Hospitals are increasingly turning to quality improvement techniques such as Lean and Six Sigma as a means for improving hospital outcomes to include responsiveness, patient safety, and cost savings. This heightened focus on quality initiatives may be partially due to the fact that manufacturing has achieved much success with these strategies (Shah, Chandrasekaran, & Linderman, 2008). Another important reason may be that Medicare recently instituted new payment policies whereby hospitals could receive additional payments for exhibiting better quality, safety, and patient experience scores (Ding, 2014). Given these changes, hospitals are even more motivated than before to identify strategies that improve these outcomes. However, research is inconclusive with regards to whether Lean or Six Sigma is the most effective means of improving performance in health care (Field, Heineke, Langabeer, & DelliFraine, 2014) or whether some combination of both Lean and Six Sigma together might provide superior results.
Some have even argued that process improvement techniques borrowed from the manufacturing sector do not translate well in the health care setting (Langabeer, DelliFraine, Heineke, & Abbass, 2009). Health care delivery is unique from manufacturing in that, like other professional service organizations, health care is characterized by high customer contact/ customization, process variability, and high labor intensity (Lewis & Brown, 2012). In this type of complex service environment, professional judgment and employee autonomy take on a much greater role than in the manufacturing setting (Dobrzykowski, McFadden, & Vonderembse, 2016). There are concerns that concepts like reducing waste, streamlining processes, and standardizing procedures can negatively impact clinician autonomy and result in less effective decision-making on behalf of patients (Graban, 2011). Exacerbating the problem is the lack of conclusive empirical evidence establishing the benefits of implementing these techniques in health care. For instance, a recent empirical study found only indirect effects through responsiveness but did not find a direct link between Lean and patient safety or between Six Sigma and patient safety (McFadden, Lee, Gowen, & Sharp, 2014). A recent article that systematically reviewed several literature reviews concluded that there is limited evidence to support a positive link of health care performance improvement with the implementation of Lean, Six Sigma, or a combination of the two practices (Deblois & Lepanto, 2016).
In addition to the mixed empirical results regarding the effectiveness of Lean and Six Sigma in hospitals, there is also limited knowledge about the current level of implementation of Lean and Six Sigma. One study surveyed a national sample of U.S. health care organizations and revealed that only 27% had adopted some form of Six Sigma, and of the organizations that stated they did not implement Six Sigma, 54% said they had no intentions of deploying it in the future (Feng & Manuel, 2008). Another study indicated that 42% of U.S. hospitals currently implement Six Sigma and 53% practice Lean (Henke, 2009). The wide discrepancy of the research findings raises the following questions regarding the actual usage of Lean and Six Sigma in hospitals: (a) What are the existing patterns of Lean and Six Sigma implementation in hospitals, and (b) does Lean, Six Sigma, or a combination of both lead to better hospital performance across different dimensions such as responsiveness capabilities, patient safety outcomes, and cost savings?
To address the above questions, this study empirically explores the patterns of Lean and Six Sigma implementation in U.S. hospitals. Specifically, we use cluster analysis to identify predominant Lean and Six Sigma implementation patterns and draw from dynamic capability theory to explain why Lean and Six Sigma practices should improve hospital performance. This study contributes to the literature by not only identifying the current patterns of Lean and Six Sigma implementation in U.S. hospitals but also providing valuable insight into the organizational characteristics of each cluster. In addition, although hospitals have implemented Lean and Six Sigma together, prior studies have failed to adequately consider the effect that simultaneous implementation of these two techniques has on hospital performance. This study addresses this research gap and offers justification for encouraging hospitals to implement more Lean tools in combination with Six Sigma for better performance outcomes.
Literature Review and Theory
Lean and Six Sigma are two process improvement approaches designed to enhance organizational efficiency and effectiveness. Although there are similarities between these two approaches, they also differ on conceptual models, implementation, benefits, and drawbacks (Andersson, Eriksson, & Torstensson, 2006). As more recent options to continuous quality improvement, Lean and Six Sigma focus on training employees and managers in specific tools and techniques. Lean thinking, however, mainly emphasizes efficiency improvement in speed and waste, in contrast to the Six Sigma focus on effectiveness for process analysis and reduction in variation and errors (Andersson et al., 2006). Although their main philosophies, practices, and tools are further discussed below, our literature review reveals the lack of empirical research, which can provide the evidence for the benefits of those quality endeavors in hospitals. Langabeer et al. (2009) also noted that more than 75% of the research on either Six Sigma or Lean in hospitals were purely subjective, conceptual, or based on a single case.
