Read the article about combining Six Sigma and knowledge

International Journal of Production Research,
Vol. 46, No. 23, 1 December 2008, 6781–6795

Simultaneous implementation of Six Sigma and knowledge
management in hospitals
C. R. GOWEN III*y, G. N. STOCKz and K. L. MCFADDENx
yDepartment of Management, Northern Illinois University, DeKalb, Illinois, USA
zCollege of Business, University of Colarado at Colorado Springs, Colorado Springs,
Colorado, USA
xDepartment of Operations Management and Information Systems, Northern Illinois University,
DeKalb, Illinois, USA

(Revision received May 2008)
Six Sigma programmes aspire to reduce variation in organizational processes and
achieve clear financial results. Six Sigma initiatives have proven to be an effective
technique for improving quality in manufacturing. Similarly, the importance
of knowledge management has grown considerably in recent years and has
emerged as a major source of competitive advantage for manufacturing firms.
From the perspective of a decision support system, knowledge management
is concerned with information acquisition, dissemination, and responsiveness.
Little research has examined simultaneous applications of Six Sigma and
knowledge management. The purpose of this paper is to explore the usefulness
of knowledge management for the implementation of Six Sigma in hospitals.
We hypothesize that knowledge management will enhance the implementation of
Six Sigma by leading to improvements of quality programme results and
sustainable competitive advantage. The results of hierarchical regression analysis
demonstrate that knowledge management does ameliorate the success of Six
Sigma initiatives, specifically for knowledge dissemination and responsiveness.
These results are discussed in terms of the contributions to existing theory and for
managers of Six Sigma and knowledge management initiatives.
Keywords: Six Sigma; Knowledge management; Competitive advantage

1. Introduction
To achieve a competitive advantage, organizations have recently adopted Six Sigma
initiatives and knowledge management systems. However, investment in quality and
information systems is not necessarily effective. Research has revealed significant
mediators of organizational performance, such as knowledge-based dynamic
capability and organizational learning (Wang et al. 2007, Yeung et al. 2007).
Likewise, knowledge management could enhance the effectiveness of quality
initiatives through a decision support system, such as an information technology
infrastructure (Hartman et al. 2002, Hsu and Shen 2005). The purpose of this paper
is to test the synergistic effects of Six Sigma and knowledge management on
*Corresponding author. Email: [email protected]
International Journal of Production Research
ISSN 0020–7543 print/ISSN 1366–588X online ß 2008 Taylor & Francis
http://www.tandf.co.uk/journals
DOI: 10.1080/00207540802496162

6782

C. R. Gowen III et al.

organizational success. Specifically, we propose that the application of knowledge
management enhances the effect of Six Sigma on quality programme results and
sustainable competitive advantage.
The concurrent implementation of Six Sigma initiatives and knowledge
management is relatively novel and the extant literature lacks an assessment of its
effectiveness. Theory development and strong empirical support for Six Sigma and
knowledge management concepts originated mainly from manufacturing settings
(Gunasekaran and Ngai 2007). Similar models of successful service Six Sigma
and knowledge management have been proposed but have attracted less scholarly
attention (Antony 2006). Given the paramount importance of patient safety issues
today, quality and knowledge management provide leading opportunities for
improvement of healthcare systems (Ruiz 2004). Silverstein (2006, p. 39) concluded
that ‘there’s no better place to apply Six Sigma than the healthcare industry’ as it is
‘process intensive business, rife with data’. Consequently, we examine the theoretical
foundations for Six Sigma and knowledge management, describe the methodology
for testing our hypotheses, and present our results, conclusions, and implications for
practice and future research.

