Lean maturity and quality in primary care

Monica Kaltenbrunner (Faculty of Health and Occupational Studies, University of Gävle, Gävle, Sweden)
Svend Erik Mathiassen (Faculty of Health and Occupational Studies, University of Gävle, Gävle, Sweden)
Lars Bengtsson (Faculty of Engineering and Sustainable Development, University of Gävle, Gävle, Sweden)
Maria Engström (Faculty of Health and Occupational Studies, University of Gävle, Gävle, Sweden) (Department of Public Health and Caring Sciences, Uppsala University, Uppsala, Sweden) (Nursing Department, Medicine and Health College, Lishui University, Lishui, China)

Journal of Health Organization and Management

ISSN: 1477-7266

Publication date: 28 March 2019



The purpose of this paper is twofold: first, to describe Lean maturity in primary care using a questionnaire based on Liker’s description of Lean, complemented with observations; and second, to determine the extent to which Lean maturity is associated with quality of care measured as staff-rated satisfaction with care and adherence to national guidelines (NG). High Lean maturity indicates adoption of all Lean principles throughout the organization and by all staff.


Data were collected using a survey based on Liker’s four principles, divided into 16 items (n=298 staff in 45 units). Complementary observations (n=28 staff) were carried out at four units.


Lean maturity varied both between and within units. The highest Lean maturity was found for “adhering to routines” and the lowest for “having a change agent at the unit.” Lean maturity was positively associated with satisfaction with care and with adherence to NG to improve healthcare quality.

Practical implications

Quality of primary care may benefit from increasing Lean maturity. When implementing Lean, managers could benefit from measuring and adopting Lean maturity repeatedly, addressing all Liker’s principles and using the results as guidance for further development.


This is one of the first studies to evaluate Lean maturity in primary care, addressing all Liker’s principles from the perspective of quality of care. The results suggest that repeated actions based on evaluations of Lean maturity may help to improve quality of care.



Kaltenbrunner, M., Mathiassen, S., Bengtsson, L. and Engström, M. (2019), "Lean maturity and quality in primary care", Journal of Health Organization and Management, Vol. 33 No. 2, pp. 141-154. https://doi.org/10.1108/JHOM-04-2018-0118

Download as .RIS



Emerald Publishing Limited

Copyright © 2019, Monica Kaltenbrunner, Svend Erik Mathiassen, Lars Bengtsson and Maria Engström


Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial & non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode


In Sweden, primary care is the first level in the healthcare system, offering basic medical care to citizens and residents (Swedish Board of Health and Welfare (SBHWb)). The healthcare system strives to deliver care of good quality using fewer resources (Wilson, 2009), while facing emerging demands (Osborn et al., 2015). National guidelines (NG) have been developed to improve healthcare quality, and all staff are expected to adhere to them (SBHWa, 2016). To achieve care of good quality and to increase productivity, Lean principles have spread to healthcare systems worldwide (Brandao de Souza, 2009; Joosten et al., 2009). In healthcare, it is common to adopt Lean using value stream mapping (VSM), improvement events and standardization (Costa and Godinho Filho, 2016). However, only a few studies of Lean have addressed the extent to which different aspects of Lean have been implemented (Brackett et al., 2013). Moreover, most Lean studies in healthcare primarily focus on productivity (Brandao de Souza, 2009; Costa and Godinho Filho, 2016; Mazzocato et al., 2010) and less on quality of care (D’Andreamatteo et al., 2014). The present study addresses Lean maturity in Swedish primary care, investigating the extent to which it is associated with quality of care.

Lean according to Liker

In the present study, Lean is addressed using Liker’s (2004) description, which includes 14 principles organized into a 4P model: philosophy, processes, people and partners, and problem solving. Briefly, philosophy involves focusing on the customer and engaging the whole organization to bring value to the customer, the organization and society as a whole. A long-term philosophy is prioritized even if this may be expensive in the short term. Processes concerns creating flow through, e.g., standardization and reduction of waste, such as decreasing waiting time, quality deficiencies, excessive inventories and unnecessary movements. People and partners concerns, e.g., showing respect for people in and related to the organization as well as creating prerequisites for them to grow. The managers should know and communicate the company’s philosophy, and the organization and staff should know what is valuable for the customer. Problem solving refers to the notion that staff are devoted to continuously improving processes and quality. Liker (2004) claimed that, if set goals are to be achieved, all Lean principles should be present throughout the organization and used by all staff. In the present paper, the extent to which this is accomplished is termed Lean maturity. Often Lean principles are only partially adopted in healthcare (Brackett et al., 2013) and not throughout the organization (Costa and Godinho Filho, 2016).

