there are challenges with their measurement some solutions are also offered.

Introduction

One of the challenges for hospital administrators is to improve both the efficiency of their facilities while simultaneously improving patient outcomes. There are certainly instances where this goes hand-in-hand, such as with advances in surgical techniques that lower the intrusiveness of surgeries and shorten post-op stays. But there are other means as well by which efficiency can be improved while improving health outcomes. Knowing that there are increased risks associated with longer hospital stays, such as the risk of infection, reducing the length of those stays would appear to be a means by which hospital administrators can both improve health outcomes while at the same time improve efficiency, throughput and profitability of their facilities. One means of reducing stays is Six Sigma, a concept that emerged in production management at Motorola but has since been adapted to a multitude of contexts, including health care. This paper will examine at the role Six Sigma can play in reducing the length of hospital stays.

What is required of such an initiative is a means by which outcomes can be measured and assessed. This is an important part of the Six Sigma process, which relies on continually reducing error rates. Error rates require having measures against that can be reduced. For the Six Sigma strategy to work, the measures have to be related to the outcomes that are desired from the strategy. Thus, special care and attention must be paid to the development of outcomes that are to be measured.

Literature Review

There are two components to the literature review. The first is an overview of Six Sigma. The Six Sigma philosophy was developed at Motorola and applied to engineering, in particular to the reduction of defects. When Six Sigma was adopted at General Electric, it became more widely known and since then the program has been professionalized and applied to a wide range of organizations. The Six Sigma advocacy website, iSixSigma.com describe it as "the implementation of a measurement-based strategy that focuses on process improvement and variation reduction through the application of Six Sigma improvement projects" (iSixSigma.com, 2016). There are guidelines for the selection of these improvement projects. The more expertise one acquires on Six Sigma and implementing its techniques, the higher one goes on the Six Sigma hierarchy, which is structured as belts in karate, the black belt being the highest.

The point of Six Sigma is to institute "systematic innovation efforts" (Koning et al., 2006). Such efforts help hospitals to be more competitive, for one, and to improve health outcomes. Part of why Six Sigma is said to work is that the organization is always working on projects to improve quality -- Six Sigma motivates managers and reduces organizational complacency with respect to quality. This is one of the best features of using measures as benchmarks for performance -- you always know where you stand and it is easy to motivate organizations to improve.

As Koning (2006) notes, "some operational inefficiencies are associated with the direct medical service delivery process" and this is especially true in organizations where those processes have never been subject to the sort of scientific, engineering-based analysis that is typical of Six Sigma practitioners. Six Sigma seeks to change the way that the organization thinks about service delivery. Some basic assumptions that exist in healthcare that Six Sigma seeks to disavow are that mistakes are unavoidable and that hierarchical chain of command is effective (Koning, 2006). What Six Sigma promotes in the pursuit of a "standard set of solutions to common problems, and a focus on the customer" (Koning, 2006).

As Revere and Black (2003) point out, Six Sigma is actually quite compatible with health care, because in health care there is near zero room for error. In many cases, treatments are routine, as are things like paperwork. There is no particular reason why random defects or errors should exist for most medical functions, yet errors are all too common, and it is organizational culture that allows for this.

The implementation of Six Sigma in the hospital setting has been conducted for several years now, and there are studies that highlight its effectiveness. DelliFraine et al. (2010) found that most studies looking at the effectiveness of Six Sigma in health care did not have statistical analysis, which seems odd given that Six Sigma specifically relies on statistical analysis. This, it could be reasonably argued, is a defect of the people performing these studies, not of the people implementing the Six Sigma, however. Some of the studies have only incomplete data. Heuvel, Does & Verver (2005) found cost savings from the use of Six Sigma in Dutch hospitals and linked these to improved quality, but provided little evidence for the latter.

Studies that specifically link Six Sigma efforts to reduce length of stay show promise. In a study of an ophthalmology clinic in Jordan found that length of stay was reduced by 44% (Mandahawi, et al., 2011). Other evidence has demonstrated that even small Six Sigma projects can yield positive results, making a case that even after an initial round of successful projects has been undertaken, it is still worthwhile to continue with other Six Sigma projects to earn incremental gains toward the program's error-free objective (Heuvel, Does & Bisgaard, 2005). So while there is not as much data as might be expected about the efficacy of Six Sigma in the hospital setting, there is room to believe that it can be effective, and many studies show that it has.

The second component of the literature review seeks to identify the different measures that should be used in the Six Sigma program to help reduce hospital length of stay while simultaneously improving health outcomes. The first step in this is to determine what factors lead to length of stay in hospital visits. Certainly, the type of medical condition is a factor, but this research is focused more on what can be controlled. Evidence from the Centers for Disease Control indicates that greater efficacy of drugs and technology have done the most to reduce the length of stay in hospitals (Frances & Hall, 2007). This makes a case for ensuring that the staff are fully educated on the latest advancements. More likely to be controlled specifically through Six Sigma initiatives is "greater development and coverage in post-acute care alternatives to hospitalization, growth in utilization review programs, increased enrollment in managed care plans, and other cost containment programs" (Frances & Hall, p.1). Some of these factors are related to the pay-for-performance culture in America (McDonald & Kirk, 2013).