Dynamic capability theory delivers a basis for the implementation of Lean and Six Sigma. Dynamic capability is defined as the capability of the organization to integrate and reconfigure internal and external resources and competences to respond to rapidly changing environments (Teece, Pisano, & Shuen, 1997). A comprehensive review of the theory suggests expanding applications to enablers of dynamic capabilities (Wilden, Devinney, & Dowling, 2016). By redesigning internal processes and routines, Lean and Six Sigma can enable organizations to improve their abilities to adapt to changing conditions. In a health care environment, Lean and Six Sigma provide effective ways to improve health care processes so that hospitals can meet the ever increasing needs of patients. On the basis of dynamic capability theory, we would therefore expect Lean and Six Sigma implementation to improve health care performance.
Six Sigma in Hospitals
Six Sigma’s systematic data-driven approach aims to resolve process errors by focusing on organizational outcomes most critical to customers. Six Sigma process improvement requires the use of quality teams and project solution systems, which are effective in resolving health care issues. Employees are trained as Black Belt or Green Belt change agents in advanced statistical and team management skills. Critical Six Sigma concepts for hospitals include the DMAIC (Define, Measure, Analyze, Improve, and Control) process, focus on business results, and competitive selection and review of people and projects (Gowen, Stock, & McFadden, 2008). Other Six Sigma tools include Statistical Quality/Process Control, Pareto chart, and Fishbone diagram (McFadden et al., 2014).
Despite its promising roles, the relationship between Six Sigma and hospital performance has not been well established. A review of research on the application of Six Sigma in health care over a 15-year period reveals that, out of more than 300 articles, only seven studies empirically tested for significant effects of Six Sigma, although all of them reported positive outcomes (DelliFraine, Wang, McCaughey, Langabeer, & Erwin, 2013). Another comprehensive review reveals a lack of conclusive evidence for the effectiveness of Six Sigma in health care, with only 28% of studies reporting cost savings and 8% describing revenue improvements (Liberatore, 2013). Furthermore, whereas 67% of the studies reported initial improvements, only 10% had sustained improvement, which suggests weak Six Sigma performance results in health care.
Lean in Hospitals
The implementation of Lean management initiatives is increasingly important among health care organizations. A main motivation for Lean methods has been the relentless pursuit of efficiency, reduced costs, and increased delivery speed of products and services (Graban, 2011). Health care non-value-added is estimated at 95% of health care operations, leaving room for substantial efficiency improvement (Hagan, 2011). Lean methods rely on five principles: (a) identifying customer value, (b) specifying the value stream for each service, (c) mapping the service flow, (d) establishing the customer-driven pull processes, and (e) pursuing perfection (Cottington & Forst, 2010). Common Lean management tools include process mapping, value stream mapping, Kaizen improvement events, just-in-time process management, and “5S” principles (i.e., Sort for necessity, Simplify the workplace, Shine for cleanliness, Standardize processes, Sustain standard processes; Graban, 2011).
A recent case study reveals possible impressive results of Lean, including reduced patient cycle time, waiting time, queue length, scheduled time of staff, and improved patient satisfaction (Jayasinha, 2016). In addition, some case studies take a contextual perspective and identify key factors that influence the outcome of Lean projects in hospitals, including CEO commitment, culture and style of leadership, absorptive capacity of the hospital, alignment of the Lean initiative with the hospital mission, dedicated resources and experts to Lean, staff training before and during projects, measurable project target, proper work load assignment, team autonomy, and communication between project team members and other affected staff in the hospitals (Harrison et al., 2016 ; Hung, Gray, Martinez, Schmittdiel, & Harrison, 2016). However, a comprehensive review of Lean and Six Sigma implementation in acute care hospitals found overall success to be poor to fair, with success primarily for linear event sequenced operations, such as emergency departments, intensive care units, and operating theaters (Deblois & Lepanto, 2016). This study calls for more research to better understand the impact of Lean and Six Sigma implementation in health care.
Co-implementation of Lean and Six Sigma
Although little theory exists about integrating the two approaches, Lean and Six Sigma practices seem to be increasingly implemented at the same time, called Lean Six Sigma. Dynamic capabilities theory provides a clear lens to help explain why one would expect that implementing both initiatives simultaneously should result in improved performance outcomes (Anand, Ward, Tatikonda, & Schilling, 2009). By taking advantage of the synergistic benefits and avoiding drawbacks of each technique alone, Lean Six Sigma can add a dynamic dimension “by institutionalizing organizational learning, manifested in the form of process improvement” (Anand et al., 2009, p. 446). The combined learning resulting from implementing both techniques together is expected to result in greater overall effectiveness.