Lean in healthcare

In healthcare, Lean is most commonly implemented and assessed using value stream and process mapping (Costa and Godinho Filho, 2016; Mazzocato et al., 2010; Poksinska, 2010), improvement events (Brackett et al., 2013; Costa and Godinho Filho, 2016; Mazzocato et al., 2010) and standardization (Costa and Godinho Filho, 2016; Mazzocato et al., 2010). The extent to which Lean is adopted has mainly been investigated using document studies (e.g. Burgess and Radnor, 2013; Langstrand and Drotz, 2016), observations (e.g. Langstrand and Drotz, 2016; Holden et al., 2015) and interviews (e.g. Langstrand and Drotz, 2016; Grove et al., 2010), while a few studies have used questionnaires involving all staff (Roszell, 2013) or managers only (Dellve et al., 2015). In the context of primary care, few studies have focused on Lean. One interview study in both primary care and hospitals in Sweden addressed the use of VSM, continuous improvements, standardization, identifying waste, teamwork, daily short meetings, supporting and coaching leadership style, visual control and 5S (Drotz and Poksinska, 2014). An interview study with staff in primary care in the UK described adoption of VSM and stakeholder mapping (Grove et al., 2010), while a study of primary care in the USA described standardization, work flow, redesigning rooms and holding daily morning meetings (Hung et al., 2016). The above studies have provided only a partial insight into Lean maturity, in that they have failed to address Lean maturity using Liker’s entire 4P model and to collect information on Lean from all staff members, both of which, according to Liker, are paramount to successful implementation and maintenance of Lean.

Effects of Lean in healthcare

Different outcomes appear to be related to Lean in healthcare, e.g., staff members’ working conditions and stress (e.g. Dellve et al., 2015), productivity measured as number of treated patients (Edwards et al., 2012; Hwang et al., 2014) and reduced waiting time (Fillingham, 2007; Simons et al., 2017). As regards quality of care, results indicate that Lean may improve patient safety (Jorma et al., 2016; Simons et al., 2017), decrease mortality and lead to faster patient recovery (Fillingham, 2007; Yousri et al., 2011).

In primary care, knowledge concerning Lean maturity and quality of care is lacking (D’Andreamatteo et al., 2014). One case study (Poksinska et al., 2016) showed no difference in patient satisfaction with units that had adopted Lean compared to units that had not. The study reports that two primary care units with purported successful Lean adoption showed only limited focus on the patient perspective, e.g., on what patients consider valuable, and whether patients are satisfied. Assessments of Lean maturity based on staff perceptions and opinions are rare, although Liker (2004) emphasized that Lean should be practiced by all staff and managers in the organization. Thus, the aim of the present study was twofold: first, to describe Lean maturity in primary care using a questionnaire based on Liker’s description of Lean and observations; and second, to determine the extent to which Lean maturity is associated with quality of care measured as staff-rated satisfaction with care and adherence to NG to improve healthcare quality. In Sweden, NG can be related to quality of care, as they contain recommendations aimed at guiding healthcare staff in selecting appropriate treatment and methods (SBHWa, 2016); staff usually strive to adhere to these guidelines.



The study had a descriptive, correlational design and used a mixed-methods approach. Results from analyzes of quantitative questionnaire data are illustrated using qualitative observation and interview data.

Setting and sample, questionnaires

All 52 eligible primary care units in a region in central Sweden were asked to participate: 42 accepted, whereof 4 were private for-profit providers. To achieve greater dispersion among participating units, one of the largest private for-profit healthcare providers in Sweden was also asked to participate, provided the units had adopted Lean; 6 of 85 units accepted. In total, 850 staff at 48 units were eligible for the study. The manager first informed the staff about the study, and then staff received a questionnaire, sent to their work e-mail address, which also contained further information about the study. A total of 351 staff members responded to the questionnaire; 30 of them were excluded because they did not completely answered the Lean items. Further analyses of the 4P’s included only participants who had responded to at least 50 percent of the Lean items (n=298 (35 percent) of eligible staff, working at 45 units). Most participants were female (86 percent) and registered nurses (41 percent). Mean age was 51 years (Table I).

Setting and sample, observations and interviews

Two units with high and two with low Lean maturity, as indicated by staff ratings in the questionnaire, were selected by convenience for observation. By coincidence, all four units were located in rural areas. At these units, a purposive sample of staff was asked to further participate; sampling aimed at variation in profession, age and gender. In total, 28 participants were observed, six to eight participants/unit. Most participants were female (n=24) and about half were registered nurses (n=13) (Table I).

Data collection, questionnaires

The survey questionnaire was web-based and sent out in Spring 2016. Two reminders and a paper version were sent to non-responders. Lean maturity was assessed using the 16-item Lean in Healthcare Questionnaire (LiHcQ) (Kaltenbrunner et al., 2017). In the LiHcQ, three of Liker’s 4P (philosophy, people and partners, problem solving) (Liker, 2004) are represented by three items, while the fourth (processes) includes seven items. Response alternatives range from 1 (low Lean maturity) to 5 (high Lean maturity). Psychometric tests of the LiHcQ showed acceptable values (Kaltenbrunner et al., 2017). Staff satisfaction with provided care was assessed using the nine-item Staff Satisfaction with Care (SSC) scale (Mårtensson et al., 2010). Responses are measured on a seven-point scale, 7 standing for high satisfaction. The reliability of the scale is satisfactory (Mårtensson et al., 2010). In the present study, one SSC item was discarded because it was not relevant to primary care. In addition, a study-specific questionnaire was developed to measure staff-rated adherence to NG. The NG questionnaire was based on existing Swedish NG (SBHWa) for treating different medical diagnoses. The questionnaire addressed, in total, 13 guidelines for patients with, e.g., diabetes, different cancer diagnoses and stroke. For each guideline, respondents were asked whether their work fully adhered to the NG. Response alternatives ranged from 1 (strongly disagree) to 5 (strongly agree).

