Applying Six Sigma and DMAIC to diversity initiatives.

Metrics are central to gauging the success of any business endeavor--whether that endeavor is improving market share, creating efficiencies in the supply chain, or reducing infection rates. Metrics can also be used in determining the effectiveness of diversity initiatives, such as ameliorating racial and ethnic disparities in care or recruiting and retaining a diverse workforce. Unless we set goals and examine the activities for reaching those goals, we cannot be sure that we are achieving our targets.

Rohini Anand, Ph.D. (2007), senior vice president and global chief diversity officer of Sodexho--a leading provider of food and facilities management services that has been repeatedly named by DiversityInc. as one of the Top 50 Companies for Diversity--describes the centrality of metrics in Sodexho's nationally recognized success:

"At Sodexho the adage 'what gets measured gets done' is a truism. In order to ensure that diversity and inclusion are imbedded in the organization, we established a sophisticated scorecard that measures both progress in diversity outcomes and, more importantly, the underlying processes and inclusive behaviors that impact the outcomes. Progress is measured and reported monthly, and managers are held accountable by linking the metrics to incentive compensation, as we do with any other measures of organizational success."

Arguably, what is measured is more important than the act of measurement itself. By looking at the hallmarks, activities, or distinguishing features of organizations that are considered diversity leaders, we can identify diversity areas in our own organization that can benefit most from quality improvement. The previous four installments of this column highlighted the following three areas of focus:

1. leadership, strategy, and climate;

2. human resource management; and

3. culturally competent care, interpreters, and translators.

Each of these areas can be initially evaluated using the "Cultural Competency Assessment Tool for Hospitals" (CCATH). (1) The CCATH assesses the extent to which the organization adheres to best practices in each given area. The organization can then take action to close the gap between current and best practices by developing the appropriate infrastructure and the processes of using that infrastructure to produce desired outcomes. A necessary next step is to directly define and assess success in achieving desired organization-specific outcomes so that the infrastructure and its processes can be continually fine-tuned. The ultimate goal here is to apply standard quality improvement methods, including Six Sigma, to diversity initiatives. According to Billington and Billington (2003), Six Sigma is "a data-driven, methodical program of continuous and breakthrough improvement focused on customers and their critical requirements."

OUTCOME METRICS FOR DIVERSITY AREAS

Two or three outcome metrics for each of the diversity areas we have identified are as follows.

1. Leadership, strategy, and climate

a. Concordance between leadership team diversity and service area or workforce diversity

b. Employee satisfaction with the workplace by key dimensions of diversity, such as gender, race/ethnicity, generation, tenure, role, and department

c. Supplier diversity, as measured by the proportion of procurement from women and minority-owned suppliers

2. Human resource management

a. Employee retention rates by key dimensions of diversity, such as gender, race/ethnicity, generation, tenure, role, and department

b. Compensation by key dimensions of diversity, controlling for human capital variables and performance

3. Culturally competent care, interpreters, and translators

a. Patient satisfaction with care across key dimensions of diversity, such as gender, race/ethnicity, and generation

b. Achievement of the Agency for Healthcare Research and Quality's (AHRQ) 2006 quality measures (see www.ahrq.gov/qual/nhdr06/measurespec/ index.html), by race and ethnicity

Note that these outcome metrics may not be the most appropriate first-tier choices for your organization, as each organization must determine its own desired outcomes in light of its unique mission, vision, and strategy.

APPLICATION OF SIX SIGMA AND DMAIC 

As reported in the first installment of this column, demographic diversity presents both challenges and opportunities. Research on diversity's impact reveals that its effect is heavily dependent on the organizational context, including business strategy, human resource practices, culture and climate, and leadership. As Kochan and his colleagues (2003) concluded from their comprehensive study, the business case for demographic diversity is nuanced. Responsible healthcare leaders ensure that their diversity dollars are well spent. They not only evaluate the presence of the best-possible diversity infrastructure and practices, they also strive to improve results. Doing so requires using outcome-oriented metrics and committing to continual assessment of organizational performance.