Experimental studies on Lean and Six Sigma practices in hospitals show positive benefits. Of 124 hospital surgical articles, 11 studies used Lean, 6 applied Six Sigma, and 6 utilized Lean Six Sigma, with six themed outcomes: improved outpatient efficiency and operating theater efficiency, plus decreased mortality, operative complications, waste-based harms, cost, and length of stay (Mason, Nicolay, & Darzi, 2015). Lean Six Sigma applied at eight hospitals to ameliorate 41 causes of hand hygiene noncompliance resulted in improvement from 47% to 81% compliance sustained for the end of the project a year later (Chassin, Mayer, & Nether, 2015). Case studies demonstrate the efficacy of health care Lean Six Sigma programs to greatly reduce costs, medical errors, emergency room times, personnel hiring, start times, and maintenance (Ahmed, Manaf, & Islam, 2013). Overall, the above research implies that co-implementation of Lean and Six Sigma might be beneficial. However, asserting that would be skeptical without knowing the answers for the following questions: (a) Are there a significant number of hospitals that implement both Lean and Six Sigma at the same time? (b) If so, do they outperform other hospitals that mainly execute only Lean, Six Sigma alone, or nothing? In order to explore the current status of Lean and Six Sigma implementation in U.S. hospitals, we conduct a cluster analysis with a moderately large data set.
After obtaining institutional review board approval, we collected data via a survey instrument. The clarity of the set of questions was checked by several Quality and Patient Safety Directors in the local area. We started with a list of U.S. hospitals by using a directory found on the website Hospitallink.com. After eliminating nonhospital organizations, we took a stratified sample of hospitals according to geographic region so our data could accurately represent the population of U.S. hospitals. We then obtained telephone numbers in order to contact survey targets who have profound knowledge about Lean and Six Sigma implementation, such as the Director of Quality, Chief Quality Officer, Patient Safety Director, Risk Management Director, and Director of Nursing. During the phone call, the purpose of this study was explained. We were able to personally speak to 500 administrators at 307 hospitals and asked them to participate in our survey. The potential respondents were promised confidentiality as well as a report of the averaged results. The survey then was sent via e-mail, followed by three rounds of e-mail reminders at monthly intervals. Multiple reminders for our survey generated multiple waves of respondents.
Completed surveys were received from 215 hospitals, achieving a 70% response rate. There were multiple raters completing the questionnaire at 105 of the 215 hospitals. The multiple rater responses for each item were averaged for each hospital, as they exhibited an excellent level of interrater reliability: The intraclass correleation coefficient values ranged from 0.731 to 0.851 (Cicchetti, 1994). Although our response rate was high, we also checked for nonresponse bias using an extrapolation method suggested by Armstrong and Overton (1977). We compared 50 early respondents to 50 late respondents in terms of all the variables used in this article, assuming that late respondents are similar to nonrespondents (Armstrong & Overton, 1977). t Tests revealed no significant mean difference between the two groups at the p < .01 level. Therefore, we conclude that nonresponse bias was not an issue in our sample.
The sample included hospitals from all 50 states. The regional distributions of the respondent hospitals were 19% Western, 32% Midwestern, 20% Southern, 11% Southwestern, and 18% Eastern hospitals. This distribution was not statistically different from the regional distribution of all hospitals included in the Hospitallink.com directory (χ2 = 5.22, df = 4, p = .27).
Lean and Six Sigma Implementation
In order to assess the extent of Lean and Six Sigma implementation of hospitals, we conducted an extensive literature review and adopted empirically validated measures for key tools and activities of the quality programs. Table 1 shows four measurement items for Six Sigma (Gowen et al., 2008) and six items for Lean program (Graban, 2011). For each item, hospitals were asked to what extent they have implemented the tool or practice. The level of deployment was measured on a 0–5 scale (0 = none; 1 = very low; 2 = low; 3 = moderate; 4 = high; 5 = very high).
Performance and Other Characteristics of Hospitals
For hospital performance, we examined responsiveness capability, patient safety, and cost. Table 1 shows four measurement items for a hospital’s responsiveness capability (Darroch, 2005) and four measurement items for patient safety level (McFadden, Henagan, & Gowen, 2009). Net cost saving is a single perceptual measure, which evaluates the cost saving performance after the implementation of quality tools and practices. It was measured on a 0–5 scale (0 = none; 1 = very low; 2 = low; 3 = moderate; 4 = high; 5 = very high).