Data collection, observations and interviews

Observations were conducted by the first author (MK), shadowing each participant for approximately 4 h to observe behavior in the context of the 4P using a structured observation protocol. Processes was studied by observing whether appointed times were held, whether any interruptions occurred in the care process, the location of storage rooms and use of standardization visible to the observer. During the observations, MK asked complementary questions and recorded answers on an MP3 player. Open-ended questions captured philosophy (communication of shared goals and long-term thinking), processes (routines, flow and standardization), people and partners (teamwork and respect) and problem solving (problem solving and participation in decision making). In total, 97 h of observation and 5.5 h of interviews were conducted. After every observation day, except for 2, MK had phone contact with the last author, ME, to reflect on the procedure and the collected data.

Data analyses, questionnaires

Data were analyzed using IBM SPSS Statistics version 22. First, bivariate correlations between variables were determined using Spearman’s correlation coefficient. Thereafter, generalized estimating equations (GEE) models for binary data were employed to study multivariate relationships, controlling for correlations within units. Because the dependent variables were not normally distributed, data were dichotomized using the median. In a first GEE model, the total score of all items in LiHcQ was used as the independent variable with either SSC or NG as dependent variables. In a second model, scores for each of the 4P were entered instead as independent variables. Internal missing data on Lean were replaced using multiple imputations (MI), as recommended by Newman (2014) and Shrive et al. (2006). When employing MI, five sets with imputed data are calculated together with the original data. The GEE models for MI data returned odds ratio (OR), 95% confidence interval (CI) and p-values for each MI data set, as well as the original data set. p-values and regression coefficients, but not OR, were also obtained for pooled data. p-values <0.05 were considered to indicate statistical significance.

Data analyses, observations and interviews

Observation data were analyzed deductively using qualitative content analysis (Patton, 2002), employing Liker’s 4P as a framework for categorizing the data. The MP3 files were transcribed verbatim, and handwritten field notes together with data from the observation protocol were transferred to a Word document. Data were read several times to get a sense of the whole. Thereafter, meaning units linked to the 4P were identified, condensed and categorized. The first (MK) and last author (ME) continuously discussed each step in the analyses process in order to enhance trustworthiness and to reach consensus (Patton, 2002). Moreover, the third author (LB) read and commented on the categorized observation data to gain further insight and increase trustworthiness.

Ethical approval

The study was approved by the Regional Ethical Review Board in Uppsala (Reg. No. 2014/525 and 2014/525/1).


Lean maturity

The questionnaire results (Table II) will be presented along with illustrations and quotes (Table III) from the observations. The focus will be on contrasting low Lean maturity (levels 1 and 2) and high Lean maturity (levels 4 and 5).

Philosophy: regarding staff members’ (Item 1 (I1)) and first-line managers’ (I2) commitment to and engagement with Lean, a maturity level of 4–5 was reported by 34 and 42 percent, respectively. During the observations, staff at all four units mentioned that the first-line managers often communicated goals. Regarding time for continuous improvement of work (I3), 46 percent rated that no time was specifically allocated to this, or that time was allocated only in rare cases (maturity levels 1–2); 27 percent rated a maturity levels 4–5.

Processes: for VSM (I6), 48 percent rated maturity levels 1–2, saying that no or only some processes were mapped, and that awareness of value and waste was low. The highest Lean maturity was found for routines (I7), where more than half (56 percent) of staff members rated maturity levels 4–5, i.e., that most teams had routines, or that everyone was highly skilled in developing, updating and following routines. However, during the observation, staff also mentioned a certain lack of routines, difficulties in finding them and low compliance concerning, e.g., hygiene routines and refilling trolleys. Item 8 concerned whether the staff planed their work based on patients’ needs; 41 percent rated maturity levels 1–2, i.e., that the unit investigated patient influx and necessary resources minimally or not at all. During observation, some staff members mentioned that they planned their work based on patient statistics. Concerning automatic quality controls, Lean maturity was also low (I9); 50 percent rated maturity levels 1–2, indicating no or little awareness of how automatic quality controls could be practiced. Regarding care based on patient demands, flow and the use of signs/signals (I10), staff ratings varied greatly: about one-third rated 1–2, and one-third 4–5. During observation of all four units, the facilities used were often organized and structured. However, some unorganized rooms were also observed, e.g., where sheets of papers were placed on the floor or behind the curtains. Interruptions were common at all units, during which staff were searching for information, retrieving material or medications or asking others for advice. The storerooms at two units were centrally placed, while they were more peripherally placed at the other two units. To facilitate overview, one unit had cupboards with glass doors and used a mirror to detect new patients coming in. Regarding visualizing improvements (I11), about one-third rated maturity to be 1–2 (no or some improvements visualized, and if so in an unstructured manner), and one-third rated 4–5 (improvements are visualized at the unit, and the unit is highly skilled in visualizing improvements). As to whether staff were involved in purchasing new products (I15), 43 percent responded that they were always involved, i.e., maturity levels 4–5. Observations of whether appointed times were held could not be performed as planned, and data on this aspect are disregarded.