Bank of America, which ranked number 1 on DiversityInc.'s Top 50 Companies for Diversity in 2007, is one of a handful of high performers that use Six Sigma and its five-step DMAIC (define, measure, analyze, improve, control) process to assess the impact of its diversity initiatives (Millman 2007). This global financial-services firm uses this practice throughout the organization, and it also applies this methodology to its diversity and inclusion initiatives. For its diversity initiatives, Bank of America uses Six Sigma and DMAIC to evaluate its affinity group mentoring program. Geri Thomas, diversity executive at Bank of America, describes the benefits of following the Six Sigma approach (Sapp 2007):

"Six Sigma methodology has helped us design, and now measure, the impact of the program. We identified improvement opportunities, ensuring [that] changes are data-driven, implemented, and refreshed and are now becoming part of the ongoing management process."

Healthcare organizations can emulate diversity-initiative leaders like Bank of America by (1) building and using a diversity infrastructure, (2) defining outcome-based measures of impact, and (3) using DMAIC to assess and continuously improve organizational performance.

Originated in the manufacturing sector by Motorola, Six Sigma uses statistical analyses to measure and reveal opportunities for process improvement by uncovering defects (Lucas 2002; Mader 2007). For a service firm, a defect is any transaction, service encounter, action, or procedure that does not meet customer expectations. The five-step DMAIC technique is the most prevalent Six Sigma method used for process improvement (de Mast and Bisgaard 2007). Its elements are as follows:

* Define: select the process that needs improvement.

* Measure: translate the process into quantifiable forms, collect data, and assess current performance.

* Analyze: identify the root causes of defects and set goals for performance.

* Improve: implement and evaluate changes (solutions) to the process to remove root causes of defects.

* Control: standardize solutions, and continuously monitor improvement.

Diversity and cultural competence are not transaction-oriented, as Bank of America's experience indicates. However, Six Sigma and the DMAIC process can be used effectively in managing diversity programs. For example, DMAIC can be used to ensure that the process that leads to patient satisfaction with quality of care is not experienced less positively by patients of color. DMAIC can help the organization to identify and ameliorate disparities in satisfaction through identifying root causes and implementing solutions that minimize or remove those root causes.

In this diversity-focused process improvement application, the DMAIC steps could be as follows:

Define

Six Sigma processes selected for improvement are normally those that are critical to the success of the organization. At the define step, assume that customer satisfaction with care is critical to success. Any defect in that care process, as viewed by the customer, leads to dissatisfaction. Thus, the view of the customer is critical to success. Because the goal of Six Sigma and DMAIC is to remove defects, this method will focus on eliminating processes or factors in care that result in dissatisfaction. A patient satisfaction survey will reveal areas of dissatisfaction, and this will define the process that needs to be improved using Six Sigma and DMAIC. One process or factor that could be improved in the diversity context is the lack of or minimal interpreter services.

Measure

Well-designed performance measurement systems use metrics that are quantitative, are easy to understand, encourage appropriate behavior, and are visible. In addition, performance measures must encompass outputs and inputs, which are defined and mutually understood by those who will use them. In short, performance measures should be multidimensional so that every member of the team has a clear understanding of what behavior patterns need to be changed.

A Pareto chart is the quality improvement tool most often used to display or array the data obtained in the measure step. A Pareto chart is a bar chart that graphically represents the frequency distribution of each performance metric. In the example we used in the define step--improving patient satisfaction with care--a Pareto chart can depict the availability of interpreter services by language. The chart can display the proportion of patients by primary language and the availability of interpreters for that language group. This way, availability versus need can be easily seen, allowing managers to uncover gaps in service and to determine which gap has to be tackled first.

The measure step also involves setting goals for each metric. In the interpreter services example, a goal might be to increase the availability of Spanish language interpreters to meet the demand of the population whose primary language is Spanish.

Analyze

The analyze step involves determining the root causes of defective products and services. Data from patient satisfaction surveys and focus groups may be analyzed by race and ethnicity of respondents. If results indicate that patients of color are less satisfied than their white counterparts, the root causes of this finding should be explored. Determining root causes can be done through assessing the organization's conformity to best practices, as measured by the CCATH or the American Hospital Association's (AHA) "Diversity and Cultural Proficiency Assessment Tool for Leaders" (see www.aha.org/aha/content/2004/pdf/diversitytool.pdf). Most importantly, the organization should also analyze its own processes.