2. Literature review
Six Sigma is a process improvement initiative designed by leading manufacturing
companies and recently adopted by service organizations (Antony 2006). It is
a systematic data-driven approach that resolves errors in processes by focusing on
organizational outcomes most critical to customers (Breyfogle 2003). The Six Sigma
quality level is characterized by only 3.4 or fewer defects per million opportunities.
Six Sigma team members are trained in problem solving and statistical techniques.
Team projects are selected based on customer requirements and on their ability to
achieve clear financial returns for the organization.
Motorola pioneered the concept of Six Sigma and won the prestigious Malcolm
Baldrige National Quality Award in 1988 largely due to their Six Sigma initiative.
Six Sigma programmes at other manufacturing companies, such as General Electric
and Allied Signal, provided a significant amount of credibility and media attention.
Over the past two decades, Six Sigma has become an increasingly popular initiative
across a range of industries (Kumar et al. 2006). Empirical results indicate that Six
Sigma has contributed to process and quality improvement, customer satisfaction,
and corporate competitiveness (Lee and Choi 2006). Furthermore, Gowen and
Tallon (2005) found that companies with higher levels of technological intensity were
more likely to implement Six Sigma, as well as more likely to achieve a competitive
advantage as a result of such implementation, compared with low technological
intensive companies.
As a means for resolving patient safety issues, many healthcare organizations
have undertaken Six Sigma initiatives targeted toward improving organizational
performance (Lloyd and Holsenback 2006). The reduction of medical errors in
healthcare can be compared to conformance quality in manufacturing. The adoption
of Six Sigma programmes has expanded only recently to healthcare organizations
(Carrigan and Kujawa 2006). Research suggests that the implementation of quality
programmes significantly improves patient satisfaction (Marley et al. 2004).

Simultaneous implementation of Six Sigma and knowledge management

6783

Specifically, case studies of Six Sigma initiatives have resulted in diverse pragmatic
improvements, such as clinical, operational, and service benefits (Carpenter 2006,
Craven et al. 2006, Sherman 2006).
Likewise, the importance of knowledge management has grown considerably
in recent years and has emerged as a major source of competitiveness mainly for
manufacturing firms (Gunasekaran and Ngai 2007). For our purposes, knowledge
management is defined as ‘the process that creates or locates knowledge and
manages the dissemination and use of knowledge within and between organizations’
(Darroch 2003, p. 41). Successful knowledge management depends on the relevant
technical infrastructure to capture, store, share, and use information common to
a decision support system (Lee and Choi 2003). Knowledge management is a
systematic and cross-disciplinary approach to improving an organization’s ability
to mobilize knowledge which supports decision making (Hsu and Shen 2005).
Moreover, applications of knowledge management as a decision support system
have proven successful in manufacturing organizations (Dayan 2003, Dayan and
Evans 2006, Irani et al. 2007, Nachiappan et al. 2007) and healthcare settings
(Hartman et al. 2002).
Knowledge management can be represented as a three-stage process of
knowledge acquisition, dissemination, and responsiveness (Darroch 2003).
Knowledge acquisition relates to the location, creation, and discovery processes.
For example, knowledge could be acquired from employees, databases, and
relationships between the firm and its customers or suppliers. Knowledge
dissemination measures how knowledge is applied and distributed in the organization. Knowledge responsiveness refers to the way that the organization utilizes
various types of knowledge, such as how a company can use knowledge about
customer behavior to improve customer satisfaction and retention. Having knowledge available to the right people at the right time is critical in building an
organization’s competencies (Alazmi and Zairi 2003). Information sharing is critical
for successful organizational processes, such as supply chain management (Chandra
et al. 2007), but only if the benefits outweigh the risks (Smith et al. 2007). Several
empirical studies have also revealed that knowledge management practices can lead
to improvement of organizational effectiveness (McCann and Buckner 2004, Yeung
et al. 2007).
Knowledge management offers a compelling complement to the success of
quality management initiatives (Choo et al. 2007), such as the Malcolm Baldrige
National Quality Award process (Meyer and Collier 2001, Lee et al. 2006).
Information technology (IT) initiatives to support knowledge management can lead
to greater organizational performance (Wang et al. 2007). The Institute of Medicine
(2000) reported that United States’ medical errors contribute to more than one
million injuries and up to 98,000 deaths annually, for which 58% were preventable.
Poor healthcare information, such as incorrect medication administration, accounts
for many of those fatalities. The Institute of Medicine report also claimed that IT
initiatives, such as electronic prescriptions, could eliminate up to 80% of dispensing
errors. Likewise, IT applications have been applied to the automation of more
routine tasks to resolve recent nursing staff shortages and so that nurses are allowed
to devote more attention to patient safety issues (Mullaney and Weintraub 2005).
Although healthcare IT initiatives have been expensive and slow, case studies reveal
that they have resulted in greater quality of patient care (Carpenter 2006).