People and partners: staff at all four units described working in fixed teams, and that multi-functional teams were formed based on patients’ needs. They mentioned that the first-line manager was not always knowledgeable about staff members’ work tasks. On the other hand, the staff said that the first-line manager showed them respect by listening to them and showing an interest in their opinions. The lowest Lean maturity concerned whether the unit had an appointed change agent (I4); 81 percent of the staff rated a maturity level of 1 (indicating that the unit did not have such an agent) or 2. The second highest Lean maturity among all items was reported for value as described from the patient perspective (I5). Nearly half (48 percent) of the staff rated maturity levels 4–5 on this item, meaning that most or all staff can see and describe the processes at the unit and identify their value for the patient. Staff contact with partners and suppliers (I16) showed the second lowest maturity of all items; 61 percent rated maturity level 1 (staff having no contact at all) or 2 (some staff had contact with partners and suppliers). During the observations, staff at all four units described good collaborations with the municipality and other healthcare service providers. Some staff mentioned difficulties in influencing suppliers and long delivery times.

Problem solving: regarding evaluations (I12), 48 percent rated maturity levels 1–2, i.e., that staff had difficulties influencing what and how to evaluate or that some staff had recently begun searching for evaluation methods. For problem solving (I13), 41 percent rated levels 1–2, i.e., that improvement work was largely unplanned and unstructured. The observations showed that all four units had daily meetings at the visual management board, however, with various contents. One unit had two visual management boards centrally placed, which were used when they had their daily discussion of problems and improvements. The same unit had a list in every room for documenting problems to be addressed later at the unit meeting. Other problem-solving methods reported by staff were based on, e.g., set goals, guidelines or statistics. Sometimes the problem was solved, and the solution presented post hoc to the first-line manager, who, according to staff, was most often positive. However, they also mentioned lack of follow-up. Item 14 concerns staff involvement in decision making, and the ratings showed that 44 percent of staff felt that they were involved and that the unit had structured meetings aimed at consensus (i.e. maturity levels 4–5).

Median and quartiles (Q1–Q3) for individual items in philosophy were: 3(2–4), in processes: 3(1–4), in people and partners: 2(1–4) and in problem solving: 3(1–4).

Positive and significant associations were found between all 4P (Table IV). Intra-class correlation coefficient (ICC) values for clustering effect within units for philosophy were: ICC 0.24, processes: ICC 0.24, people and partners: ICC 0.25 and problem solving: ICC 0.22. An overview of Lean maturity across all units is presented in Figure 1, illustrating that units differed, with respect to both each of the 4P and overall Lean maturity. As seen in Figure 1, the possible total score for philosophy, people and partners and problem solving can range from 3 to 15, whereas processes can range from 7 to 35; Lean in total, i.e., the sum of all scores from all items in the LiHcQ, can take values from 16 to 80. In Figure 1, where the graph rises rapidly, this indicating that several units have this same score.

Lean maturity, SSC and adherence to NG

For the pooled MI data, bivariate correlations analyses showed that the LiHcQ total score was significantly associated with both SSC (rs 0.19, p=0.002) and NG (rs 0.39, p<0.001) (Table IV). Each of the 4P was also significantly associated with SSC (p<0.001 to 0.040) and NG (p<0.001 for all). Thus, in units with high Lean maturity, SSC and NG also received high ratings.

The GEE models (Table V) confirmed that the LiHcQ total score was significantly associated with both SSC and NG. Multivariate analyses including each of the 4P as independent variables revealed that, for the MI data, processes was significantly associated with both SSC and NG. Multivariate analyses on original data without imputation showed that processes and problem solving were significantly associated with NG, while none of the 4P was significant for SSC.


Our study is one of the first to report on Lean maturity in primary care based on a comprehensive instrument using staff ratings to measure Lean. The results showed large dispersion in Lean maturity, both between care units and within the same care unit for the 4P, some principles receiving considerably higher maturity ratings than others. Variation in Lean adoption between organizations has been described in several reviews (Antierens et al., 2018; Brackett et al., 2013; Costa and Godinho Filho, 2016). Reasons for Lean adoption only being partial, and for not achieving high Lean maturity, include staff showing limited interest in and having little experience of Lean (Dellve et al., 2015; Drotz and Poksinska, 2014; Hasle et al., 2016), factors also identified in the observational data in our study. According to Liker, to achieve success based on Lean, all staff need to be involved and all Lean principles need to be used. Nonetheless, improvements in quality of care and patient safety have been reported even in cases of partial adoption of Lean (Antierens et al., 2018). Our results indicate that quality of care increases when Lean maturity increases. However, when multivariate relationships were analyzed, controlling for correlations within units, only processes was significantly associated with quality of care. It would seem that some principles contribute more to quality of care than others do, or that certain principles are easier to implement than others, depending on setting and organizational culture.