Six Sigma initiatives can affect some of these factors, in particular the expanded coverage and development of post-care alternatives, and the greater number of people in managed care plans and facilities. The primary tool that will be used to address these processes is Kano's model to capture customer critical requirements (Trusko et al., 2007). One of the core facets of Six Sigma is the determination of customer behavior and how that in turn influences an organization and its attempts to best serve its customers. In this particular application of the Six Sigma methodology, it could be that some customers are simply unaware of these options. However, using Kano's model can determine this fact as well as the truly relevant reasons for why customers are not using these additional options. Finally, it would also benefit the hospital to incorporate the usage of another tool, just to reinforce and measure the results of Kano's model. In that case, the hospital would likely utilize SIPOC (Suppliers, Inputs, Process, Outputs, and Customers).

There are even simpler means, which can be used as a starting point. Six Sigma works well with processes, because it asks the people working on the process to find inefficiencies in the process and eliminate them. In one study, the use of Six Sigma was applied to a hospital's discharge process, which in turn both reduced the length of hospital stay and allowed the hospital to increase its patient throughput and revenues (Allen et al., 2010).

Measuring Outcomes, Assessment, and Research

There are four areas of focus, and within these a number of different measures that can be applied to this problem. Working with such areas of focus will give the organization the opportunity to measure past and current performance, plus the progress towards attaining future organizational goals (Grigoroudis et al., 2012). The first area is business value, measured by return on investment, cash flow and net profit. These are standard accounting measures that are presently gathered by the hospital's accounting department. They represent some of the basic financial measures that the organization's management wishes to see.

The second area of focus is the value of the customer. Patient satisfaction is one useful metric, as are clinical outcomes. To these, patient length of stay will also be added. It is important to remember that having multiple different measures orients the solutions towards things that benefit all measures, as opposed to sacrificing one measure to improve another. For example, there is an established relationship between clinical outcomes and length of stay, but also there is a link between length of stay and patient satisfaction wherein the less time a patient stays in hospital, the more satisfied they are, other factors being equal.

Other metrics that can be taken into consideration are value of employee metrics such as professional development, and employee retention. There are also learning and growth metrics like overall quality improvement and new service development that can be looked at. These are summarized in Appendix A. But the most important two sets of metrics are the patient-focused metrics and the financial metrics. Six Sigma has a strong customer focus on the premise that reducing errors will improve customer satisfaction, and that there is a link between customer satisfaction and an organization's financial outcomes. Thus, these are going to be the focal points of the measures that are put in place for the Six Sigma program to reduce patient length of stay.

As part of the implementation process, behavioral changes are encouraged. A Six Sigma project typically requires of staff and management that they do something differently, but over time and many projects, the objective is to change the way that the organization thinks about its operations. Staff and management are to constantly seek out means by which they can improve, and Six Sigma contains a specific structure for that inquiry.

At the heart of Six Sigma is that the process has to be linked to the outcome. Thus, behavioral changes need to be linked to the outcomes that the organization is seeking. An example would be changing the discharge process. By eliminating needless steps -- especially ones that require multiple different people -- and streamlining the essential steps, behavior is changed with the net effect being lower length of stay, cost savings and increased revenue for the hospital, and greater patient satisfaction. The organization needs to be cognizant of where links do not exist. For example, speeding up the discharge process does not appear to have much influence on patient health outcomes -- it neither helps nor harms them.

Another one of the strategies that the literature identified was to look at alternatives, such as diverting patients to managed care more frequently, or more quickly. It is then difficult to measure what the effects of this are on patient health, even if the organization can measure the other effects. This is one of the big challenges with measurement. Neiger et al. (2010) identified that having listed metrics will assist organizations to learn what things work and what things do not. Examining measures in a controlled, statistically -- rigorous manner will help management to understand the effects of its changes. Not all changes are going to be successful, but measures are required to help management understand the influences of its changes on the different metrics that it has determined have the most value.

The next is to understand how to measure outcomes. As noted, the financial outcomes are easy because they are already being recorded as part of the financial accounting process. Patient length of stay is something that is normally measured as well. If such statistics are not presently being gathered, they can be, because there is intake and discharge information that is recorded for each patient. The one caveat with this is that if length of stay is being lowered by moving more patients to managed care, or by diverting the toughest patients, that does not truly measure the actual performance of the people within our organization. Six Sigma appreciates diversions and other things that have an effect on outcomes, but ultimately it is about improving processes. Thus, the comparable benchmark for something like length of stay is length of stay for the same type of patient. This requires arguably more refined data collection than many hospitals currently have.

If, for example, the hospital wishes to reduce the time for discharge processing, time stamps at the beginning and end of this process will be required, and must be implemented before the changes are made. How this would work is that there are time stamps on the initiation of the process and its completion. The different steps of the process can be measured in order to find out the areas where improvements are possible, by identifying steps in the process that seem to be the most inefficient, or the ones that take the most time. Then, ideas are generated with respect to performance improvements. After implementation, the same measurement process is undertaken in order to provide a clearer picture of whether or not the changes were successful. Sometimes what occurs is that a change improves one area, but has a negative consequence on another, so it is important to measure both the parts and the whole to understand the effect of the change.

One of the bigger challenges will be to measure health outcomes. When academics study health outcomes, they will look at things like readmission rates, but when patients are diverted to other forms of care, it can be more difficult to learn about what happens to them when they leave the facility. Measurement of health outcomes is therefore a greater challenges. It requires creativity to find measures that accurately reflect what happens to someone after they leave the system. Usually, manufacturing firms using Six Sigma look at defect rates before the product leaves, and that might be the best approach that is available. Return rates are the same thing as readmission rates, however, so that might be one of the easier ways to measure the effectiveness of the changes on patient health.