In addition to performance measures, we examined three hospital characteristics: (a) the number of beds, (b) the number of physicians, and (c) the number of full-time equivalent employees (FTE) who are exclusively dedicated to quality improvement. These measures are a proxy for the size of each hospital as well as the amount of resources available for any quality improvement efforts.
Publically available data from the Centers for Medicare & Medicaid Services and the U.S. Department of Agriculture allowed us to collect more information for our sample hospitals. We examined whether a hospital is a teaching hospital or not, and a private hospital or not. In addition, hospital urban influence code posted by U.S. Department of Agriculture was applied to our samples to indicate the location of the hospital (1 = in large metro area of 1+ million residents, 2 = in small metro area of less than 1 million residents, 3 = metropolitan area adjacent to large metro area, … 12 = most rural, noncore not adjacent to metro or micro area and does not contain a town of at least 2,500 residents).
Reliability and Validity of Variables
Although all of our measures are grounded in the previous literature, we evaluated their reliability and validity through confirmatory factor analysis (Anderson & Gerbing, 1988). The measurement model fits the observed data well. Its chi-square value was 190.594 (p < .001, df = 127). Specifically, multiple indices of model fit (such as comparative fit index, goodness of fit index, and Bollen’s incremental fit index) were greater than 0.9. The point estimate of root mean square error of approximation was smaller than 0.048.
Table 2 includes the specific reliability- and validity-related information. Regarding reliability, Cronbach’s alpha scale reliability values for those factors with multiple measurement items ranged from 0.791 to 0.904, which indicate acceptable internal consistency (Nunnally & Bernstein, 1978). All composite reliability values of the factors also well exceeded the minimum acceptable level of 0.70 (Fornell & Larcker, 1981). With respect to validity, all factor loadings were above 0.5 and statistically significant (p < 0.05), which supports convergent validity for each construct (Fornell & Larcker, 1981). The average variance extracted for each construct was greater than the squared correlations of the construct with the other constructs, which describes good discriminant validity (Fornell & Larcker, 1981).
Finally, we conducted Harman’s one-factor in order to check for the possible presence of common method bias. The χ2 difference between the original measurement model and a new measurement model with one common factor was statistically significant at p < .001, implying that the original model fit the data better (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Therefore, common method bias due may not be a substantial risk.
Cluster analysis has been a main methodology in order to develop empirical taxonomies for business strategy, customer and patient segments, quality management, and other contexts (Brusco, Steinley, Cradit, & Singh, 2012). Although either K-means clustering or hierarchical clustering might have been traditionally popular, they can be venerable for many reasons. Hierarchical clustering is not good at dealing with big data sets, is easily distorted by outliers, and is dependent on researchers’ subjective decision on how many clusters should be retained after data analysis. With K-means clustering, researchers should specify the number of clusters in advance for actual data analysis, which is always questionable and does not guarantee the solution optimality (Brusco et al., 2012 ; Norušis, 2012).
As a way to overcome the disadvantages of the K-means clustering and hierarchical clustering, we used the two-step approach, which combines the advantages of the two traditional techniques (Gore, Tinsley, & Brown, 2000 ; SPSS, 2001). First, cases are assigned to subgroups, and the initial estimate for the number of clusters is obtained. This stage makes the big data analysis easier. Second, those preclusters are grouped into the desired number of clusters, whereas the initial estimate is refined further by finding the greatest change in distance between the two closest clusters. A more detailed algorithm of this two-step approach is available in the SPSS technical report (SPSS, 2001).
Next, in order to verify the obtained clusters, we used multiple statistical tests, including t test, chi-square test, and regression and compared the performance (i.e., responsiveness capability, patient safety, and cost savings) as well as various characteristics (i.e., the number of beds, the number of physicians, and the number of FTEs exclusively dedicated to quality improvement, whether the hospital is a teaching hospital or not, hospital ownership, and hospital urban influence code) of the different hospital groups obtained by the above cluster analysis.