We found the highest Lean maturity for processes involving routines. During the observations, staff expressed diverse opinions about routines. Describing value from the patient perspective also showed high Lean maturity, a finding related to person-centered care (McCormack and McCance, 2010). Our findings are in accordance with results from previous research on Lean in healthcare (Antierens et al., 2018). One explanation for the results can be the tradition in healthcare of having both routines and a patient perspective, making these principles easier to address in efforts to increase Lean maturity.

The principle of VSM has been described as being frequently adopted, among those principles investigated at all (Antierens et al., 2018; Costa and Godinho Filho, 2016). Our results showed low Lean maturity for VSM. This confirms the findings of Drotz and Poksinska (2014), who focused primarily on primary care units, finding that VSM was only partially adopted and that it was terminated after improvements had been made. One reason for low Lean maturity in primary care may be that care processes are more complex because they, e.g., involve different professions working independently, each profession with its own patients, and because processes may extend over a longer period of time than, e.g., in an emergency unit. According to Hasle et al. (2016), complex processes involving different professions may be one reason for low Lean maturity. In our study, having a change agent showed low Lean maturity. Greenhalgh et al. (2004) claimed that having a change agent or a champion is essential in disseminating an organizational innovation. Low Lean maturity was also seen concerning the question of whether time was allocated to improvement work and problem solving. However, during the observation at one unit with high Lean maturity, different problem-solving methods were used and time was allocated so that all staff could participate. Having resources and being engaged can be essential when striving for high Lean maturity.


The present study used a cross-sectional design, which limits our ability to draw conclusions about causality in the relationship between Lean maturity and quality of care. Moreover, we employed non-random sampling and faced a rather low response rate in the questionnaire part of the study, which limit generalizability. The qualitative data in the study could have been explored more thoroughly; future work may address these data in more depth. However, the study also had several strengths: mixed methods were used and the sample is representative according to national data registers on primary care staff, showing that female registered nurses are the largest licensed group in healthcare (SCB, 2015).

Implications for practice

Lean is present in primary care and associated with quality of care. Therefore, striving to increase quality of care by adopting Lean may be fruitful, although more research is needed on the nature of associations and possible interventions. For managers, when adopting Lean, a long-term aim should be to achieve high Lean maturity even though adoption might have started on a small scale. The manager also needs to repeatedly evaluate Lean maturity and base initiatives for continuous improvement on results from such evaluations.


Lean maturity varied in primary care, both in terms of adoption of different Lean principles and Lean adoption between units. Lean maturity was positively associated with quality of care, indicating that Lean has the potential to meet some of the challenges facing present-day healthcare.


Cumulative distributions across the 45 investigated units of their median LiHcQ scores for Lean in total, and each of the 4P

Figure 1

Cumulative distributions across the 45 investigated units of their median LiHcQ scores for Lean in total, and each of the 4P

Demographic data for participants

Survey Observations
Total number of participants, n 298 28
Participants at public non-profit/private for-profit provider healthcare units, n 230/68 28/0
Women, n 255 24
Men, n 43 4
Md (Q1Q3) 53 (43−59) 51 (45−59)
Mean (SD) 51 (10.2) 51 (9.0)
Profession, n
Registered nurses 121 13
Licensed practical nurse 17 6
Manager 21
Physiotherapist 37 3
Occupational therapist 9
Physician 45 6
Administrator and secretary 35
Dietician 1
Social worker and psychologist 24
Years worked at the present unit
Md (Q1Q3) 6 (3−13) 5 (2−15)
Mean (SD) 9 (8.8) 9 (10.0)
Years worked in the profession
Md (Q1Q3) 20 (13−30) 24 (14−36)
Mean (SD) 21 (11.4) 24 (12.0)

Notes: Md, median; Q, quartiles. Some participants have multiple functions and therefore the number of professions does not add up to 298

Number of participants (with percent in parentheses) answering each item in the LiHcQ, organized according to Liker’s 4P