6784

C. R. Gowen III et al.

3. Sustainable competitive advantage and hypotheses
In the dynamic capabilities theory, the effective implementation of Six Sigma and
knowledge management could result in sustainable competitive advantage (Barney
2002). Dynamic capabilities are organizational processes that effectively utilize
organizational resources (Winter 2003). Knowledge-based dynamic capability has
been reported as the link between IT support for knowledge management and firm
performance (Wang et al. 2007). Competitive advantage can be achieved and
sustained from resources and dynamic capabilities that are characterized by four
factors: value, rareness, imitation cost, and non-substitutability (Hitt et al. 2007).
Value refers to the degree that the firm’s resources enable the organization to
respond to external threats and opportunities. Rareness concerns the degree that
competing firms do not possess the organization’s particular valuable resources,
such as a pharmaceutical firm’s patented products. Imitation cost focuses on the
cost disadvantage faced by other firms that do not possess a certain resource.
Non-substitutability captures the degree that a resource has no strategic equivalent.
Practically, resources limitations force an organization to capture only some measure
of each factor. Certain Six Sigma dimensions, such as Black/Green Belt training and
teams, other strategic human resource practices, and DMAIC (define, measure,
analyse, improve, and control) process management, can be dynamic capabilities
which lead to sustainable competitive advantage, as reported in empirical studies
(deMast 2006, Lee and Choi 2006).
Competitive advantage could result from other dynamic capabilities, such as
appropriately designed knowledge management initiatives (Gunasekaran and Ngai,
2007). The value and rareness of knowledge management applications can improve
through greater efficiency in their implementation. Also, adaptation of Six Sigma
and knowledge management to unique hospital conditions and patient needs could
increase imitation cost and non-substitutability. Therefore, more appropriate
implementation of knowledge management could enhance the success of Six Sigma
programmes beyond that of employing only Six Sigma practices.
As described above, the implementation of Six Sigma is associated with
improvement in quality programme results, such as quality, customer satisfaction,
net cost savings and reduction of errors, as well as improvement in competitive
advantage. Furthermore, it is expected that the three dimensions of knowledge
management will also enhance Six Sigma initiatives in terms of greater quality
programme results and sustainable competitive advantage. Therefore, the previous
literature review leads to the following research hypotheses:
H1: Six Sigma initiatives will have a positive effect on quality programme
results.
H2: Knowledge acquisition will have a positive effect on quality programme results,
in the context of Six Sigma initiatives.
H3: Knowledge dissemination will have a positive effect on quality programme
results, in the context of Six Sigma initiatives.
H4: Knowledge responsiveness will have a positive effect on quality programme
results, in the context of Six Sigma initiatives.
H5: Six Sigma initiatives will have a positive effect on sustainable competitive
advantage.

Simultaneous implementation of Six Sigma and knowledge management

Six Sigma initiatives

6785

H1
H5
H2

Quality
programme
results

Knowledge acquisition
H6
H3
Knowledge dissemination

H7
H4

Knowledge responsiveness

Sustainable
competitive
advantage

H8

Figure 1. Framework for the effects of Six Sigma and knowledge management on quality
programme results and sustainable competitive advantage.

H6: Knowledge acquisition will have a positive effect on sustainable competitive
advantage, in the context of Six Sigma initiatives.
H7: Knowledge dissemination will have a positive effect on sustainable competitive
advantage, in the context of Six Sigma initiatives.
H8: Knowledge responsiveness will have a positive effect on sustainable competitive
advantage, in the context of Six Sigma initiatives.
These hypotheses are diagrammed by the conceptual model in figure 1 that guides
this study.

4. Methodology
This research employs a survey methodology to collect data in order to test our
research hypotheses, using the hospital organization as the unit of analysis.
To obtain a list of US hospitals for this survey, we utilized a comprehensive
directory of the 6000 hospitals posted on the website Hospitallink.com. From the
hospital websites, we were able to obtain the addresses and telephone numbers for
a random sample of the hospitals. An initial questionnaire was tested in a pilot
survey sent to several hospital Quality Directors. Phone interviews were also initially
conducted to improve the clarity and to reduce the ambiguity of our questions.
We contacted the Quality Director and Information Systems Director at each
hospital to obtain multiple raters who could complete our survey. For additional
raters, we also contacted the Director of Nursing and Risk Manager at each hospital.
By calling the hospitals directly, we were able to ensure that the surveys were emailed
to the appropriate people. Flynn et al. (1990) advocates this approach as an effective
means for improving the response rate. Another method we used to increase our

6786

C. R. Gowen III et al.

response rate was to send two email reminders with the questionnaire attached at
three-week intervals.
We limited our data set to those hospitals from which we received multiple
responses. The final sample of 112 hospitals yielded a response rate of approximately
61%. This response rate compares favourably with the response rates cited in other
published survey-based research studies in the field of operations management
(Flynn et al. 1990). The Cronbach Alpha (CA) inter-rater reliability, which was
calculated for the responses from multiple raters for each hospital, exhibited an
average CA value of 0.71, which is deemed acceptable (Flynn et al. 1990). For each
survey variable, the multiple rater responses were averaged to give a value for each
item from each hospital. Finally, we used Harman’s one-factor test to check whether
common method bias was present (Podsakoff et al. 2003). Harman’s one-factor
test resulted in nine factors accounting for 68.2% of the variance, with the first factor
at 11.6%. Because no factor accounted for most of the variance, the single method
of data collection was an acceptable risk.