Results of Cluster Analysis
The two-step cluster analysis with the two variables Lean and Six Sigma yielded two clusters, which reflect two discernible patterns of the quality program implementation in U.S. hospitals. In order to further validate these two groups, we applied traditional hierarchical clustering method and K-means method individually to our data. Those traditional methods also generate very similar and consistent results. K-means clustering with three-cluster and four-cluster models was not supported by our data. Consistent cluster assignments despite different methods imply the stability of the two identified groups. The size of each group and membership profile information including the extent of Lean and Six Sigma implementation are summarized in Figure 1. Two clusters are comparable in terms of size but show different implementation patterns of Lean and Six Sigma programs.
Cluster 1 (n = 116) represents hospitals that are focusing on Six Sigma activities. The average extent of Six Sigma implementation is 2.96 (out of 5). The average Lean implementation level is only 0.49 (out of 5). As opposed to Cluster 1, Cluster 2 (n = 99) shows a high level of average Lean implementation score, which is 3.26 (out of 5). The average extent of Six Sigma implementation for hospitals for this cluster is 3.90 (out of 5), which is even higher than those in Cluster 1. Apparently, Cluster 2 represents hospitals that are deep into implementing both Lean and Six Sigma activities for their quality improvement.
Having obtained these two clusters, we used a t test (for noncategorical data) and a chi-square test (for categorical data) to determine whether the two groups are significantly different with regards to various performance and hospital characteristic measures. Cluster 1 and Cluster 2 are indeed statistically different for almost all the variables we examined. Table 3 shows the results.
For hospital performance, we find that Cluster 2 is consistently better than Cluster 1. Hospitals in Cluster 2 show a statistically higher level of performance in terms of responsiveness capability (p < .01) and patient safety (p < .05) than hospitals in Cluster 1. For the cost saving performance via quality improvement activities, Cluster 2 also outperforms Cluster 1, although it is marginally significant (p < .1).
For hospital characteristics, hospitals in Cluster 2 have more beds (p < .05) and more physicians (p < .1) than hospitals in Cluster 1. In addition, Cluster 2 has more FTEs who are exclusively dedicated to quality improvement (p < .05) than Cluster 1. Hospitals in Cluster 2 are located in more urban areas (p < .05). Chi-square tests imply that teaching hospitals and private hospitals are more likely to be in Cluster 2 than Cluster 1.
In order to compare the performance of the two clusters more systematically, additional regression analyses were conducted. The major independent variable is a dummy variable that indicates whether the hospital is in Cluster 2 or not. Five control variables were entered including (a) the number of beds, (b) the number of FTEs who are exclusively dedicated to quality improvement, (c) whether the hospital is a teaching hospital or not, (d) hospital ownership, and (e) urban influence code. The number of physicians was not controlled due to a multicollinearity issue since the correlation between the variable and the number of beds was too high. Table 4 shows consistent results: Cluster 2 is more positively related to responsiveness capability and patient safety. However, the relationship between Cluster 2 and cost saving becomes weak after other control variables are included.
After interpreting each cluster’s natures, we labeled the two clusters as the Moderate Six Sigma group (Cluster 1) and the Lean Six Sigma group (Cluster 2). Specifically, the Moderate Six Sigma group represents hospitals that implement Six Sigma activities at a moderate level but with little to no Lean implementation. This might imply that they employ the project-based Six Sigma approach in order to resolve any identified quality problems, without incorporating Lean to manage the entire system and its internal processes on a daily basis for continuous improvement.
In contrast, the Lean Six Sigma cluster represents hospitals that embrace both Lean and Six Sigma activities for continuous improvement. As opposed to the Moderate Six Sigma group, hospitals in the Lean Six Sigma group exhibit a much higher level of Lean implementation, combined with a high level of Six Sigma implementation. Lean’s holistic view to span the entire system and employees seems to allow hospitals to consistently achieve a higher performance level in terms of responsiveness capability, patient safety, and probably cost savings.
The t test and chi-square results on hospital characteristics indicate that the Lean Six Sigma group hospitals tend to be larger and have more resources for quality improvement activities than the Moderate Six Sigma group hospitals, in terms of the average number of beds, the number of physicians to serve patients, and the number of FTEs who are exclusively dedicated to quality improvement. In addition, teaching hospitals and private hospitals are more likely to be in the Lean Six Sigma group compared to non-teaching hospitals and public hospitals. We interpret this to suggest that teaching and private hospitals tend to pay more attention to quality improvement.