Maturity, response LiHcQ
4P Items LiHcQ 1 2 3 4 5 Md (Q1Q3) Mean (SD) Valid n
Philosophy 1. Staff engagement and commitment to Lean 42 (14%) 59 (20%) 96 (32%) 54 (18%) 46 (16%) 3 (2–4) 3.0 (1.3) 297
2. First-line managers’ engagement and commitment to Lean 29 (10%) 34 (12%) 98 (36%) 73 (27%) 41 (15%) 3 (3–4) 3.2 (1.2) 275
3. Allocation of time to improvement work 63 (20%) 81 (26%) 82 (27%) 39 (13%) 42 (14%) 3 (2–4) 2.7 (1.3) 307
Processes 6. Use of value stream mapping 45 (16%) 91 (32%) 83 (29%) 45 (16%) 21 (7%) 3 (2–3) 2.7 (1.1) 285
7. Developing and following routines 5 (2%) 53 (18%) 77 (25%) 109 (36%) 59 (20%) 4 (3–4) 3.5 (1.0) 303
8. Planning work based on patients’ needs 34 (12%) 82 (29%) 93 (33%) 50 (17%) 25 (9%) 3 (2–4) 2.8 (1.1) 284
9. Implementing and using automatic quality controls 74 (27%) 63 (23%) 69 (26%) 46 (17%) 18 (7%) 2(1–3) 2.5 (1.2) 270
10. Basing the provided care on what patients desire 32 (12%) 59 (21%) 101 (36%) 40 (14%) 47 (17%) 3 (2–4) 3.0 (1.2) 279
11. Visualizing improvements and placing them strategically 14 (5%) 77 (28%) 106 (38%) 66 (24%) 15 (5%) 3 (2–4) 3.0 (1.0) 278
15. Purchasing and implementing new products 17 (6%) 33 (11%) 118 (40%) 109 (36%) 20 (7%) 3 (3–4) 3.3 (1.0) 297
People and partners 4. Access to a change agent at each unit 204 (69%) 36 (12%) 24 (8%) 15 (5%) 17 (6%) 1 (1–2) 1.7 (1.2) 296
5. Skills in describing value from the patient perspective 23 (8%) 55 (19%) 75 (25%) 106 (36%) 35 (12%) 3(2–4) 3.3 (1.1) 294
16. Collaborating with partners and suppliers 35 (12%) 141 (49%) 78 (27%) 19 (6%) 18 (6%) 2 (2–3) 2.5 (1.0) 291
Problem solving 12. Staff evaluating the care 86 (31%) 46 (17%) 64 (23%) 68 (25%) 12 (4%) 3 (1–4) 2.5 (1.3) 276
13. Solving problems 49 (17%) 70 (24%) 90 (31%) 50 (17%) 28 (10%) 3 (2–4) 2.8 (1.2) 287
14. Decision making involving both staff and manager 12 (4%) 72 (24%) 83 (28%) 95 (32%) 34 (12%) 3 (2–4) 3.2 (1.1) 296

Notes: LiHcQ, Lean in Healthcare Questionnaire; 4P, philosophy, processes, people and partners, problem solving; Md, median; Q, quartiles. Response alternatives ranging from 1 (low maturity) to 5 (high maturity)

Quotes related to each of the 4P

Dimension Context Quote
Philosophy The participant was asked whether shared goals were expressed by the first-line manager and whether the communicated goals were important and relevant for the participant. The response also illustrates that the participant was confident that trust supports accurate decisions “Yes, for sure, for sure, they are (the goals). And we have these performance-related payments and that’s kind of […] at the moment it’s, for instance, home visits by health visitors from the baby clinic and that’s because it’s important, they (the healthcare trust) think, to get it underway, that routine, and well you get extra money too”
Processes The quote considers routines, whether the participant knew if there were any routines for the work tasks he/she conducted “Well, I think we have written routines somewhere here, but I don’t know where they are. When it comes to typical wound dressings and things like that, then we use the document developed by the trust or The Handbook for Healthcare, or there are a lot of routines on the intranet. But, regarding (routines) for a workday, I don’t know where they are. If I need them I’ll find them”
People and partners The participant was asked whether the first-line manager listened to and respected staff and their problems, ideas and questions. The quota also illustrates that all staff helped each other in efforts to meet the patient’s needs “Yes, definitely! She/he always listens, and if you go to her/him and need some help or information you get a pretty quick answer, she/he looks it up at once. It’s almost like a family here I think, everyone, everybody helps each other, you can ask anybody and everybody just helps, it’s great”
Problem solving The quote illustrates how problem solving can be conducted. First, the participant describes how one aspect of problem solving was improved when a visual management board was implemented. The information board showed one month, and the colors made it easy to see whether work was generally satisfying, and it highlighted issues or problems of importance for continuous improvements in care processes “We just got a visual management board with green, yellow, red and orange, colors you can fill in. Well, green is good things and red means that some disturbing patient harm has occurred. It’s a kind of assessment of how we do our work. This is all new, earlier we just talked about it, now it is more graphic”

Note: The “Context” column provides a description of the background of each quote

Internal consistency (Cronbach’s α) and associations (Spearman; with p-values) between staff assessments of Lean, Staff Satisfaction with Care (SSC) and adherence to national guidelines (NG)

Variables SSC NG Philosophy Processes People and partners Problem solving LiHcQ tot
SSC α 0.90
NG 0.111 α 0.94
p-value 0.118
Philosophy 0.142 0.293 α 0.76
p-value 0.022 <0.001
Processes 0.199 0.403 0.664 α 0.87
p-value <0.001 <0.001 <0.001
People and partners 0.164 0.314 0.627 0.729 α 0.59
p-value 0.009 <0.001 <0.001 <0.001
Problem solving 0.128 0.347 0.597 0.771 0.687 α 0.81
p-value 0.040 <0.001 <0.001 <0.001 <0.001
LiHcQ tot 0.189 0.394 α 0.93
p-value 0.002 <0.001

Notes: α, Cronbach’s α values are given along the diagonal for SSC, Staff Satisfaction with Care; NG, national guidelines; LiHcQ, Lean in Healthcare Questionnaire. LiHcQ tot includes a total score for all items in the LiHcQ. The results are based on imputed data (MI, multiple imputation). Boldface type indicates statistically significant p-values