4.1 Variables
The key constructs in our conceptual framework are:
.
.
.
.
.
.

Six Sigma initiatives.
Knowledge acquisition.
Knowledge dissemination.
Knowledge responsiveness.
Quality programme results.
Sustainable competitive advantage.

The questionnaire items for each construct were drawn from the previous literature
and are shown in the Appendix. The Six Sigma initiatives (SSI) variable was
measured by four items asking the respondent to assess the level of implementation
of each of these items (Six Sigma system, black/green belt training, DMAIC process,
and quality system financial rewards). The knowledge acquisition (KA), knowledge
dissemination (KD), and knowledge responsiveness (KR) constructs were measured
by items from those scales developed and validated by Darroch (2003, 2005)
and Darroch and McNaughton (2003). The quality programme results (QPR)
construct was measured by four items (quality improvement, patient satisfaction
increase, net cost savings, and reduction in the severity of errors) based on
prior research in healthcare quality (Kazandjian and Lied 1999, Spath 2000, Gowen
et al. 2006a, McFadden et al. 2006a, b). The final construct in the conceptual
framework, sustainable competitive advantage (SCA) was measured by four items
(value added, rareness, costly-to-imitate, and non-substitutability), described
previously (Gowen et al. 2006b) and based on the dynamic capabilities model
(Barney 2002, Winter 2003).
Principal components factor analysis for all of the constructs of this study
confirmed those scales, as reported in table 1, using varimax rotation with Kaiser
normalization advocated by Hinkin (1995). The Cronbach Alpha scale reliability
values for these six constructs consisted of a range of 0.65 to 0.84, which is beyond
the minimum acceptable level of 0.60 for exploratory research (Flynn et al. 1990).

Simultaneous implementation of Six Sigma and knowledge management
Table 1.

6787

Results of factor analysis and Cronbach Alpha scale reliability for all constructs.

Construct

Items

Loading

Alpha

Six Sigma initiatives

Six Sigma system
Black/green belt training
DMAIC process
Quality programme rewards

0.844
0.838
0.559
0.564

0.653

Knowledge acquisition

Survey employees regularly
Managers ask employees work feelings
Appraisals for employees needs
Employees attend training seminars
Staff meetings with employees
Employees take college courses

0.765
0.831
0.828
0.730
0.624
0.747

0.842

Knowledge dissemination

Marketing assesses patient needs
Marketing information accessible
Meetings for marketing trends
Patient information database accessible
Patient satisfaction data sent to all levels
Records of internal best practices

0.804
0.819
0.739
0.641
0.747
0.775

0.848

Knowledge responsiveness

Immediate action on quality issues
Respond to new patient service needs
High effort for patient service requests
Quick response to patient complaints
Quick response to employee concerns

0.771
0.825
0.807
0.838
0.814

0.868

Quality programme results

Quality improvement
Patient satisfaction increase
Net cost savings
Reduction in the severity of errors

0.825
0.809
0.777
0.541

0.723

Sustainable
competitive advantage

Value added
Rareness
Costly-to-imitate
Non-substitutability

0.707
0.852
0.838
0.690

0.778

The items for each scale were averaged to create the variables used in the subsequent
regression analysis.
In order to control for four possible confounding variables, our analysis included
the level of ‘experience’ (EXP) the hospital had with quality systems, the ‘size’ of the
hospital (measured by the number of beds), the number of ‘full time equivalent
employees’ (FTE) dedicated to quality programmes, and the primary mission
of hospital in terms of the ‘type’ (TYP, i.e. community, teaching, or other type of
hospital). Table 2 shows descriptive statistics for these variables. In addition, table 2
includes Pearson correlation coefficients showing the strength of the bivariate
relationships between the variables.

5. Analysis and results
To test our hypotheses, we used hierarchical regression analysis (Cohen et al. 2002).
In this approach, control variables were entered into multiple regression analysis.