We should clarify that we did not know in advance whether hospitals in our sample were using Lean or Six Sigma. The extent to which of these programs are utilized by U.S. hospitals was our main research question (i.e., figuring out the current status of Lean and Six Sigma implementation). Because of lack of prior empirical research and well-established theories, we were not able to assure in advance how many or what kinds of clusters would exist. After obtaining a large and representative sample of data and using cluster analysis, we identified only two clusters (Lean Six Sigma and Moderate Six Sigma) that reflect the current status of Lean and Six Sigma implementation in U.S. hospitals. An important finding from our analysis was that a cluster was not obtained for hospitals using only Lean. Thus, our study determined that implementing Lean without Six Sigma was not a predominant implementation strategy in U.S. hospitals. The study offers justification for encouraging hospitals to implement more Lean tools in combination with Six Sigma for better performance outcomes.
Overall, our results propose meaningful implications for practice and theory in quality management in health care, expounding upon the possible complementarity of Lean and Six Sigma initiatives. First, our research provides empirical evidence to those advocates of Lean Six Sigma. The benefit of combining Lean with Six Sigma approaches has been assumed and tested in the manufacturing industry (Shah et al., 2008). Although the need to synthesize Lean thinking with a Six Sigma approach for health care systems has also been proposed by practitioners and researchers, the dominant research methodology in the health care literature has been the case study method (e.g., Ahmed et al., 2013 ; Chassin et al., 2015). Although we appreciate the ability of the case study method to generate holistic knowledge about a system, we also acknowledge its limitations in generalizability. We therefore conducted empirical analyses with cross-sectional data and systematically identified the benefits of implementing the two approaches simultaneously.
Second, these results also support the dynamic capabilities theory in providing empirical evidence for the superiority of the combination of Lean and Six Sigma for health care improvement. Anand et al. (2009) suggests that continuous quality improvement techniques such as Lean and Six Sigma can serve as dynamic capabilities, since they are designed to improve the firm’s ability to integrate and restructure functional competencies, institutionalize organizational learning, and increase an organizations strategic response to changing environmental conditions. By executing these quality improvement programs together, health care systems would be better able to nurture proper routines in which employees quickly acknowledge problems, identify solutions, and reorganize resources to implement the solutions (Teece et al., 1997). The routines become new valuable intangible assets to build long-term competitive advantages through continuous improvement.
Third, we assert that Lean should be further encouraged among hospitals. Although previous research found that more U.S. hospitals are implementing Lean (53%) than Six Sigma (between 27% and 42%; Feng & Manuel, 2008 ; Henke, 2009), our results reveal somewhat different perspectives on the current status of the quality improvement initiatives. As both groups implement Six Sigma at moderate to high levels (see Figure 1), most U.S. hospitals now seem to sufficiently practice Six Sigma. This might be because they can obtain a lot of support from educators and consultants for the adoption of Six Sigma tools. However, Lean does not seem to be implemented thoroughly in the U.S. hospitals. The average Lean implementation level of the entire hospitals in our data was only 1.736 (out of 5; see Table 2). Moreover, the Moderate Six Sigma group, which accounts for 54% of the sample hospitals, has a significantly low level of average implementation score of Lean, which was only 0.49 out of 5 (see Figure 1). It is worth reminding that we examined the extent to which hospitals employ each quality program’s tools and practices, rather than whether or not the hospitals simply implement the tools. Therefore, our results might reveal that U.S. hospitals are still not capable of utilizing the complete set of Lean tools developed by manufacturing companies. As the combination of Lean and Six Sigma seems to generate better outcomes, the low level of Lean implementation in many hospitals would be a serious concern. Future research should identify the Lean-specific barriers and remedies in hospitals.
We attempted to discover existing patterns of Lean and Six Sigma implementation from 215 hospitals in the United States via a cluster analysis. Two clusters, the Moderate Six Sigma group and the Lean Six Sigma group, were identified. We then found that superior performance can be achieved by putting effort into both quality improvement initiatives at the same time. By doing so, this study provides an important step toward a better understanding of the Lean and Six Sigma implementation in hospitals and offers some useful insights for practitioners and managers.
First, our results suggest that combining Lean and Six Sigma can generate synergetic effects. The Lean tools and practices to eliminate unessential non-value-added elements within the system and to optimize work flows and processes could provide a better environment in which Six Sigma problem-solving tools play significant roles. Managers thus need to understand philosophies, practices, and tools for each initiative and employ them simultaneously to eliminate any wastes in health care systems in a more scientific way.