Associations of overall Lean maturity (LiHcQ tot, total score for all items, Model I) and each of the 4P (Model II) with Staff Satisfaction with Care (SSC) and national guidelines (NG)

Original data MI data set, variation from 5 data sets Pooled MI data
Outcome Predictors OR 95% CI p-value OR (min−max) p-value
SSC Model I
LiHcQ tot 1.025 1.007; 1.043 0.006 1.044−1.045 <0.001
Age 0.968 0.947; 0.989 0.003 0.968−0.969 0.001
Model II
Philosophy 0.994 0.891; 1.108 0.913 0.977−1.017 0.916
Processes 1.030 0.952; 1.114 0.460 1.076−1.110 0.027
People and partners 1.044 0.896; 1.217 0.579 0.954−0.993 0.830
Problem solving 1.028 0.908; 1.162 0.666 1.027−1.102 0.407
Age 0.968 0.948; 0.989 0.002 0.965−0.968 0.001
NG Model I
LiHcQ tot 1.071 1.046; 1.096 <0.001 1.063 for all 5 MI data sets <0.001
Model II
Philosophy 1.030 0.879; 1.207 0.716 1.029−1.040 0.653
Process 1.128 1.040; 1.223 0.004 1.126−1.138 0.002
People and partners 0.857 0.734; 1.000 0.050 0.886−0.933 0.203
Problem solving 1.152 1.022; 1.298 0.020 1.064−1.114 0.221

Notes: SSC, Staff Satisfaction with Care; LiHcQ, Lean in Healthcare Questionnaire; NG, national guidelines; 4P, philosophy, processes, people and partners and problem solving; OR, odds ratio. In all GEE models, an unstructured working correlation matrix was used. Results from GEE models are presented for original data, imputed data (multiple imputation (MI)) and pooled data. Italic face type shows statistically significant values (p<0.05)


Antierens, A., Beeckman, D., Verhaeghe, S., Myny, D. and Van Hecke, A. (2018), “How much of Toyota’ s philosophy is embedded in health care at the organisational level? A review”, Journal of Nursing Management, pp. 1-10.

Brackett, T., Comer, L. and Whichello, R. (2013), “Do lean practices lead to more time at the bedside?”, Journal for Healthcare Quality, Vol. 35 No. 2, pp. 7-14.

Brandao de Souza, L. (2009), “Trends and approaches in lean healthcare”, Leadership in Health Services, Vol. 22 No. 2, pp. 121-139.

Burgess, N. and Radnor, Z. (2013), “Evaluating lean in healthcare”, International Journal of Health Care Quality Assurance, Vol. 26 No. 3, pp. 220-235.

Costa, L.B.M. and Godinho Filho, M. (2016), “Lean healthcare: review, classification and analysis of literature”, Production Planning & Control, Vol. 27 No. 10, pp. 823-836.

D’Andreamatteo, A., Iannu, L., Lega, F. and Sargiacomo, M. (2014), “Lean in healthcare: a comprehensive review”, Health Policy, Vol. 119 No. 9, pp. 1197-1209.

Dellve, L., Williamsson, A., Strömgren, M., Holden, R.J. and Eriksson, A. (2015), “Lean implementation at different levels in Swedish hospitals: the importance for working conditions and stress”, International Journal of Human Factors and Ergonomics, Vol. 3 Nos 3/4, pp. 235-253.

Drotz, E. and Poksinska, B. (2014), “Lean in healthcare from employees’ perspectives”, Journal of Health Organization and Management, Vol. 28 No. 2, pp. 177-195.

Edwards, K., Nielsen, A.P. and Jacobsen, P. (2012), “Implementing lean in surgery – lessons and implications”, International Journal of Technology Management, Vol. 57 Nos 1-3, pp. 4-17.

Fillingham, D. (2007), “Can lean save lives?”, Leadership in Health Services, Vol. 20 No. 4, pp. 231-241.

Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P. and Kyriakidou, O. (2004), “Diffusion of innovations in service organizations: systematic review and recommendations”, The Milbank Quarterly, Vol. 82 No. 4, pp. 581-629.

Grove, A.L., Meredith, J.O., Macintyre, M., Angelis, J. and Neaily, K. (2010), “Lean implementation in primary care health visiting services in National Health Service UK”, Quality & Safety in Health Care, Vol. 19 No. 5, p. e43.

Hasle, P., Nielsen, A.P. and Edwards, K. (2016), “Application of lean manufacturing in hospitals – the need to consider maturity, complexity, and the value concept”, Human Factors and Ergonomics in Manufacturing & Service Industries, Vol. 26 No. 4, pp. 430-442.

Holden, R.J., Eriksson, A., Andreasson, J., Williamsson, A. and Dellve, L. (2015), “Healthcare workers’ perceptions of lean: a context-sensitive, mixed methods study in three Swedish hospitals”, Applied Ergonomics, Vol. 47, pp. 181-192.

Hung, D., Gray, C., Martinez, M., Schmittdiel, J. and Harrison, M.I. (2016), “Acceptance of Lean redesigns in primary care: a contextual analysis”, Health Care Management Review, pp. 1-10.