13.23
150.14
3.79
2.04
1.49
3.60
3.09
3.82
3.28
2.52

6.880
172.000
9.610
0.371
0.855
0.746
0.828
0.564
0.631
0.702

SD

0.127
0.109
0.160
0.007
0.180
0.245**
0.157
0.096
0.100

EXP

SD, standard deviation; *p50.05; **p50.01.

EXP
SIZE
FTE
TYP
SSI
KA
KD
KR
QPR
SCA

Mean

0.023
0.152
0.175
0.334**
0.311**
À0.014
0.112
0.268**

SIZE

0.001
À0.127
À0.062
0.142
À0.068
0.006
À0.008

FTE

0.271**
0.132
0.303**
0.195*
0.182
0.211*

TYP

0.296**
0.192*
0.068
0.290**
0.290**

SSI

0.632**
0.528**
0.454**
0.407**

KA

0.547**
0.555**
0.369**

KD

0.537**
0.344**

KR

0.497**

QPR

Table 2. Descriptive statistics and Pearson correlation coefficients for quality system experience (EXP), hospital size (SIZE), full-time equivalent
number of quality programme employees (FTE), type of hospital (TYP), six sigma initiatives (SSI), knowledge acquisition (KA), knowledge
dissemination (KD), knowledge responsiveness(KR), quality program results (QPR), and sustainable competitive advantage (SCA).

6788
C. R. Gowen III et al.

Simultaneous implementation of Six Sigma and knowledge management

6789

Theoretically grouped sets of variables were then entered into the regression, and an
F statistic was calculated to determine whether the change in variance explained (R2)
by the additional variables was statistically significant. Tables 3 and 4 show the
cumulative result of entering the control variables, then the SSI construct, and finally
the three knowledge management variables into the overall regression model.
In table 3, the dependent variable was QPR. In the first model, the control variables
were entered and none was statistically significant. In the second model, the SSI

Table 3. Regression results for quality programme results as the
dependent variable (standardized coefficients shown).
Model
1
EXP
SIZE
FTE
TYP
SSI
KA
KD
KR
Overall R2
Overall F
Change in R2
F for change
n ¼ 112

2

3

0.060
0.081
À0.002
0.160

0.070
0.043
0.030
0.093
0.260**

0.044
1.226

0.104
2.462*
0.060
7.127**

À0.038
À0.021
0.013
À0.040
0.212*
0.013
0.346***
0.341***
0.429
9.693****
0.325
19.585****

*p50.05; **p50.01; ***p50.005; ****p50.001.

Table 4. Regression results for sustainable competitive advantage
as the dependent variable (with standardized coefficients shown).
Model
1
EXP
SIZE
FTE
TYP
SSI
KA
KD
KR
Overall R2
Overall F
Change in R2
F for change
n ¼ 112

0.045
0.237*
À0.019
0.168

0.104
3.093*

*p50.05; **p50.01; ***p50.005.

2
0.054
0.205*
0.010
0.110
0.225*

0.149
3.706***
0.045
5.621*

3
À0.008
0.167
0.029
0.060
0.181*
0.145
0.049
0.223*
0.263
4.586***
0.114
5.303***

6790

C. R. Gowen III et al.

variable was also entered. SSI was positive and significant (at p50.01); the change
in R2 was statistically significant as well. In the third model, the three knowledge
management variables were entered as a group and the change in R2 was significant
(at p50.001). Also, both KD and KR were significant (at p50.005), but the control
variables and KA were not significant. In this last model, SSI, KD, and KR were
significant and positive, which indicates support for hypotheses H1, H3, and H4.
Table 4 shows the results of the hierarchical regression where SCA was the
dependent variable. In the first model with only control variables, hospital size
was statistically significant (at p50.05). In the second model, the change in R2 was
significant; SSI and hospital size were significant and positive (at p50.05). In the
third model, SSI and KR were both significant and positive (at p50.05) and the
change in R2 was significant (at p50.005). For this final model, the control variables,
KA, KD were not significant. Therefore, the results in table 4 show support for
hypotheses H5 and H8.

6. Discussion and limitations
These results extend the literature by examining the efficacy of concurrent
implementation of Six Sigma initiatives and knowledge management. Our support
for the effects of Six Sigma on increasing quality programme results and competitive
advantage aligns with the previous descriptive literature (Barry et al. 2002) and
empirical research (Lee and Choi 2006). Similarly, our findings demonstrate that
knowledge management practices improve quality programme results and competitive advantage, which...