Second, our research encourages practitioners to critically think about how to support Lean implementation in hospitals. Given that the Lean concept was actually developed and introduced earlier than Six Sigma in the manufacturing field (Shah et al., 2008), our result that Lean is implemented less than Six Sigma in hospitals in terms of intensity is quite interesting. To our best knowledge, there has been no research on whether Lean is more difficult to implement than Six Sigma in hospital settings. Some studies only report that hospitals feel reluctant to implement Lean for reasons such as limiting health care professional’s decision-making autonomy (Graban, 2011) and taking extra time to train and supervise workers (Kelly, Bryant, Cox, & Jolley, 2007). In order to overcome these challenges and to ensure the successful Lean implementation, managers should make an extra effort to arrange the proper set of resources not only at the organizational level but also at the team and individual level. Those resources include team-based organizational structures, empowerment cultures, training programs to introduce Lean tools, and philosophies to health care practitioners (Harrison et al., 2016 ; Hung et al., 2016).
Limitations and Future Research Directions
Although this study provides both a practical and theoretical contribution to our understanding of Lean and Six Sigma implementation in health care, it is not without limitations. The first shortcoming is that this is an exploratory study that uses perceptual data. A potential limitation of using perceptual data in this study is the possibility of self-reported bias from respondents about Lean and Six Sigma implementation and performance. Future research can be conducted that might overcome some limitations of this exploratory study. For example, future research could utilize some increasing available objective performance measures, such as the Centers for Medicare & Medicaid Services database. Although such data have not been reported by all hospitals, the contribution of Lean and Six Sigma to hospitals might become clearer across different outcome measures.
A second limitation of this study is that it is based on cross-sectional survey data. Future research should include longitudinal studies on the patterns of Six Sigma and Lean implementation to confirm and refine the observed patterns. By doing so, although we identified just two clusters, more refined or different patterns could be identified with a bigger data set, which can examine the evolution of Lean and Six Sigma implementation in hospitals.
A third limitation of this study is that there may be other organizational variables capable of influencing the phenomena under study that have not been considered here. In order to further verify the effectiveness of Lean and Six Sigma initiatives, more sophisticated experiments or more rigorous regression models can be designed and carried out. Although testing the interaction effect of Lean and Six Sigma on performance measures goes beyond the scope of this research study, we consider this a fruitful direction for future research. Nonetheless, our exploratory results could provide interesting insights on the kinds of variables that can be examined and should be controlled for in future studies.
Ahmed S., Manaf N. H., Islam R. (2013). Effects of Lean Six Sigma
application in healthcare services: A literature review. Reviews on Environmental Health
, 28(4), 189–194.
Anand G., Ward P. T., Tatikonda M. V., Schilling D. A. (2009). Dynamic capabilities through continuous improvement infrastructure. Journal of Operations Management
, 27(6), 444–461.
Anderson J. C., Gerbing D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin
, 103(3), 411–423.
Andersson R., Eriksson H., Torstensson H. (2006). Similarities and differences between TQM, Six Sigma
. The TQM Magazine
, 18(3), 282–296.
Armstrong J. S., Overton T. S. (1977). Estimating nonresponse bias in mail surveys. Journal of Marketing Research
, 14(3), 396–402.
Brusco M. J., Steinley D., Cradit J. D., Singh R. (2012). Emergent clustering methods for empirical OM research. Journal of Operations Management
, 30(6), 454–466.
Chassin M. R., Mayer C., Nether K. (2015). Improving hand hygiene at eight hospitals in the United States by targeting specific causes of noncompliance. The Joint Commission Journal on Quality and Patient
, 41(1), 4–12.
Cicchetti D. V. (1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological Assessment
, 6(4), 284–290.
Cottington S., Forst S. (2010). Lean healthcare: Get your facility into shape
. Danvers, MA: HCPro.
Darroch J. (2005). Knowledge management, innovation and firm performance. Journal of Knowledge Management
, 9(3), 101–115.
Deblois S., Lepanto L. (2016). Lean
and Six Sigma
in acute care: A systematic review of reviews. International Journal of Health Care Quality Assurance
, 29(2), 192–208.
DelliFraine J. L., Wang Z., McCaughey D., Langabeer J. R. II, Erwin C. O. (2013). The use of Six Sigma
in health care management: Are we using it to its full potential? Quality Management in Health Care
, 22(3), 210–223.
Ding D. X. (2014). The effect of experience, ownership and focus on productive efficiency: A longitudinal study of U.S. hospitals. Journal of Operations Management
, 32(1), 1–14.