Hwang, P., Hwang, D. and Hong, P. (2014), “Lean practices for quality results: a case illustration”, International Journal of Health Care Quality Assurance, Vol. 27 No. 8, pp. 729-741.

Joosten, T., Bongers, I. and Janssen, R. (2009), “Application of lean thinking to health care: issues and observations”, International Journal for Quality in Health Care: Journal of the International Society for Quality in Health Care/ISQua, Vol. 21 No. 5, pp. 341-347.

Jorma, T., Tiirinki, H., Bloigu, R. and Turkki, L. (2016), “LEAN thinking in Finnish healthcare”, Leadership in Health Services, Vol. 29 No. 1, pp. 9-36.

Kaltenbrunner, M., Bengtsson, L., Mathiassen, S.E. and Engström, M. (2017), “A questionnaire measuring staff perceptions of Lean adoption in healthcare: development and psychometric testing”, BMC Health Services Research, Vol. 17 No. 1, p. 235.

Langstrand, J. and Drotz, E. (2016), “The rhetoric and reality of Lean: a multiple case study”, Total Quality Management & Business Excellence, Vol. 27 Nos 3-4, pp. 398-412.

Liker, J.K. (2004), The Toyota Way, McGraw-Hill, New York, NY.

McCormack, B. and McCance, T. (2010), Person-Centred Nursing: Theory and Practice, Wiley Blackwell, Oxford.

Mårtensson, G., Carlsson, M. and Lampic, C. (2010), “Is nurse–patient agreement of importance to cancer nurses’ satisfaction with care?”, Journal of Advanced Nursing, Vol. 66 No. 3, pp. 573-582.

Mazzocato, P., Savage, C., Brommels, M., Aronsson, H. and Thor, J. (2010), “Lean thinking in healthcare: a realist review of the literature”, Quality and Safety in Health Care, Vol. 19 No. 5, pp. 376-382.

Newman, D.A. (2014), “Missing data: five practical guidelines”, Organizational Research Methods, Vol. 17 No. 4, pp. 372-411.

Osborn, R., Moulds, D., Schneider, E.C., Doty, M.M., Squires, D. and Sarnak, D.O. (2015), “Primary care physicians in ten countries report challenges caring for patients with complex health needs”, Health Affairs (Project Hope), Vol. 34 No. 12, pp. 2104-2112.

Patton, M.Q. (2002), Qualitative Research and Evaluation Methods, 3rd ed., Sage, London.

Poksinska, B.B. (2010), “The current state of lean implementation in health care: literature review”, Quality Management in Healthcare, Vol. 19 No. 4, pp. 319-329.

Poksinska, B.B., Fialkowska-Filipek, M. and Engström, J. (2016), “Does Lean healthcare improve patient satisfaction? A mixed-method investigation into primary care”, BMJ Quality & Safety, Vol. 26 No. 2, pp. 95-103.

Roszell, S.S. (2013), “Measuring lean management penetration on the hospital nursing frontline: instrument development”, doctoral dissertation, University of North Carolina at Chapel Hill.

SBHWa, “National guidelines”, Swedish Board of Health and Welfare (SBHWa), Stockholm, available at: www.socialstyrelsen.se/riktlinjer/nationellariktlinjer (accessed March 9, 2018).

SBHWb (2016), “Primärvårdens uppdrag (in Swedish)”, Swedish Board of Health and Welfare (SBHWb), Stockholm, available at: www.socialstyrelsen.se/Lists/Artikelkatalog/Attachments/20066/2016-3-2.pdf (accessed March 9, 2018).

SCB (2015), “Statistikdatabasen (Elektronisk) (in Swedish)”, SCB (Statistics Sweden), Stockholm, available at: www.statistikdatabasen.scb.se/pxweb/sv/ssd/START__AM__AM0208__AM0208E/YREG54/?rxid=ac057f9a-5651-4746-a532-93faea3d72cc (accessed November 10, 2017).

Shrive, F.M., Stuart, H., Quan, H. and Ghali, W.A. (2006), “Dealing with missing data in a multi-question depression scale: a comparison of imputation methods”, BMC Medical Research Methodology, Vol. 6 No. 1, p. 57.

Simons, P., Backes, H., Bergs, J., Emans, D., Johannesma, M., Jacobs, M., Marneffe, W. and Vandijck, D. (2017), “The effects of a lean transition on process times, patients and employees”, International Journal of Health Care Quality Assurance, Vol. 30 No. 2, pp. 103-118.

Wilson, G. (2009), “Implementation of releasing time to care – the productive ward”, Journal of Nursing Management, Vol. 17 No. 5, pp. 647-654.

Yousri, T.A., Khan, Z., Chakrabarti, D., Fernandes, R. and Wahab, K. (2011), “Lean thinking: can it improve the outcome of fracture neck of femur patients in a district general hospital?”, Injury, Vol. 42 No. 11, pp. 1234-1237.


The authors would like to thank all participants and statistician Hans Högberg for statistical advice. The research was financially supported by the University of Gävle.

Corresponding author

Monica Kaltenbrunner can be contacted at: moakat@hig.se