A cause-and-effect diagram (also called a fishbone diagram or Ishikawa diagram, after its developer, Kauro Ishikawa) can be used to determine potential causes. Ideally, such a diagram is the product of participatory brainstorming or an assessment exercise in which potential causes of the identified defect are categorized into meaningful and manageable groupings. Figure 1 is a cause-and-effect diagram that displays the root causes of a gap in patient satisfaction by race and ethnicity. In this figure, four categories are used: policies, procedures, people, and plant/technology. Insights from the CCATH, the AHA assessment tool, or other similar inventory of performance against best practices are invaluable to the brainstorming process.

Figure 1 displays the relationship among related causes on each of the four major categories. As shown in the figure, we performed only a first-level analysis. In an actual application of the diagram, the brainstorming group can jot down multiple levels of analysis, underlining causes behind causes. Once all possible causes have been entered, the group would then review the finished diagram, gain consensus on the top five or so root causes, and assign a priority ranking to each cause.

Improve

At the improve step, action is taken to address the previously identified root causes. For example, in Figure 1, the root cause "cultural competence training is not offered to clinicians" might be ameliorated by providing cultural competence training for clinicians. This training could improve the process that in turn leads to patients' satisfaction with their care. In this step, solutions are identified and implemented for each of the root causes revealed in the analyze step and selected for improvement.

Control

The control step involves monitoring the status of the improvements to the process to ensure that goals are achieved and acceptable behavior patterns are maintained throughout the organization. When patient satisfaction metrics are monitored and made visible to everyone concerned, cultural competence becomes a critical success factor. At this stage, new possibilities for process improvement are revealed, enhancing the concept of continuous improvement inherent in the Six Sigma method. By beginning with satisfaction and expanding to other metrics, such as the AHRQ 2006 quality measures for patients, organizations can systematically improve their diversity performance.

[FIGURE 1 OMITTED]

CONCLUSION

Although Six Sigma and DMAIC are most often associated with the manufacturing sector, they can be used effectively to improve an organization's diversity strategies and management. As Bank of America has demonstrated through its mentoring program, diversity metrics can be improved by using this quality improvement technique. By committing to the use of metrics and with continuous improvement, targets for high-quality care and goals of equitable treatment of all patients and employees can be reached.

References

Agency for Healthcare Research and Quality (AHRQ). 2006. "National Healthcare Disparities Report 2006. Appendix C: Measure Specifications Quality of Health Care." [Online information; retrieved 7/23/07.] www.ahrq.gov/qual/nhdr06/measurespec/index.html.

Anand, R. 2007. Personal communication with authors, July 30.

Billington, M. G., and P. J. Billington. 2003. "Six Sigma: Quality Performance." [Online information; retrieved 8/14/07.] www.clomedia.com/content/templates/clo_feature.asp?articleid=207. de Mast, J., and S. Bisgaard. 2007. "The Science of Six Sigma." Quality Progress 40 (1): 25-29.

Kochan, T., K. Berukova, R. Ely, S. Jackson, A. Joshi, K. Jehn, J. Leonard, D. Levine, and D. Thomas. 2003. "The Effects of Diversity on Business Performance: Report of the Diversity Research Network." Human Resource Management 42 (1): 3-21.

Lucas, J. M. 2002. "The Essential Six Sigma." Quality Progress 35 (1): 27-31.

Mader, D. R 2007. "How to Identify and Select Lean Six Sigma Projects." Quality Progress 40 (7): 58-60.

Millman, J. 2007. "Why You Need Diversity to be Competitive: Case Studies from the 2007 DiversityInc. Top 50 Companies for Diversity." DiversityInc Magazine (June): 24-44.

Sapp, K. 2007. Personal communication with the author, September 27.

Janice L. Dreachslin, Ph.D., professor of health policy and administration, and professor in charge, Biotechnology and Health Industry Management, Penn State Great Valley, School of Graduate Professional Studies, Malvern, Pennsylvania, and Peggy D. Lee, Ph.D., assistant professor of management, Penn State Great Valley, School of Graduate Professional Studies

Note

(1.) For further information or to obtain the instrument, contact Dr. Weech-Maldonado at [e-mail].