Dobrzykowski D. D., McFadden K. L., Vonderembse M. A. (2016). Examining pathways to safety and financial performance in hospitals: A study of Lean
in professional service operations. Journal of Operations Management
, 42, 39–51.
Feng Q., Manuel C. M. (2008). Under the knife: A national survey of six sigma
programs in US healthcare organizations. International Journal of Health Care Quality Assurance
, 21(6), 535–547.
Field J. M., Heineke J., Langabeer J. R., DelliFraine J. L. (2014). Building the case for quality improvement in the health care industry: a focus on goals and training. Quality Management in Health Care
, 23(3), 138–154.
Fornell C., Larcker D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research
, 18(1), 39–50.
Gore P. A., Tinsley H. E., Brown S. D. (2000). Handbook of applied multivariate statistics and mathematical modeling. Chapter cluster analysis
, (pp. 298–318). San Diego, CA: Academic Press.
Gowen C. R. III, Stock G. N., McFadden K. L. (2008). Simultaneous implementation of Six Sigma
and knowledge management in hospitals. International Journal of Production Research
, 46(23), 6781–6795.
Graban M. (2011). Lean hospitals: Improving quality, patient safety, and employee satisfaction
. Boca Raton, FL: CRC Press.
Hagan P. (2011). Waste not, want not: Leading the Lean
health-care journey at Seattle Children’s Hospital. Global Business and Organizational Excellence
, 30(3), 25–31.
Harrison M. I., Paez K., Carman K. L., Stephens J., Smeeding L., Devers K. J., Garfinkel S. (2016). Effects of organizational context on Lean
implementation in five hospital systems. Health Care Management Review
, 41(2), 127–144.
Henke C. (2009). ASQ study: Healthcare catches on to Lean
, Six Sigma
. Quality Progress
, 42(5), 17.
Hung D., Gray C., Martinez M., Schmittdiel J., Harrison M. I. (2016). Acceptance of Lean
redesigns in primary care: A contextual analysis. In Health Care Management Review
. [Epub ahead of print].
Jayasinha Y. (2016). Decreasing turnaround time and increasing patient satisfaction in a safety net hospital-based pediatrics clinic using Lean Six Sigma
methodologies. Quality Management in Health Care
, 25(1), 38–43.
Kelly A. M., Bryant M., Cox L., Jolley D. (2007). Improving emergency department efficiency by patient streaming to outcomes-based teams. Australian Health Review
, 31(1), 16–21.
Langabeer J. R., DelliFraine J. L., Heineke J., Abbass I. (2009). Implementation of Lean
and Six Sigma
quality initiatives in hospitals: A goal theoretic perspective. Operations Management Research
, 2(1), 13–27.
Lewis M. A., Brown A. D. (2012). How different is professional service operations management? Journal of Operations Management
, 30(1), 1–11.
Liberatore M. J. (2013). Six Sigma
in healthcare delivery. International Journal of Health Care Quality Assurance
, 26(7), 601–626.
Mason S. E., Nicolay C. R., Darzi A. (2015). The use of Lean
and Six Sigma
methodologies in surgery: A systematic review. The Surgeon
, 13(2), 91–100.
McFadden K. L., Henagan S. C., Gowen C. R. III (2009). The patient safety
chain: Transformational leadership’s effect on patient safety
culture, initiatives, and outcomes. Journal of Operations Management
, 27(5), 390–404.
McFadden K. L., Lee J. Y., Gowen C. R. III, Sharp B. M. (2014). Linking quality improvement practices to knowledge management capabilities. The Quality Management Journal
, 21(1), 42–57.
Norušis M. J. (2012). IBM SPSS Statistics 19 Advanced Statistical Procedures Companion
. Englewood Cliffs, NJ: Prentice Hall.
Nunnally J. C., Bernstein I. H. (1978). Psychometric theory
. New York: McGraw-Hill.
Podsakoff P. M., MacKenzie S. B., Lee J. Y., Podsakoff N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology
, 88(5), 879–903.
Shah R., Chandrasekaran A., Linderman K. (2008). In pursuit of implementation patterns: The context of Lean
and Six Sigma
. International Journal of Production Research
, 46(23), 6679–6699.
SPSS. (2001). The SPSS two-step cluster component
. White paper technical report.
Teece D. J., Pisano G., Shuen A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal
, 18(7), 509–533.
Wilden R., Devinney T. M., Dowling G. R. (2016). The architecture of dynamic capability research identifying the building blocks of a configurational approach. The Academy of Management Annals
, 10(1), 997–1076.